Copyright 2012, Austin Sims

Economic perceptions and the mass media: The role of exemplification within economic news stories

by

Austin Sims, BA, MS

A Dissertation

In

MASS COMMUNICATIONS

Submitted to the Graduate Faculty of Texas Tech University in

Partial Fulfillment of the Requirements for

the Degree of

DOCTOR OF PHILOSOPHY

Approved

Dr. Trent Seltzer Chair of Committee

Dr. Coy Callison

Dr. Weiwu Zhang

Dr. Klaus Becker

Dominick Casadonte Dean of the Graduate School

December, 2012

Copyright 2012, Austin Sims

Texas Tech University, Austin Sims, December 2012

ii

ACKNOWLEDGMENTS I would like to thank my committee for their dedication and patience as I

worked through the process of completing my dissertation. Dr. Zhang took the time to

sit with me and go over theory and its importance. Dr. Becker did a phenomenal job in

helping me navigate and understand the subject of economics. I never would have

been able to design and create my experiment without Dr. Callison taking the time to

sit down and go over my instrument and story manipulations. And most importantly I

would like to think Dr. Seltzer for his patience in putting up with me and walking me

through each stage of the dissertation. It was very much appreciated.

I would also like to thank the graduate school for their generosity in funding

my dissertation. The funds helped cover the costs of hiring a third party sampling firm

to collect the data needed to power the experiment. Secondly, the graduate program at

the College of Media and Communications for funding my tuition and providing

scholarships that enabled me to continue on in academia.

Also to Dr. Jerry Hudson for providing me an opportunity to work on

academic and applied research at the Center for Communications Research. The

experience I gained from him taking a chance on me has allowed me to continue to

grow as a researcher and prepared me for the next step in my career. And lastly to my

mom, Debbie Sims, for taking the time to read over my work when I was tired and ask

questions when things didn’t make sense. And to my dad, John Sims, for encouraging

me to stick with it and to continue to move forward.


iii

TABLE OF CONTENTS ACKNOWLEDGMENTS ............................................................................................. ii ABSTRACT ............................................................................................................... v LIST OF FIGURES ................................................................................................... vi I. CHAPTER I INTRODUCTION ................................................................................ 1 II. CHAPTER II LITERATURE REVIEW ................................................................... 8

Overview ............................................................................................................. 8 Economic and Financial Information in the Media .................................................................. 8 Economic News and the President ......................................................................................... 13 Economic News and the U.S. Electorate ............................................................................... 16 Economic News and the Media Elite ..................................................................................... 24 Economics as an Obtrusive Issue ........................................................................................... 27 Economic Literacy and Assessment ....................................................................................... 28

Theoretical Framework ..................................................................................... 30

Media Effects and Public Opinion ......................................................................................... 30 Exemplification ...................................................................................................................... 32 Information Processing .......................................................................................................... 36 Exemplification of Known Event Populations ....................................................................... 41 Exemplification of Unknown Event Populations ................................................................... 43 Personal Characteristics ......................................................................................................... 43

Research Questions and Hypotheses ................................................................. 45

III. CHAPTER III METHODOLOGY ....................................................................... 50

Overview ................................................................................................................................ 50 Research Participants ............................................................................................................. 50 Experimental Procedures for Testing Economic Communications ....................................... 52 Procedure ................................................................................................................................ 55 Stimulus Material ................................................................................................................... 57

Measures ........................................................................................................... 61

Story Version and Manipulation ............................................................................................ 61 Economic Literacy ................................................................................................................. 63 Arithmetic Aptitude ............................................................................................................... 64 Political Ideology ................................................................................................................... 65 Economic Perceptions ............................................................................................................ 66 Base-rate Recall ..................................................................................................................... 67 Data Analysis ......................................................................................................................... 69

IV. CHAPTER IV RESULTS ................................................................................... 72

H1a: Overall Economic Perceptions ...................................................................................... 72 H1b: Accuracy in Recall ........................................................................................................ 76


iv

H2a: Economic Literacy and Economic Perceptions ............................................................. 76 H2b: Accuracy of Recall and Economic Literacy .................................................................. 82 R1: Arithmetic Aptitude, Economic Literacy, and Perceptions of the Economy .................. 82 R2: Political Ideology and Perceptions of the Economy ....................................................... 89

V. CHAPTER V DISCUSSION ................................................................................. 93

Analysis of Basic Exemplification Theory and Economic Perceptions ................................. 94 Analysis of the Effect of Recall ............................................................................................. 97 Analysis of Economic Literacy .............................................................................................. 98 Analysis of Arithmetic Aptitude and Economic Literacy .................................................... 103 Analysis of Political Ideology within Exemplification Theory ........................................... 104 Theoretical Implications ....................................................................................................... 108 Practical Implications ........................................................................................................... 110 Limitations ........................................................................................................................... 112 Conclusion ............................................................................................................................ 113

BIBLIOGRAPHY ................................................................................................... 117 APPENDICES ........................................................................................................ 133

A. Economic news story manipulations .......................................................... 133 Base-rate Positive: Exemplar Representative ...................................................................... 133 Base-rate Positive: Exemplar Non-representative ............................................................... 141 Base-rate Positive: Exemplar Mixed .................................................................................... 149 Base-rate Negative: Exemplar Representative ..................................................................... 157 Base-rate Negative: Exemplar Non-representative .............................................................. 165 Base-rate Negative: Exemplar Mixed .................................................................................. 173

B. Measures ..................................................................................................... 181


v

ABSTRACT The persuasiveness and influence of exemplars over base-rate information has

been well researched throughout exemplification studies. Few would argue the

narratives, vivid language, and sometimes colorful statements of exemplars have an

overwhelming effect, suppressing the less than vibrant base-rate. Current studies

though have begun to illustrate individual characteristics that have the ability to short-

circuit the effects of exemplars. For instance, recent research has shown an

individual’s arithmetic aptitude plays a dominant role in how they process numerical

base-rate data versus exemplars. This study continues that line of research, focusing

on how subject matter knowledge, in this case economics, can also reduce and

diminish the influence of exemplars.

Through the implementation of a 2 (high economic & low economic aptitude)

x 2 (base-rate positive & base-rate negative) x 3 (representative, non-representative, &

mixed exemplars) experiment the results showed individuals with higher economic

literacy were less influenced by exemplars and held relatively consistent attitudes

toward the overall strength of the economy across all conditions. Individuals

possessing lower economic literacy were found to be significantly more affected by

exemplars as their attitudes and opinions varied across the different manipulations,

indicating their lack of knowledge regarding the subject left them susceptible to

exemplars.


vi

LIST OF FIGURES 3.1 Within-Subjects Design. .............................................................................. 54

3.2 Experimental Design .................................................................................... 55 3.3 Story Manipulation ...................................................................................... 60

3.4 Base-rate Manipulation ................................................................................ 61 3.5 Exemplar Manipulation ................................................................................ 62

3.6 Representative Manipulation ....................................................................... 62 3.7 Graphical Representation of Measures ........................................................ 71

4.1 Economic Perceptions across Story Manipulations ..................................... 73 4.2 Economic Perceptions by Story Version ..................................................... 74

4.3 Economic Perceptions across Exemplar Conditions .................................... 75 4.4 Economic Perceptions, Economic Literacy, & Story

Version ......................................................................................................... 78 4.5 Low Literacy, Economic Perceptions across Story Version ........................ 79

4.6 Economic Literacy, Economic Perceptions, & Exemplars .......................... 80 4.7 Economic Literacy, Economic Perceptions, & Base-rate

Information ................................................................................................... 81 4.8 Economic Literacy, Recall, & Representativeness ...................................... 82 4.9 Arithmetic Aptitude and Economic Perceptions .......................................... 83

4.10 Low Aptitude, Economic Literacy, & Economic Perceptions ................................................................................................... 84

4.11 High Aptitude, Economic Literacy, & Economic Perceptions ................................................................................................... 84

4.12 Low Arithmetic Aptitude, Economic Literacy, & Story Version ......................................................................................................... 85

4.13 High Arithmetic Aptitude, Economic Literacy, & Story Version ......................................................................................................... 86

4.14 High Arithmetic Aptitude, Economic Literacy, & Exemplars ..................................................................................................... 87

4.15 Arithmetic Aptitude, Economic Literacy, & Base-rate Information ................................................................................................... 88

4.16 Political Ideology and Economic Perceptions ............................................. 89


vii

4.17 Political Ideology, Economic Perceptions, & Story Version ......................................................................................................... 90

4.18 Political Ideology, Economic Perceptions, & Exemplars ............................ 91 4.19 Political Ideology, Economic Perceptions, & Base-rate .............................. 92


1

CHAPTER I

INTRODUCTION

Beginning in December 2007 and continuing into the present, the United States

and much of the world are still rebuilding from the financial downturn that destabilized

much of the global economy (Wessel, 2010). Economic stories dominated both the media

and public agendas from 2008 until 2011 (The Pew Research Center, 2011a) with 44

percent of respondents stating they followed economic news stories very closely, far

outpacing the 2012 elections and the war in Afghanistan. In 2009, 20 percent of news

stories covered the economy while coverage of health care was a distant second with only

9 percent. The trend continued in 2010 as the economy finished first among all other news

reporting with 14 percent while the midterm elections only received 10 percent. At the

same time, 67 percent of viewers stated that economic news coverage has increasingly

become more negative (The Pew Research Center, 2011b). This perception of negative

coverage extends across stories related to jobs (75%), financial markets (69%), real estate

(63%), and food and consumer prices (62%) and is relatively even among Republicans

(71%), Democrats (62%), and independents (69%).

Unfortunately, almost five years later communications scholars have seemingly

ignored what role, if any, the mass media may have played in consumer confidence in the

months and years leading up to and following the event. This may be due to the difficult

and ubiquitous nature of the economy. Alternatively, this may be because there seems to

be little to no consensus among economists, politicians, and academics about the current

state of the economy then and now.


2

Even credible sources that everyday citizens rely upon struggled to understand and

explain the economic situation. In February of 2008, President Bush addressed the nation

stating that we were not headed into a recession, but an economic slowdown (MSNBC,

2008). However, eleven months later the National Bureau of Economic Research

declared the economy was indeed in a recession (Abate, 2008). From that point on, the

name and description of the economic downturn was framed differently based on politics,

ideology, and expertise. Vice President Biden on NBC’s Today Show described it as the

“Bush Recession” (Diemer, 2010). On the other side of the spectrum, radio political

pundit Rush Limbaugh called it the “Obama recession,” and cable television political

pundit Sean Hannity blamed the economy on “Obama’s bear market” (Media Matters,

2009). The Economist (2009) expanded the geographic scope of the event, calling it a

"slowdown" to "recession" and then to a “Great Recession.” From there it became a

“lesser depression” (Krugman, 2011). Finally, the events that began as a slowdown ended

as a “second great depression” (Paul, 2008). President Obama’s former economist

speculated there is a 50/50 chance the United States is headed into a “double-dip

recession” (Daly, 2011) which presents the conundrum of whether politicians and the

media help create self-fulfilling prophecies that, in essence, lead us toward or away from

negative economic conditions. As President Obama tries, much like President Bush did, to

alleviate fears of an economic recession, economists, business leaders, and the media are

asking if the recession ever really ended (Tseng, 2010; The Economist, 2010).

When not creating anxiety and panic in the market, the media also creates

confusion by presenting seemingly contradictory information in its reporting. Take for

instance two headlines drawn from a cable news station: “Risk of double dip recession:


3

Unlikely, but rising” and “Economists see no 2nd recession, no real recovery” (Isidore,

2010; Wiseman, 2011). The first story reported that the risks of a double dip recession are

unlikely but that those risks are rising. At the same time, a viewer can feel relief that a

recession is unlikely to occur and immediately feel threatened that the risks of a recession

are rising. The reporting of numerical probabilities is relatively uneven as the reporter

states there is a “one-in-four chance” of a double-dip recession occurring while later on

reporting the risk is as high as 40 percent. The reporter changed the numerical ratio

multiple times throughout the story, assuming readers process the information the same.

In the second story “Economists see no 2nd recession, no real recovery,” the

viewers are presented with a situation in which the state of the economy is ambiguous.

The author does make a consequential statement that indicates the duality or “catch-22”

logic of this study. He further states, “What makes a solution so difficult is that the fear

gripping investors isn’t just a symptom of economic distress; it’s also a cause of it” (p. 1).

The question becomes “where does this fear begin?” Studies show economic perceptions

are derived from two sources: personal experience and the general state of the economy

(Kinder & Kiewet, 1979). Common sense would indicate it is reasonable for fear to come

from personal experience. In fact, economic measures such as the Consumer Sentiment

Index and Consumer Confidence Index try to gauge an individual’s personal experience as

much as possible. However, research (Kinder & Kiewet, 1979) has shown a difference

exists between personal experience and perceptions on the overall state of the economy; in

this case, the fear transfixing consumers and investors must come from another location.

To clarify this situation, if economic perceptions are derived from personal experience,

then individuals gather this information from their employment, family and friends, their


4

community, social networks, and affiliations. However, if personal experience of

economic conditions is significantly different from how people view the general state of

the economy, then Kinder and Kiewet indicated that economic information must come

from another source outside of personal experience. While they did not elaborate on the

nature of the outside source, it is logical that the media most likely has an influential role

in shaping those perceptions.

There is little doubt the news media play a role in forming economic perceptions

(Shah, Watts, Domke, Fan, & Fibison, 1999) while often providing a misleading or

inaccurate representation of the economic “reality” (Goidel & Langley, 2005;

Hetherington, 1996; Patterson, 1993;). Some of this may be explained by Mullainathan

and Shleifer (2005) in their analysis of how the media use a type of spin or “slant” in its

quest to create a meaningful story. Unfortunately, even the most genuine and well-

meaning editorializing of economic news can underscore the true nature of economic

policy, sending the wrong signals concerning the economic impact of those policies

(Alsem, Brakman, Hoogduin, & Kuper, 2008). For instance, a recent Forbes article

“When journalists meet numbers: Not a pretty sight” attacks journalists for their lack of

knowledge on the subject of economics and for reporting stories about which they have

little understanding (Worstall, 2011). In the news article referenced in the Worstall piece,

a newspaper journalist reported that a recent change in tax law would constitute a tax hike

for consumers. The Forbes article refuted each of the reporter’s explanations line-by-line

by focusing on her mathematic acuity and terminology. The Forbes author finally

reasoned that the new law was in fact a tax deduction and not a hike at all.


5

Another problem is that the media often display characteristics of a “herd

behavior” by focusing on buzzwords such as “economic crisis” and reporting on negative

aspects of the crisis and how it might become worse (Shiller, 1995). The ramifications of

“herd behavior,” also known as pack journalism, can lead to a “groupthink mentality that

yields news coverage in a one-sided manner” (Breen & Matusitz, 2008, p. 3). Haiman

(1999) found herd behavior among journalists created stories that are full of rhetoric and

misleading statements which may affect both short- and long-term perceptions of the

audience. Another approach in how reporters cover stories is based upon Mullainathan

and Shleifer’s (2002, 2005) idea that there are two types of media bias: ideological and

spin. They stated that competition among media outlets reduces ideological bias as

viewers amassed opposing views and found the truth most often existed in the middle (p.

531). However, their results also found that reducing ideological bias through competition

leads to greater spin as outlets seek to create memorable stories. Alsem et al. (2008) used

the concept of spin bias to determine that media reports magnified actual changes in the

economy, leading to higher volatility in consumer confidence.

Hatzinikolaou (2010) found evidence such descriptions shock or augment

“economic agents to adjust their expectations of economic activity downwards and to

reduce their spending much more than they should, thus fulfilling these expectations” (p.

144). An additional side-effect resulting from the herd behavior of the media was not only

economic destabilization but also longer lasting business cycle contractions and shorter

expansions. In essence, “the news media are doing more than communicating news about

economic events; they are defining the meaning of the events” (Tims, Fan, & Freeman,

1989, p. 763).


6

However, as the economy is an obtrusive issue (Horner, 2008), based upon one’s

own personal experience or observation, it would seem less likely the media could set the

public agenda, but Zucker (1978) stated it is possible through heavy media exposure.

Perhaps this is due, in part, to the fact that most citizens do not understand the basic

intricacies of how the economy works, or that even when presented with correct

information are unable to decipher the numbers, ratios, and meanings behind the

information. This leads viewers to take media cues and use heuristics to form economic

judgments that considerably influence their perceptions (Mondak, 1993; Pan & Kosicki,

1997). While critics lambast the idea of powerful media effects (Severin & Tankard,

1992), there is little doubt that the way in which the media frames economic information

is a significant indicator of consumer confidence and perceptions of the general state of

the economy (Fan, 1993).

The purpose of this paper is to answer three primary questions regarding the

relationship of the news media and economic news coverage with how people process or

understand economic information presented in the news. The first question focuses on,

whether individuals or audiences are affected by base-rate information or the exemplars

within economic information as reported by the media? Building upon the first question,

the second delves further by asking, what effect does either have on how individuals

perceive the state of the economy and their ability to accurately recall information? And

lastly, do personal characteristics, such as subject-matter knowledge or political ideology,

determine how they are influenced by base-rate information versus the use of exemplars in

news stories? Results from these questions will help answer questions regarding how

methodical the media should be in their reporting of economic indicators while also


7

stressing the importance of how audiences’ basic understanding of the economic

indicators affects their overall perceptions of the economy.


8

CHAPTER II

LITERATURE REVIEW

Overview

The academic field of economic communication is relatively scattered and

discombobulated, preventing it from becoming a more thoroughly studied area. This may

be due in part to the complicated nature of understanding economics, the lack of cross-

academe studies between mass communications and economics, as well as a lack of

shared theoretical backgrounds. Although economics and political science often blend

their fields together and political science and mass communications have similar ties,

there is very little to indicate that a nexus exists between economics and mass

communications. It could be said there is little doubt the mass media have an intricate

relationship with national and global economies based upon the news coverage they

provide on economic information and events. However, the role the media play with

respect to economic perceptions and the manner in which audiences process economic

information are still relatively unknown. The major focus of this chapter is how the media

have historically reported economic news and what role exemplification theory has had on

audience perceptions as reported by the media.

Economic and Financial Information in the Media

Economists, business leaders, and academics hold a general opinion that the media

is relatively inaccurate in its portrayal of economic and financial information (Kurtz,

1990). At the very least, it is generally believed that the media is misleading in the

representation of economic figures. Media coverage of economic information is often

factually reported, but issues with accuracy arise during the explanation of the information


9

or portrayal of facts (De Boef & Kellstadt, 2004; Goidel, Procopio, Terrell, & Hu, 2010).

Worstall’s (2011) aforementioned Forbes article illustrated how a journalist can

accurately report the facts of an economic policy and through the simple explanation of

those facts inaccurately communicate a misleading story. Research on trends related to

economic news and the news media has found that traditionally the news is negatively

skewed (Shen, 2009; Soroka, 2006), and according to Goidel and Langley’s (1995) study

of the New York Times, economic media coverage “bears no relationship with economic

reality” (Shen, 2009, p. 382). If this state of misinformation is a lasting trend, then two

conflicting scenarios exist: (1) it is a violation of what Shen (2009) calls the media’s role

of informing and motivating audiences, or (2) it supports Martin’s (2008) assessment that

bad news is good for political and civic participation (p. 389).

The implications of the dissemination of potentially ambiguous and confusing

information, especially in the current economic climate, are far-reaching. Misinformation

can affect consumer behavior, financial institutions, and political voting (Alsem et al.,

2008; Hatzinikolaou, 2010; Starr, 2010). It is well established that repercussions related to

irresponsible reporting can weigh heavily on today’s society (Matusitz & Breen, 2007).

Within the realm of economic reporting, the reasons for these common oversights are not

completely understood. It is possible that the communication of incorrect information by

the media is a reflection of media ignorance due to a lack of economic experience,

exposure, or even laziness.

Harrington (1989) stated, “there is a common perception among some critics that

the networks’ coverage of the economy distorts reality” (p. 17). Harrington’s study

showed that economic downturns such as inflation, unemployment, and stagnation were


10

given significantly more coverage than economic prosperity and that in nonelection years,

broadcast television was more likely to provide coverage of poor economic conditions

such as unemployment, inflation rates, and the growth rate of the GNP than during an

election cycle. As economic conditions declined, negative economic news stories were

“approximately 34 percent longer and twice as likely to lead the evening newscasts” (p.

34). Evidence suggests that in situations in which the media misrepresented economic

information during a recession, the misrepresentation may have stunted economic

recovery due to reduced spending (Blanchard, 1993). Overly positive or negative news

coverage can lead to heavy consumer spending or saving (Starr, 2010). Unfortunately, it is

not just the tone of the report that can lead to misunderstanding; Starr (2010) was also

quick to point out that even when the media uses “hard data,” there is still a great

probability that viewers will make incorrect inferences.

Research showed the media provides cues on a host of issues that, in turn,

influence an audience (Sniderman, Brody, & Tetlock, 1991). Kuklinski and Hurley (1994)

believe these cues act as heuristics, allowing audience members to make quick mental

judgments and calculations about political issues. In many situations, heuristics provide

individuals the ability to make rapid decisions based upon the small amount of

information which they have gathered or have been exposed to. Ideally, an individual has

enough knowledge or experience with an issue to make an informed decision. However, in

terms of economics, few people have the basic education or understanding to construct

well-informed, and deliberate opinions, (Roos (2007). Because of these situations, it is

“plausible that shifts in campaign coverage of the economy would considerably influence

citizens’ political judgments” (Shah, Watts, Domke, Fan, & Fibison, 1999, p. 918).


11

Heuristics are not directly tied into Starr’s (2010) hypothesis that correct “hard

data” can lead to incorrect inferences. Brosius and Bathelt (1994) explained that base-rate

information is the facts that provide the basic, general information that may include the

causes, consequences, or importance of an issue. Even so, it is easy to see how even the

most accurate, base-rate information can be incorrectly inferred if an individual makes

erroneous or poor associations with other information or is not provided with additional

information from the source as to a cause or reason. Therefore, an example of how base-

rate information could be misinterpreted might be a news report regarding unemployment,

minorities, and drug use. If the story stated that those who frequently abused drugs were

more likely to receive welfare, that in times of economic downturn minorities were more

likely to face unemployment than Caucasians, and that individuals who find themselves

unemployed were more likely to accept welfare, then three factually distinct base-rate

statements taken together might lead one to believe that minorities were frequent drug

users. Without a qualifier differentiating the three it is easy to see how incorrect

correlations could occur using base-rate information. Unfortunately, the difficult nature of

understanding economic factors, indicators, and news may lend itself to false syllogisms

in the media.

Coverage of financial information and the economy is similar in many ways to the

coverage of the local weather forecast. Erikson, MacKuen, and Stimson (2000) stated that

people often watch the news to determine “whether or not tomorrow’s economy will be a

nice economic day” (Goidel, Procopio, Terrell, & Wu, 2010, p. 3). Local media coverage

provides little in analytical analyses, focusing on stock prices and consumer prices, and

often avoids the presentation of economic data unless it is considered a national issue (Wu


12

& Day, 2005). The national news, on the other hand, mixes opinions with sources while

overall providing a broad examination and overview of the economy. In contrast to

national coverage, local affiliates or news outlets’ rare coverage of economic news almost

always contains a regional “spin” that incorporates local businesses in an attempt to

personalize the story and give it a more local feel. Many of these stories are softball pieces

meant to highlight (Goidel et al., 2010) and boost awareness of an issue without

necessarily engaging the audience (Kaniss, 1991).

In a comparison of national and local economic news, Goidel et al. (2010) found

that although both primarily focus on negative aspects, national news media was

significantly more negative than local news. This was true for both television and

newspaper coverage. Respondents in the study indicated they received the vast majority of

their information from national news (54%), followed by local television (17%), and local

newspapers (10%). National print media was more closely associated with economic

conditions and followed changes in the economy quicker than local news outlets.

Results showed that correlations existing between “pocketbook” expectations and

leading economic indicators did not pertain to the state unemployment rate and only

extended to the national level. The overall tone was not flattering; the authors noted that

while most people receive their information from broadcast news, broadcast news and

local newspapers were “not closely aligned with real economic conditions” (p. 15).

This overemphasis in reporting in many ways mirrors how media coverage of

economic news is similar to media coverage of violent crime. As the news often

overstates rising levels of violent crime, the media shows a similar pattern in the

overemphasis of negative economic conditions in the U.S. (Altheide, 1997). Soroka


13

(2006) attributes this very real occurrence to journalists’ asymmetric interests; many

journalists adhere to the belief that audiences value negative information and find it more

informative than positive information. If asymmetrical news is indicative of a society that

holds the government accountable and thus a component of the watchdog role, Soroka

writes that asymmetric news may be an “underappreciated dynamic in both public opinion

and media coverage” (p. 374).

Given the capricious state of economic news coverage in the United States,

citizens’ perceptions and evaluations of the economy are only as good as the information

they are provided. Mutz (1992) believes that if people derive their attitudes from personal

experience and the news media, then any information not directly reflective of the

economic reality will skew public opinion. Because much of the research has centered

around the tone of media coverage in reports on economic news, recent studies

additionally focused on the type of economic information most likely to be covered

(Fogarty, 2005). Basic changes in economic indicators and inflation rarely cause spikes in

the amount of news reported. However, changes in unemployment rates cause significant

variance in the trends and level of reporting among news providers. As unemployment

rises, so does the amount of news coverage given to the subject; this increase in news

coverage sometimes exacerbates the situation as journalists tend to use emotionally driven

stories that magnify the current conditions (Fogarty, 2005). As unemployment decreases,

the media responds much more slowly, and the reporting on improved employment rates

does not reflect the speed at which unemployment improves.

Economic News and the President


14

News coverage of the economy extends well beyond economic cycles of low and

high unemployment, inflation and deflation, and the GDP. Often the state of the economy

plays a prominent role throughout political campaigns and more specifically throughout

presidential elections. The 1992 presidential election is an illustration of the nation’s

economic reality juxtaposed with media portrayal. In July 1990, the nation’s economy

slowed into a recession, and by March 1991 it was over. Hetherington (1996) explored

how the media portrayed an ailing economy during the 1992 election, 20 months after the

recession officially ended. Interestingly, he found that increased exposure to the media

resulted in more common perceptions that the economy was still in a downturn. Thus,

voters who turned to the media for their economic and political news, regardless of

personal experience, were more likely to view the recession as on-going.

Studies (Edwards, Mitchell, & Welch, 1995; Patterson, 1993) during this period

demonstrated that as the media shifted its focus from the Gulf War to the economy, news

coverage was overwhelmingly negative. Even as the economy continued to improve after

the recession was officially over, the media emphasized the economic “crisis” facing the

nation. Patterson (1993) showed news coverage with references to the economy was 90

percent negative and 80 percent of the coverage targeted the Bush administration.

Hetherington (1996) believed “had voters cast their ballots based on the actual condition

of the national economy in 1992, they more than likely would have returned George Bush

to office” (p. 373).

Some researchers believe that negative economic reporting grew from President

Bush’s increase in approval ratings garnered during the Gulf War (Stevenson,

Gonzenbach, & David, 1991). Although this may seem conspiratorial, Blood and Phillips


15

(1995) referred to the concept as the “adversarial press hypothesis” (p. 6). Supporting this

hypothesis is evidence to suggest the nature, positive or negative, of national news stories

is highly correlated with presidential approval ratings (Graber, 1993). According to Blood

and Phillips (1995), the potential dominance of the media’s focus on the economy over all

other issues may have “primed” news audiences’ perceptions regarding the president

(Iyengar & Kinder, 1984). Under priming, as news audiences are exposed to specific

messages, they cognitively process stimuli and construct future evaluations based upon

the stimulus (Hetherington & Rudolph, 2008), and because “most citizens are not

generally interested in politics, they cannot assimilate the complex political information

with which they are inundated” (Hetherington, 1996, p. 375). That citizens are unable to

cohesively understand the information forces them to synthesize information accessible

from memory. This synthesized information becomes the basis on which evaluations of

the president are made.

In expanding on the 1992 election, Shah et al. (1999) examined how economic

issues and the president were covered by the media during the 1984, 1988, 1992, and 1996

campaigns. Not unexpectedly, they found incumbent presidents garnered significantly

more economic coverage than their opponents did. Also, during times of economic

prosperity, the sitting president received not only more economic coverage but also more

positive coverage. The inverse is true as well. An incumbent president presiding over

economic downturns was more likely to receive substantially more negative coverage.

This is evident in the 1992 presidential election and from an unsubstantiated observation

of the 2012 presidential election.


16

Four years following the 1992 presidential election, the economy continued to

grow at record levels, and throughout the 1996 presidential election, negative opinions of

the economy were relatively low. This led Domke et al. (1997) to observe that during

periods of a healthy economy people are more likely to employ retrospective attitudes

toward the economy while prospective attitudes are more prevalent during a downturn or

recession. Domke et al. (1997) believed that in times of economic prosperity

(retrospective) individuals focus on how well they are doing relative to how well they

have done in the past. This type of rationale is reflected in measures of consumer

confidence in which questions ask how people feel they are (better or worse) financially

than they were the year prior. However, in times of economic downturn or recession the

authors felt people became more forward thinking (prospective) about economic

conditions and the general direction of the economy. As times become better, people tend

to make comparisons to the past, perhaps taking a glass is half full approach; and as times

grow worse, people focus on the direction in which they are going while looking for signs

of economic turnaround or changes.

Economic News and the U.S. Electorate

Research has shown media coverage of economic news is often misrepresentative

of economic conditions and can play a substantial role in presidential elections, but little

has been discussed so far as to how voters form decisions based upon economic news

reporting. In examining Hetherington’s (1996) aforementioned statement regarding

voter’s cognitive processing of political information, a conflicting question arises: “Do

people base their voting and consumer behaviors on the environmental conditions around

them or are their attitudes and opinions shaped by the general state of the nation’s


17

economy?” These questions are studied extensively within economic and political

research culminating in Kinder and Kiewiet’s (1979, 1981) seminal findings. In the

context of economic voting, the authors identified two types of economic voters: egotropic

or “pocketbook” voters who focus on the instantaneous, palpable economic events that

shape their lives (Bloom & Price, 1975) and sociotropic voters who focus on the general

economic welfare of the nation and base their voting decisions on the current economic

climate and the party in power. A better way of understanding this idea is that

“pocketbook voting reflects the circumstances and predicaments of personal economic

life; sociotropic voting reflects the circumstances and predicaments of national economic

life” (Kinder & Kiewiet, 1981, p. 132).

Understanding how people potentially act on their perceptions of the economy

helps explain the rationale behind voting behavior. However, the 1992 presidential

election and simultaneous negative media coverage of the economy create a situation in

which an individual’s economic perceptions and the economic reality are called into

question. Kinder and Kiewet (1979) found a difference existed between voters’ personal

economic experiences and their subjective opinions of the overall state of the economy. In

deciding on the qualifications and economic rationale behind their voting decision, voters’

perceptions of the economy significantly outweighed their own personal economic

experiences. Surprisingly, the way through which individuals applied sociotropic versus

egotropic voting decisions was based upon the level of office they were voting for (Mutz,

1992). For individuals voting in a presidential election, the media played a much greater

role in shaping perceptions of the current state of the economy. However, as the elections

moved down the ballot, different sets of criteria are taken into account. In statewide and


18

local elections the results showed people indirectly allowed pocketbook issues to enter

into their decision-making. As the races came closer to voters in proximity, voters began

to cognitively associate closer economic conditions to those races.

Kramer (1983) attempted to explain the discrepancies between economic

perceptions and the economic reality of individuals living under the same economic

conditions by attributing the differences to partisanship and measurement error. Gerber

and Huber (2010) indicated partisanship was shown to have positive effects on

perceptions of the economy when a particular individual’s party is in office. The results

offered complimentary evidence toward the predictive nature of economic indicators

because the authors found partisanship exerted less influence on perceptions of how the

economy is currently performing and more on perceptions of how a particular party would

be able to bring about substantial change in the economy. This can be explained in today’s

political climate; Republicans are more likely to have doubts or negative opinions of the

economy when the country has a Democrat as president or has a Democrat controlled

Congress. Consequently, partisans’ biased perceptions of the economy are based more on

the economic competency of a particular party and less on their current personal economic

status (Gerber & Huber, 2010). This does not mean though that partisan individuals and

their biased perceptions can’t be controlled in a survey or experiment.

There is evidence that partisan bias in economic surveys can be manipulated

through the use of questions that ask about local gas prices (Ansolabehere, Meredith, &

Snowberg, 2011). When asked about the economy, individuals’ subjective evaluations

may include perceptions of unemployment of friends and family as well as partisanship

resulting in measurement error. However, the authors found perceptions of gas prices are


19

based upon an individual’s most recent trip to the gas station and are not reflective of

partisan ideals or biases. Similar to the economy, gas prices are obtrusive and highly

salient issues among most individuals, yet neither party identification nor socioeconomic

status had any significant effect on perceptions of gas prices.

In contrast to this, other researchers offered that individuals have different

methods of evaluating the economy based on varying preferences (Kinder, Adams, &

Gronke, 1989). In these situations, everyday economic indicators such as wages, salaries,

gas prices, and groceries offer objective assessments through which people can judge

economic performance (DeBoef & Kellstadt, 2004).

Parallel with perceptions of the current state of the economy, there are

expectations “that party control has real effects on economic policy and performance”

(Ladner & Wlezien, 2007, p. 591). Ladner and Wlezien’s research reaffirms the reasons

why economic reports such as the consumer confidence and consumer sentiment indexes

ask questions regarding consumer opinions on goods, prices, unemployment, and gas

prices a year in advance. Voters use future economic indicators as predictive tools to

indicate which party will have the most positive impact on the economy 12 months in

advance, as they realize that meaningful effect will not occur in the immediate period

directly following the election. MacKuen, Erickson, and Stimson (1992) stated that

“citizens are primarily concerned with the economic future rather than the economic past”

(p. 314).

In follow-up studies on sociotropic voting, Ansolabehere, Meredith, and Snowberg

(2008) asked the question, “If a voter’s evaluation of the overall state of the economy does

not include their personal economic experience, where does their information about the


20

economy come from?” (p. 3). Through the use of a preliminary ANES Pilot (2006) study,

the authors found that those with high exposure to radio news were more likely to

perceive the country as having lower unemployment rates. Exposure to television news

was shown to be correlated with perceptions of higher unemployment rates. This partially

supports Hetherington’s (1996) hypothesis on the relationship between the media and

voter’s evaluations of the president during the 1992 presidential race. In this scenario,

heavy exposure to the news and media was significantly correlated with negative

retrospective evaluations of the nation’s economy. Unfortunately, due to the nature of the

ANES Pilot (2006) study, questions regarding individual’s acquisition and processing of

economic information could not be ascertained, leaving a void in the literature.

The connection between voters and consumers and how their perceptions of the

economy influence their attitudes and opinions regarding the government and voting is

robust within economic and political science literature. Following along these lines, the

relationship between the public and the media is well established with the media seen as

influential in telling people not only what to think about but also how to think (McCombs

& Shaw, 1972; Entman, 1993). Consequently, McCombs and Shaw’s agenda setting

research may prove instrumental in answering the question Ansolabeher, Meredith, and

Snowberg (2008) asked: “If a voter’s evaluation of the overall state of the economy does

not include their personal economic experience, where does their information about the

economy come from?” (p. 3).

That question can be broken down into three parts in an effort to explicate in more

detail what the ANES Pilot study (2006) could not answer: (1) where do perceptions of


21

the economy come from, (2) how do individuals acquire economic information, and

perhaps most importantly, (3) how do individuals process economic information?

Zucker (1978) claimed the media was more influential on the public agenda when

exposure or “direct experience” of the issue was limited, inherently finding those issues to

be unobtrusive. However, the economy in many ways presents a conundrum. During the

1980 and 1988 presidential campaigns, working class individuals were less likely to view

inflation and unemployment rates as highly salient issues due to the positive economic

environment (Southwell, 1996). Because of the climate and because individuals viewed

those issues as low importance, voting during those two elections was lower. In contrast,

the 1984 and 1992 presidential elections delivered opposite findings. As the economic

climate began to sour and concerns regarding inflation and unemployment rose, both the

saliency of those issues and voter turnout were greater.

These two studies show the duality regarding the nature of economics. Zucker

(1978) found people’s personal experience toward an obtrusive issue makes the media less

influential. However, Southwell (1996) showed that as the economy grew healthier it

became less of an issue among the public and was less salient as well. To better

understand this, Lang and Lang (1981) explained that as an issue is deemed of greater

importance by a proportion of the population, media influence wanes. When an issue is

seen as less important or as affecting a smaller portion of the population, the media’s

influence over that issue grows. Simply, the more an issue affects a group of people, the

more they’ll discuss the issue together, limiting the media’s role in the discussion.

In comparing Lang and Lang’s (1981) study with Kramer (1971), the answer is not

so clear. According to Kramer, individuals rationalize their voting decisions on their


22

perception of the general state of the economy and not on personal experience. Therefore,

the idea of “pocketbook” voting becomes secondary to how people view the national

economic climate. Lang and Lang's research is supported when voter perceptions of the

general state of the economy are drawn from others around them and are enhanced by

their own personal experience. Building upon Lang and Lang’s hypothesis, if people’s

perceptions of the general state of the economy come from external sources outside of

personal experience or social networks or if an individual has limited contact or

engagement within social environments, then the chances of the media being more

influential on economic voting attitudes and behaviors increases.

Throughout the literature there is a general consensus that people judge the

economy through personal “pocketbook” observations or upon perceptions of the general

state of the economy. While individuals can objectively feel and observe the effects of the

economy on their lives, they can only subjectively view the national economy.

Reeves and Gimpel (2011) detailed these observations:

No one experiences national conditions. The state of the economy is but a set of

summary measures averaged across thousands of communities and millions of

individuals. Individuals do not directly experience the national gross domestic

product or the national unemployment rate. They do have personalized knowledge

of economic conditions obtained through conversations with family, friends,

coworkers, and neighbors. In their daily drives to work they might see factory

closings, foreclosure signs being posted, or gas prices on the rise. They see nearby

families disintegrate under the pressure of economic stress. These are geotropic

conditions. (p. 3)


23

As studies downplay the “pocketbook” voter aspect, it is important to remember

many of these studies were conducted before the “Great Recession” that began at the end

of 2007. Geotropic conditions appear to be more common as people watch their towns and

cities struggle under the current economic conditions, and due to the recency of the

continuing economic stagnation, few studies are available to examine these conditions.

Fifteen years before the recession, Mutz (1992) stated there was a difference between

reading about unemployment in a newspaper and being handed a pink slip at work. The

psychological and personal impact, he believed, was so great that the gravity of such an

event is inescapable

As the nation’s economy continues to dominate news coverage and create hyper-

saliency of the issue, individuals may be turning back toward their communities to look

for support and understanding from social groups (Mutz & Mondak, 1997). Surprisingly

though, Hopkins (2011) determined the wealth of an individual did not play a significant

role in their perception of the economy. Poorer citizens tended to think of themselves in

terms of those around them and did not make significant comparisons across wealth gaps.

Reeves and Gimpel (2011) did report that individual and local circumstances had an effect

on participants’ perceptions of the national state of the economy. As most studies before

the “Great Recession” focused solely on unemployment rates, inflation rates, and the

GNP, the newer study included foreclosure rates, a variable rarely seen or tested in

economic studies. The results showed high correlations. Increases in foreclosure rates

produced an increase in negative perceptions of the overall state of the economy. The

authors hypothesized the 2008 presidential election was a referendum of the national

economy; the 2012 election may be a referendum of the local economy.


24

The Reeves and Gimpel (2011) study highlights a potential shift toward personal

and community perceptions of the economy and away from national media coverage

focused on the overall national economy. Group membership through social dilemma

studies that focused on “the public good problem (in which the individual must decide

whether to contribute to a common resource) and the commons dilemma (in which the

individual must decide whether to take from a common resource)” (Brewer & Kramer,

1986, p. 543) indicated that when an issue is framed as a public goods problem (in which

the individual must decide whether to contribute to a common resource, p. 543) or a

commons dilemma (in which the individual must decide whether to take from a common

resource, p.543), there are significant differences in how individuals think about the

saliency of an issue. Part of this is based upon strong collective identities versus individual

identity positions or one’s social identity within the group (Park, 2002). Therefore,

political and economic issues and decision-making can be framed in a variety of ways that

are pertinent to communities or the individual.

Economic News and the Media Elite

If the media have the ability to frame issues in ways that make them more

prominent, then logically one might ask who chooses the frames. As discussed earlier,

some journalists erroneously frame economic information based upon a general deficiency

in understanding the issue (Worstall, 2011). But it has also been shown that political

pundits and media elites have placed political blame for and provided descriptive

impressions of the economic downturn through the way they framed the issue (Diemer,

2010; Krugman, 2011; Media Matters, 2009). Nadeau, Niemi, Fan, and Amato (1999)

acknowledge there is an intersection of economic and political evaluations generated


25

through what they say is a media “expert-driven” environment. Unfortunately, this media-

dependency model relies on audiences believing economic figures and information are

“correctly reported” and expectations that the audience and experts are rationally formed

(Haller & Norpoth, 1994; Nadeau, Niemi, Fan, & Amato, 1999, p. 110;). Discovering the

nature of economic rationality is not an easy undertaking as there are disagreements as to

whether viewers cognitively process economic information using retrospective or

prospective evaluations. Norpoth (1996) stated most individuals evaluate their current

economic situation when processing information. MacKuen, Erikson, and Stimson (1996)

found people apply the information in a predictive manner, focusing on expectations of

the economy.

Evidence shows that prospective evaluations of the general public “are not likely

to be derived primarily from personal experiences or directly from the objective condition

of the economy; rather, elite judgments and media accounts play a crucial role in their

development” (Nadeau, Niemi, Fan, & Amato, 1999, p. 114). The media, however, are

able to blend retrospective news, for example the Gulf War and foreign policy, with

prospective news, the economy, in effect forcing an audience to separate present

observations from future predictions. The complicated nature of economic information,

coupled with an audience's expectation that the information is reported correctly, allows

media elites to define and frame the issues. Some scholars have shown evidence

supporting the idea that political elites and the news media influence the public through

cues (Sniderman, Brody, & Tedlock, 1991) which act as heuristics or mental shortcuts

when processing information (Kuklinski & Hurley, 1994). This perhaps explains why


26

some research has shown that audiences and voters form temporary and often incorrect

judgments of the economy (MacKuen, Erikson, & Stimson, 1992) and related events.

Much of the research that has focused on economic conditions in the news

concentrated on the individual or the “workingman’s” well-being. Very few examine

media coverage of corporate economics and how corporate interests and welfare work in

accordance with media reports. Kollmeyer (2004) sought to scrutinize the relationship

between corporate portrayals in the media and the traditional economic news story. A

recurring theme appeared in which the author suggested “powerful social forces, arising

from the structural relationship between the news industry and the capitalist economy,

help align the norms of economic journalism with the general interests of the corporate

community and investors” (p. 450). From this theme, four perspectives were identified in

the media coverage of the economy: the news media emphasize bad economic news, the

news media “de-politicize” the economy, the news media emphasize the views of the

elites, and the news media rely on “official sources” of information. This reinforces

Croteau’s (1998) argument that while the journalists may be perceived as the “liberal

press,” they tend to have center-right views when it comes to economic issues.

Kollmeyer (2004) argues the relationship between the power elite and national

news media has such strong ties that they share many political views with the business

community. Because of this relationship, national media outlets focus on corporations and

the issues they face while virtually ignoring the common worker and laborer. The

resources at a corporation’s disposal and the relationships cultivated over time help

explain why “official sources” in news stories are rarely union laborers and factory


27

workers. The data provided evidence toward a “pro-business” economic environment in

newsrooms while nationally presenting overwhelmingly negative economic conditions.

It is apparent a relationship, arguably shaky, exists between the two fields of

economics and mass communications. And while research indicates personal experience

and observation are an important factor in people’s economic perceptions, evidence also

suggests that the media play a very crucial role in those perceptions as well. The next

logical step is to determine the narratives through which the media frame economic issues

to individuals and how those individuals accurately recall economic reporting.

Economics as an Obtrusive Issue

Due to the obtrusiveness of economic issues and the economy, the subject is one

that is highly significant to most individuals (Zucker,1978). Whether obtrusiveness is due

to heavy exposure from media coverage or an individual’s social and economic

surrounding is up for debate as there is little consensus among researchers as to which

factor has a greater impact. Horner (2008) interpreted the results from Zucker’s (1978)

agenda-setting study as he perceived an issue to be obtrusive when that issue is derived

from personal experience and observation. The greater the degree to which feelings,

attitudes, or opinions are based upon personal experience or observation of an issue

essentially lessens the media’s effects because of the strength of that relationship.

Therefore, a less obtrusive issue allows for stronger media effects.

One emerging issue is disagreement over the operationalization of obtrusiveness.

Zucker (1978) pointed out obtrusiveness is one’s direct experience with or observation of

an issue. Lang and Lang (1981) took the approach that obtrusiveness is based upon how

many people in a given population have been affected by an issue. Others have stated


28

someone can be just as affected indirectly by an event or issue as someone directly

affected by it (Curtin, 2003). Horner (2008) stated additional problems arise in that only

experience or observation and the media have traditionally been used in operationalizing

obtrusiveness. That potentially excludes the influence of interpersonal communication.

The problem with potentially isolating interpersonal communication is that it may be a

fundamental aspect of one’s personal experience. This raises the question of whether or

not interpersonal communication directly or indirectly affects one’s experience or

observation.

Economic Literacy and Assessment

Research (Walstad & Rebeck, 2002) has shown that economic knowledge plays a

vital role in explaining perceptions of public opinion, making it imperative that the media

provides thoughtful and understandable analyses when explaining the economy.

Unfortunately, Krause and Granato (1998) found media reporting of economic news

makes inherent assumptions “that the information or knowledge of the respondents is

homogeneous" (Walstad & Rebeck, 2002, p. 922), and sadly, a 1992 macroeconomic

study (Walstad, 1997) showed that less than one half of adults could correctly answer the

economic test questions. More recent studies (Blendon et al., 1997; FED, 1998; Harris &

Associates, 1999) continue to indicate the population is significantly deficient in their

awareness and understanding of the economy, affecting their “capability of understanding

major economic issues” (Walstad & Rebeck, 2002, p. 924).

Understandably in their study (Walstad & Rebeck, 2002), demographics and

socioeconomic status played an important role in people’s economic literacy. Age has

been shown to be a significant contributor in a person’s understanding of economics but


29

as individuals get older their age becomes less of a factor, peaking between 40 to 60 years

of age; males ranked higher than females while whites performed better than non-whites;

and as one would assume, level of education played a significant role. The results were,

however, mixed on the role that political affiliation played. The overall findings of the

research showed that higher levels of economic knowledge and literacy resulted in those

individuals’ views most likely mirroring the same views and perceptions as those of

economists on economic issues.

An interesting perspective put forth by Roos (2007) described the inherent risks

and pitfalls of the “layman’s” views of economics and economic policy. He stated that

with regard to everyday macroeconomic beliefs, most people do not have a strong grasp of

nor try to understand how the economy works. However, their ability to affect matters

and to impact issues they are not familiar with creates economic casualties based upon an

inefficient system of people who share the same incorrect attitudes and perceptions.

Blinder (1997) stated that politicians, many being non-experts themselves, are subject to

and influenced by the views of other non-experts. Birney and Shapiro (2005) took this

idea one step further and found that politicians are exposed to political exertion and direct

pressure by many non-expert, economically illiterate constituents. Consequently, voters

who do not have a basic understanding of how economics work can influence economic

and public policy through grassroots interaction, campaign contributions, and elections.

Walstad (1997) found that with regard to economic issues and positions, economic

knowledge had a significant effect on public opinion because “people will state an opinion

about an economic issue despite having little knowledge of the subject” (p. 203).


30

Because of these implications, the media has a responsibility and should be aware

that the tone, language, and information used in economic news stories do indeed play a

role in how people perceive the economy. Eccles and Nohria (1992) perhaps phrased it

best when they wrote, “The way people talk about the world has everything to do with the

way the world is ultimately understood and acted in” (p. 29). Ferraro and Pfeffer (2005)

described how economic language “evokes certain associations, certain motives, and

certain norms” (p. 16) and found that simple priming was able to produce differences

among participants, triggering “competitive behavior, beliefs in the efficacy of markets,

and individual self-interest” (p. 16). One study (Lerner & Tetlock, 1999) illustrated how

economic language could alter perceptions of accountability while another (Ratner &

Miller, 2001) demonstrated how economic language spurred participant’s self-interests.

According to Neikirk (1991), the blame can be placed on both the media and the

public; he stated the media treated the economy as if it were “teaching college calculus to

bored third graders” (p. 3). In essence, the media fails the public by presenting

complicated illustrations and graphics and then does little to explain them, often offering

just a blurb and little clarification. On the flip side, the public focuses on more

entertaining stories that are dramatic and compelling, avoiding complex information

(Neikirk, 1991).

Theoretical Framework

Media Effects and Public Opinion

While the connection between economics and political science has been clearly

articulated and studied long before the advent of mass media, the field of research that

examines the relationship between economics and mass communications is not well


31

developed. However, the shared association of political science may help in the

establishment of a link between media effects and economic behavior to fill the missing

gap.

The history of modern day economic study goes back almost 200 years before

media effects hypotheses were discussed. Citizens’ understanding of economics during the

1700s was rudimentary at best as few publications existed on the subject, and news

distribution was limited. After the advent of mass communication technology,

publications, reporting, and research on economics enjoyed a much wider distribution

network. Since the implementation of these developments, the complex nature of

economics has only become more multifaceted, and audiences’ understanding of the

subject has only slightly improved.

Understanding how the media shape public perception and the impact they have on

the “public’s policy priorities” (p. 16) has quickly become a growing body of research

(Cook et al., 1983). The strength of the media is focused on its ability to “construct social

reality” (McQuail, 1994, p. 331) while creating a “cognitive representation that codifies

expectations and assumptions about familiar events” (Elliott & Hayward, 1998, p. 16). In

exploring the relationship between the media and the economy, the media’s subjective

portrayal of the economy can essentially create a scenario in which journalists use the

truth to tell a false, misleading, or inaccurate story. While the economic information

within a news article is often both factual and correct, the inaccurate representation of

attributes of the economy may adversely depict economic indicators, providing a false

portrayal of those indicators.


32

Exemplification theory may help us to better understand how this scenario is

possible. Multiple studies (Aust & Zillmann, 1996; Brosius, 1995; Gibson & Zillmann,

1994; Zillmann, 1999, p. 70) have shown that even the most precise reporting of base-rate

information within a story may be disrupted and misconstrued by exemplars that are not

representative or reflective of the base-rate facts provided. These exemplars tend to be

more illustrative (Koehler, 1996), both sticking out and remaining in audience’s memory

longer, and potentially detract a viewer’s understanding as to the “reality” of the situation.

Exemplification research has explored this often-times complicated relationship between

base-rate and exemplars and has tried to determine how people process the information

and the effects it has on their overall perceptions.

Exemplification

The core foundation of exemplification theory is built upon the idea that

information provided within news coverage consists of two types of figures: base-rate and

exemplars (Gibson & Zillmann, 1994). Base-rate information, while largely precise and

factual on its face, is often regarded as an inferior instrument of influence and

persuasiveness in comparison to exemplars (Brosius, 1995). Typically it is the

fundamental, numerical data that provides just basic facts. It may incorporate proportional

aspects of a news story or report on basic, relevant information that is not detailed or

expanded upon. Base-rate information was presented precisely as well as vaguely in

exemplification experiments, but in each case, base-rate data was inconsequential at best

when perceptions were formed around exemplars (Zillmann, Perkins, & Sundar, 1992).

Base-rate data does possess an informational-utility (Zillmann, 2000) that increases with


33

“a) the perceived magnitude of threats and incentives, b) the perceived likelihood of their

materialization, and c) their perceived proximity in time” (Zillmann, 2002, p. 9).

One hypothetical example of base-rate is if a university newspaper simply stated

that an increase in university tuition caused 40 percent of students to attend local and

community colleges; the event within the story revolves around the 40 percent of students

now attending community college or the (unstated) 60 percent that remained enrolled. The

statement that tuition was increased provides general information and in some semblance

a direction regarding the present state of tuition and how it has potentially affected

students attending or looking to enroll at any particular university. Second, the figure 40

percent is simply a way of quantifying the percent of students affected by the increase.

Both are purely objective in their summation of the state of tuition and its effect on

students.

Exemplars, on the other hand, may consist of quotes or statements given by

sources within news reports, descriptions representing the information, or “emotion-

evoking imagery… that create perceptions and dispositions…” (Zillmann, 1999, p. 70).

They are used to highlight and illustrate certain aspects in an effort to paint a narrative for

the viewer. Through an investigation of the effectiveness of exemplars, Aust and Zillmann

(1996) found emotional exemplars, even those employing threats, were more effective

than exemplars projecting no emotion. Exemplars often are the drivers behind a news

story, and research has shown exemplars are often more persuasive on individual

perceptions than base-rate information (Brosius & Bethelt, 1994; Gibson & Zillmann,

1994; Zillmann, Gibson, Sundar, & Perkins, 1996; Zillmann, Perkins, & Sundar, 1992).


34

Going back to the previous story regarding tuition increases and its effect on

students, a writer may provide impartial base-rate information about the state of higher

education in a headline such as “Rise in tuition forces 40 percent of students to attend

local community colleges.” However, the story could easily include exemplars that

contradict the headline, such as a student providing a statement saying that the tuition

increases were less dramatic and had less effect than the headline publicized. For example

the student may be quoted in the story stating, “While tuition increases do affect us and

our parents’ financial situations, I don’t know anyone who has decided to attend

community college because of it. Most of our student loans cover the costs.” The

exemplar could also easily support and reflect the base-rate information as well. The

headline tells the audience everything they need to know about the state of higher

education, but it provides little emotion or commentary about any real effects. The

exemplar, while providing no numerical data, provides the reader with a better

understanding from a student’s perspective.

Zillmann (1999) revealed that when accurate base-rate information, very similar to

the earlier Forbes story, is reported along with “qualitatively distorted, atypical cases,”

that is to say exemplars (p. 70), the result is that the base-rate’s influence is greatly

diminished. Other factors that diminish the influence of base-rate information are the

engagement of audience’s emotions through the use of exemplars, as well as other

descriptive or disproportionate messages that reduce attention down to abstract

information. While exemplars are capable of all these things, they also draw in viewers,

capture their attention, and increase the saliency of an issue (Aust & Zillmann, 1996;

Brosius, 1996; Brosius & Bathelt, 1994).


35

Exemplars, however, go beyond emotional or elaborative descriptions to make a

story more accessible. While they do illustrate aspects of a story and help individuals

better understand the topic or issue (Burns,1992; Tversky & Hemenway, 1984; Zillmann

& Brosius, 2000), exemplars are not used “to show an individual experience but to

exemplify a societal problem with an individual experience” (Brosius & Bathelt, 1994, p.

49). Therefore, this conceptualization of exemplars can lead to two potential judgments

from an audience.

The first judgment states that audiences generalize the exemplars to be “typical of

all people affected by the problem” (Brosius & Bathelt, 1994, p. 94; Hamill, Wilson, &

Nisbett, 1980). The second states that audiences, when examining the problem, will

overestimate its importance (Brosius & Bathelt, 1994; Zillmann, Perkins, & Sundar,

1991). Individuals can then compare the exemplars with their own aggregately stored

direct experiences to determine how typical or atypical a situation is. While Zillmann and

Brosius (2000) stated that these comparisons, based upon personal experience, potentially

limit one’s understanding and knowledge, the assessments are generally reliable (Bates,

2010).

However, it’s not possible for individuals to carefully process all the information

they acquire. Petty, Cacioppo, Strathman, and Priester (1994) stated that if everyone paid

close attention to all the messages and signals received just in the morning alone, then “if

you ever made it out of the house in the morning, you probably would be too mentally

exhausted to do anything else!” (p. 118). Because of this, people take cognitive shortcuts,

which according to Zillmann (2002) may be understood by looking at how people process

information.


36

Information Processing

Even as audiences pay attention to persuasive and informative messages, there are

incidences in which they shut off the information valve, effectively ignoring persuasive

communications. In an effort to understand the cognitive processes that are activated

under exemplification, research determined that three mechanisms exist: priming, the

availability heuristic, and the representative heuristic (Zillmann, 1999; Zillmann & Gan,

1996). Priming works under the premise that exposure to certain stimuli will produce

future evaluations or reactions (Hetherington & Rudolph, 2008). Most studies include

recency as an indicator of priming through the presumption that issues or objects recently

accessed from memory are more likely to stimulate cognitive functions (Druckman,

2004). Currently, there is an on-going debate as to whether priming is a mechanism of

accessibility (Price & Tewksbury, 1997) or an aggregate-level phenomenon (Miller &

Krosnick, 2000). Real world examples of priming include the use of stereotyping to form

quick judgments of others (Bargh, Chen, & Burrows, 1996).

Researchers (Bargh & Pietromonaco, 1982; Iyengar, Kinder, Peters, & Krosnick,

1984) have stated that “priming occurs when media attention to an issue causes people to

place special weight on it when constructing evaluations…" (Miller & Krosnick, 2000, p.

301). Higgins and King (1981) stated that specific events potentially enhance and

increase the accessibility of an issue or object within an individual’s mind, and during this

incident priming occurs (Miller & Krosnick, 2000). This “accessibility hypothesis”

(Iyengar et al., 1984; Miller & Krosnick, 2000) has been reaffirmed by Price and

Tewksbury (1997) who argued:


37

First, a media message renders one or another construct applicable, and that

construct – say unemployment – is activated. By virtue of its activation, and in

direct proportion to the recency and frequency of its activation, that construct

remains temporarily accessible. Subsequently, when a person is called on to

evaluate the performance of the president, unemployment is likely to be activated.

Unless screened out as inapplicable to the evaluation, it will be used as a basis for

making a judgment of presidential performance (p. 197).

Essentially, an individual subconsciously evaluates an issue based upon that issue

being activated within their memory and not by choice. Miller and Krosnick (2000) said

audiences are “victims of the architecture of their minds” (p. 302) as information is

accessed internally and effortlessly. Reinforcing this assumption, Holbrook and Hill

(2005) stated that while the media may not be proficient in telling viewers what to think,

they are good at telling viewers what to think about while steering their attention to certain

aspects in which viewers are able to make evaluations. Therefore, individuals, instead of

using all available information, use only the bits and pieces of information that are most

accessible. Based upon Anderson’s (1983) research, this type of knowledge storage occurs

from an associative network model that is formed around the idea that information is not

stored within the memory hierarchically, but rather as nodes and links within memory

(Holbrook and Hill, 2005, p. 279). As information is “activated” within memory, the

likelihood of similar issues or nodes being activated increases. This phenomenon,

according Holbrook and Hill, leads to a spreading of activation that causes individual

changes in issue accessibility and public opinion.


38

Similar to priming, the availability heuristic is focused on the ease of accessibility

of certain responses from memory, but it is less attentive to time constraints (Kahneman,

Slovic, & Tversky, 1982; Tversky & Kahneman, 1973). In some situations, individuals

may use faulty logic or imprudent reasoning because other information was not available

or accessible. At other times, the only information people have at their disposal may be

incorrect. For example, if they had never been on an elevator before and another

passenger told them that if the elevator cable were to snap, they should both jump right

before the elevator hit the ground for safety. The new rider would most likely jump based

on a judgment from the only information accessible to him, whether accurate or not.

In essence, within each individual there is a hierarchy that allows each of us to

recall memories and associations easier than others (Busselle & Shrum, 2003; Weaver,

2009). Weaver (2009) stated “the easier a thought is to remember, the more influence it

has, which means even the most disinterested of introspections are biased by the media’s

all-encompassing drive toward parsimony (p. 7). Therefore, information that is readily and

easily accessible within memory will have disproportional influence over an individual

(Zillmann & Brosius, 2000). Tversky and Kahneman (1973) found that distinctive

characteristics of a message which make information more easily remembered are “more

likely to exert greater influence on judgment than would seem rational” (Zillmann &

Brosius, 2000, p. 43).

Two particular characteristics that make messages stand out are the vividness and

salience of an issue. Under vividness, individuals experience a sort of “color commentary”

that creates a presentation where “information may be described as vivid, that is likely to

attract and hold our attention and to excite the imagination to the extent that it is (a)


39

emotionally interesting, (b) concrete and imagery-provoking, and (c) proximate in a

sensory, temporal, or spatial way” (Nisbett & Ross, 1980, p. 45). Taylor and Thompson

(1982) found that during the initial processing stages, vivid information is more easily

accessed and retained and during later stages is more often remembered.

The second characteristic, salience, is conceptually related to vividness. Zillmann

and Brosius (2000) conceptualized salience “as an orienting motive that directs attention

toward specific facets of the stimulus environment, usually but not necessarily those that

have personal relevance” (p. 45). Salience focuses on the concept that the strongest

influence on judgment is those specific elements of an environment most likely to grab an

individual’s attention (Taylor & Fiske, 1978; Zillmann & Brosius, 2000). Any element

that an individual deems to have personal relevance is generally going to be more salient

within the mind.

In the context of exemplification theory, vividness and salience are more likely to

shed light on why exemplars are often more influential than base-rate information.

According to Zillmann and Brosius (2000), while base-rate information is often more

comprehensive and valid due to its quantitative, barebones nature, there is a general lack

of vividness and salience attached. Exemplars, because of their “emotion-evoking

imagery…” (Zillmann, 1999, p. 70), tend to be more persuasive, vivid, and salient.

Paivio (1971, 1986) hypothesized that exemplars are remarkably influential

because they “consist of concrete imaginable stimuli that evoke both a visual and verbal

representation in memory; whereas, summary-type descriptions of reality would only

leave a verbal memory trace because of their abstract contents” (Brosius, 2000, p. 11). In

a study focusing on the persuasive nature of statistical data versus exemplars, Limon and


40

Kazoleas (2004) found individuals exposed to exemplars were significantly less likely to

produce counterarguments and cognitive responses than those individuals exposed to the

statistical data. Their research indicated that exemplars reduce critical thinking, are more

likely to be accessed throughout the decision making process, but are not as concrete or as

accurate as base-rate information.

The representativeness heuristic is the third mechanism activated under

exemplification theory and is substantially different from priming and the availability

heuristic. Under this heuristic, judgments about situations are based on the perceptions an

individual holds regarding the probability or frequency in which an event may occur. It

also focuses on individual's “assessing the likelihood that a certain object or person

belongs to a certain class of objects or persons” (Zillmann & Brosius, 2000, p. 41).

Tversky and Kahneman (1974) found that people who use a representativeness heuristic

believe quantitatively and through the use of probabilities that causes and effects will be

equal. If a statistic stated one out of every two people will divorce, someone might

reasonably expect that half of all marriages end in divorce. However, that assumption may

not be correct as some people have been married multiple times, distorting the actual

percentage.

While people may hold tight to the assumption they are correct in their judgment,

studies have shown most people do not make decisions based on probabilities or

percentages (Tversky & Kahneman, 1974; Zillmann & Brosius, 2000). In a few incidences

in which anecdotes or stories are contradictory of one another, statistics and probabilities

have been proven to have significant effects (Zillmann, 1999). According to Bates (2010),

individuals rarely process information in a systematic manner focused on data collection,


41

assessment, and evaluation. Bates goes on to state that to better understand

exemplification theory, individuals often search out the simplest process through the use

of representative and availability heuristics, allowing them “to gather, sample, select, and

integrate information more quickly but not necessarily thoroughly” (p. 20). Due in part to

the influential nature of exemplars over base-rate information, Zillmann and Brosius

(2000) laid out how exemplars should be representative of the population with regards to

the phenomena being discussed.

Exemplification of Known Event Populations

Exemplification is based on the idea that events can be grouped in such a way that

each event shares similar attributes to other events within the narrative and would

therefore exemplify group attributes (Zillmann & Brosius, 2000). Each group then is

representative of the overall population being discussed. Because of this, exemplars must

be characteristic and typical of the group, or they will be deemed inappropriate because

they fail to provide reliable information regarding the group (p. 5). For instance, if most

commercial airlines struck by lightning were never affected or suffered mechanical

damage, then a story about airline safety that only focused on the one or two airline

disasters that were caused by lightning might be deemed inappropriate. The information

potentially creates a cause and effect scenario in which someone watching or reading the

story might not realize that the event, an airplane brought down by lightning, is actually

atypical or uncharacteristic of the actual group and reality. The individual may believe

that all planes struck by lightning have a high probability of crashing and therefore may

feel anxiety when flying through bad weather.


42

In order to counter an exemplar that is not typical of an event or group, research

(Zillmann & Brosius, 2000) has shown the use of multiple exemplars can reduce

ambiguity and misinformation. However, greater precision is not always assured through

the use of multiple exemplars. According to the authors, arbitrary selection of exemplars

potentially leads to duplication which in turn can increase misrepresentation. In the

example of lightning strikes on an airplane, if four exemplars were arbitrarily drawn and

three focused on planes crashing because of lightning, then once again a reader may think

there is a seventy-five percent chance of being harmed by lightning. In this case the reality

of the situation does not reflect the presentation of the story. Therefore, during the

selection process Zillmann and Brosius stated that a blind sample of exemplars has a

greater capacity to ensure accuracy but still remains limited in its overall

representativeness. They termed this concept selective exemplification.

Another sampling technique used to increase the precision of representativeness

among groups is known as stratified or proportional exemplification (Zillmann & Brosius,

2000). Stratified exemplification takes an equal number of exemplars from each

subpopulation in order to have an equal representation for the sample of events. An issue

that arises with this method is that one subpopulation may be much larger than the others.

If this is the case, then misperceptions may still occur. Once again in the airline example,

if the story contains two exemplars—one airplane was not harmed after being struck by

lightning but another plane crashed after a lightening strike--then an individual may

believe airplanes have a fifty-fifty survival or crash rate based upon the information. Here

stratification has been partially successful in creating precision, but once again

misperceptions may occur. The last sampling method, proportional exemplification, tends


43

to be the most accurate measure when constructing a sample of events. Under this model

exemplars are divided into subpopulations according to their group size within a

population of events. Larger subpopulations have more samples drawn than smaller

subpopulations in order to ensure greater accuracy. In this method, if less than ten percent

of lightning strikes cause airline accidents, then less than ten percent of the overall sample

will include exemplars of this nature. At this point an individual should have a better

understanding of the actual event and hopefully realize that lightning strikes on airplanes

pose very little risk.

Exemplification of Unknown Event Populations

In situations where the events of a population are unknown or there is very little

information regarding the population, then determining the representativeness of

exemplification is difficult. Because a population is not easily ascertained, it would be

arrogant to presume that audiences do not make general assumptions regarding the event.

In fact individuals are still just as likely to infer that the limited characteristics of that

event are representative of the larger whole (Higgins, 1996). Those few exemplars, even if

uncharacteristic, are viewed as typical of the group.

To diminish misconceptions, Zillmann and Brosius (2000) stated that information

conveyers should alert audiences that the exemplars are only unique to a special, singular

case; and in situations where the extent of the exemplars' prevalence is not known, the

conveyor needs to indicate the lack of knowledge surrounding the exemplar or case.

Offering qualifying information may reduce misinterpretations; however, there is a lack of

research examining these types of messages.

Personal Characteristics


44

With a strong focus on testing base-rate information against exemplars,

exemplification has been surprisingly deficient in determining whether people understand

base-rate information. Lipkus, Samsa, and Rimer (2007) stated that even adults who

possessed advanced educations struggled to answer rudimentary ratio transformation

questions. If a large portion of the population lacks the general ability to process

numerical information, then the rate of misinterpretation and lack of understanding had to

be substantially high (Reyna & Brainerd, 2008). Logically, individuals who fall into a low

numeric ability category are more likely to be persuaded by stories lacking quantitative

analyses and that provide nonnumeric information (Zillmann, Callison, & Gibson, 2009).

Through the creation and testing of an arithmetic aptitude exam, the authors found

evidence that those who scored higher on the test had higher recall than those with low

scores.

From the Zillmann, Callison, and Gibson (2009) study, the concept that there may

exist a quantification heuristic was presented (Zillmann, 1999, 2002). This concept

assumes that people are able to reasonably keep track of “their memory for exemplars of

phenomena of interest and attain from that a sense of quantity and proportion” (Zillmann,

Callison, & Gibson, 2009, p. 412). This idea posited that those who possess low numeric

abilities are more likely to avoid quantitative information while seeking out descriptive

exemplars. Emotional and other distracting forms of exemplars are also more likely to

cause a person with low numeric abilities to deviate away from quantitative information

and toward more descriptive explanations. Reinforcing this finding, Reyna, Nelson, Han,

and Dieckmann (2009) found individuals who possessed high numeric ability were more

likely to accurately interpret quantitative information, leading the authors to state that


45

individuals possessing lower numeric ability were more susceptible to external or

extraneous factors (p. 943).

In a follow up study, Callison, Gibson, and Zillmann (2009) tested how people’s

ability to process numeric information would affect their recall abilities of the quantitative

information. Their prediction that recall of number frequencies would be greater than that

of ratios was not supported. Their second hypothesis stating the recall of percentages

would be greater than ratio expressions was only partially supported. Once again, those

possessing low numeric abilities struggled in processing ratios while those with

intermediate and high numeric abilities had little difficulty. The verbalizing of ratios by

spelling out the numbers, i.e. three out of nine doctors, was strongly supported as an

inferior method and was largely seen as “detrimental to processing and recall” (p. 53).

In the most recent study of arithmetic aptitude, Gibson, Callison, and Zillmann

(2011) investigated how individuals with low numeric abilities were more receptive to

health care exemplars that provided personalized information than reports focusing on

impersonal quantitative specifications. Similar to the results of other arithmetic aptitude

studies (Callison, Gibson, and Zillmann, 2009; Zillmann, Callison, & Gibson, 2009),

people with low numeric abilities are more likely to be drawn toward relevant exemplars

than those who possess higher numeric abilities. This study, if examined from an applied

approach, demonstrates how the successful use of exemplars and base-rate information is

critical to accurately communicating messages to an audience.

Research Questions and Hypotheses

This study seeks to determine if the media’s reporting on economic figures and

data affects individual’s perceptions of the economic conditions and whether individuals


46

are able to accurately recall base-rate information and exemplars in the context of an

economic story. Price and Tewksbury (1997) stated, “when the media distribute messages

of highly consistent design, then chronic accessibility of constructs relating to those

attributes and features might be expected" (p. 199). Therefore, the more consistent the

message is, the more likely people will remember and access that information later.

However, individuals who focus more on exemplars within a story are also much less

likely to recall substantial information from the story or message (Brosius, 2003;

Kazoleas,1993). Studies have shown the use of exemplars is more effective over base-rate

or statistical information when influencing attitudes and perceptions (Brosius & Bathelt,

1994; Limon & Kazoleas, 2004; Martin & Powers, 1980; Taylor & Thompson, 1982).

One reason is the ability of exemplars to provide more vivid images and narratives within

a story (Nisbett & Ross, 1979, 1980) while another focuses on individual’s reliance of

exemplars to provide mental and cognitive heuristics (Kahneman & Tversky, 1972).

Zillmann and Brosius (2000) stated that an overabundance or oversample of a

subpopulation that does not accurately represent or is typical of the sample of events may

cause misconceptions and misinterpretation of the event itself. Research has shown the

use of multiple exemplars can reduce ambiguity and misinformation in order to counter an

exemplar that is not typical of an event or group (Zillmann & Brosius, 2000). Therefore,

based on the exemplification literature the following hypotheses were offered:

H1a: Base-rate positive information, as well as positive exemplars, is more

likely to lead to perceptions that the overall U.S. economy is currently

strong/healthy or is in a strong recovery than will negative base-rate

information or negative exemplars.


47

H1b: Representative exemplars will lead to greater accuracy in recall of base-

rate information than will non-supportive or mixed exemplars.

Logically, individuals with higher economic aptitude or literacy should be able to

correctly understand and provide answers that are more reflective of actual economic

conditions, i.e. numerical base-rate figures, as opposed to mediated exemplars that may

only tell one side of the narrative or issue (Walstad & Rebeck, 2002). Individuals with

high economic aptitude should derive their information and perceptions from the base-rate

data provided in the story or at least correctly identify those statistics and information later

(Walstad & Rebeck, 2002); whereas, those low in economic aptitude will most likely

focus on exemplars to better understand the economy (Blendon et al., 1997; FED, 1998;

Harris & Associates, 1999).

Higher economic aptitude individuals may also be more likely to answer questions

that more accurately reflect real-world indicators of the economy and/or direction of the

economy than those who score lower due to their more innate understanding of how the

economy works (Blendon et al., 1997). Part of this centers on the hypothesis that

individuals possessing higher economic literacy have the ability to separate base-rate

information they view in the news while placing less emphasis on narrative exemplars.

Those with lower aptitude scores are more likely to focus on exemplars to provide cues,

creating mental heuristics that affect the way they answer questions on the Index (Caplan,

2002; Roos, 2007).

H2a: The effect of H1 is mediated by economic knowledge such that people with

high economic literacy will have perceptions that more accurately reflect

the base-rate information showing the overall U.S. economy is currently


48

getting stronger (recovering) or weaker (remaining stagnate or sliding

back), while those with low economic literacy will tend to be overly

influenced by positive and negative exemplars.

H2b: The effect of H1 is mediated by economic knowledge such that people with

high economic literacy will have better recall of factual data regardless of

the representativeness of exemplars, while those with low economic

literacy will tend to be overly influenced by exemplars.

Recent empirical findings into exemplification have explored how individuals'

arithmetic aptitude affects their ability to understand and accurately recall base-rate

information (Gibson, Callison, & Zillmann, 2011). According to the authors, “it was

concluded that variation in arithmetic ability is associated with distinct differences in

information processing…” (Callison, Gibson, & Zillmann, 2011, p. 6). While these

findings are relevant in understanding how people process numeric and quantitative

information, there is little research into content specific levels of aptitude such as

economics. Blendon et al., (1997) stated the population is significantly deficient in their

knowledge of the economy, hampering their understanding of major economic issues.

This presents an issue in determining whether arithmetic ability or economic literacy play

a more dominant role in influencing an individual’s ability to accurately recall economic

base-rate information correctly.

R1: Under varying conditions, does either arithmetic aptitude or economic

literacy have a more dominant influence over an individual’s evaluations as

to the health/state of the economy or ability to accurately recall economic

base-rate information?


49

As Walstad and Rebeck (2002) found, demographics do play a substantial role in

peoples’ perceptions of the economy. Age, ethnicity, gender, and level of education were

all significant factors in people’s economic literacy which in turn affected their economic

outlook. However, one interesting factor that produced mixed results was political

affiliation. As discussed earlier, partisanship does have a positive impact on perceptions of

the economy when an individual identifies with the party or official in office (Gerber &

Huber, 2010). The results, however, are inconclusive as to whether peoples' economic

literacy and political affiliation have any significant effects on how they perceive or

process economic information.

R2: Under varying conditions, what role does political ideology have over an

individual’s perception of the overall strength of the economy?


50

CHAPTER III

METHODOLOGY

Overview

The research within this paper focused specifically on answering three questions:

1). Are individuals affected by base-rate information or the exemplars within economic

news stories, 2). What effect does this have on how individuals perceive the state of the

economy and their ability to accurately recall information, and 3). Do personal

characteristics determine how audiences are influenced by base-rate information versus

the use of exemplars in news stories? The study implemented an experiment to determine

how individuals were affected by economic base-rate information and exemplars within

the media. The objective of this inquiry is to try to establish if the way the media reports

economic news affects how individuals perceive the strength of the general state of the

economy, as well as whether the economic aptitude of individuals affects their

perceptions. The overall goal of the study will potentially provide evidence illustrating the

need for the media to be more careful and meticulous in their presentation of economic

news and that individuals should be more knowledgeable regarding how the economy

works.

Research Participants

An online sampling firm, The Sample Network, asked a panel of 5,553 participants

from its network to take part in this experiment. They distributed the link to the study’s

website through email at which time 721 respondents within the panel began the study

with 550 completing it. The study had an overall response rate of 10 percent. At that point

66 participants were removed for a number of reasons. The first factor that automatically


51

disqualified participants was if they skipped any question throughout the study. Although

they may have completed the study, they were cut entirely from the dataset if they left

blank answers. Part of the rationale for this is based upon the robust number of

participants within the research. Two of the measures focused on individuals’ aptitude and

while logically a missed question might be the same as an incorrect answer, it would

might also affect the individual’s placement within high and low levels of economic and

arithmetic literacy. Because the study exceeded the number of participants needed, it was

better to eliminate those who did not answer every question then to systematically use

case-wise or list-wise deletion and allow them to remain.

The second disqualifier was based upon the length of time it took each participant

to complete the study. The Qualtrics system provided the exact time each participant

began and ended the experiment. Based upon both those numbers, Qualtrics estimated the

average completion time was 24 minutes. In order to potentially eliminate respondents

who hastily speed through the questionnaire, a time-stamp was provided to determine the

length of time each respondent took. In determining how to eliminate from the sample

those who took too long and those who did not spend enough time on the study the

Qualtrics system indicated that examining the outliers was the best tool. The results

showed almost 95 percent of the outliers within the study took less than 16 minutes

reading and answering the questions or spent more than an hour on it. While this does not

guarantee successfully eliminating all those who rush through without providing

thoughtful answers, it does help to identify cases where that incident may have occurred.

From this, the final overall sample consisted of approximately 484 participants.


52

Only participants 18 and older were allowed to participate with the largest age

group falling between 55 and 64 years of age (33.9%), closely followed by those 35 to 54

years of age (33.7%). Males made up 38 percent of the sample while 62 percent were

female. Fifty-five percent of the participants had a high school education to some college

while 45 percent had at least a two-year college degree or more.

When asked about political affiliation and ideology 38 percent of respondents

stated they affiliated more with the Democratic party, 34 percent affiliated with the

Republican party, 21 percent stated they were independents, only two percent felt they

were more aligned with the Libertarian party, while the remaining five percent said other

or did not wish to say. Among the participants 19 percent identified themselves as liberal

to very liberal; 55 percent identified themselves as left of center, moderate, or right of

center; and 26 percent identified themselves as conservative to very conservative.

The majority of respondents (87.4%) identified themselves as “white/Caucasian,”

followed by “African-American” (7.2%), and “Hispanic” (3.1%). California and Florida

were tied for the top state from which participants were drawn at 8.3 percent. Next was

New York at 8.1 percent and Texas at 6.8 percent, with Pennsylvania rounding out the top

five at 5.6 percent.

Experimental Procedures for Testing Economic Communications

Although aspects of exemplification are rooted in the idea that recently

encountered conditions or events are aggregated and accessible in one’s memory, there are

both advantages and limitations when using experiments. In defense of experimental

methods, Morton and Williams (2008) stated that it is understood that there is “no perfect

or true experiment” (p. 5), but that choosing a particular methodology should be


53

determined by the research questions being asked. Addressing concerns of validity, the

authors posit that experimentalists are criticized for manipulating variables and controlling

the environment, and yet on the other side, observational researchers, in deciding what

measures to observe and study, manipulate the measurement choices. Essentially, the

simple decision of choosing and ignoring different factors is a control technique used by

all scientists. Other researchers (Beck, 2000; Campbell, 1957; Kinder & Palfry, 1993;

McGraw & Hoekstra, 1994) have rigorously defended the use of experiments observing

that external validity is “not about whether the experiment resembles the hypothesized”

(Morton & Williams, 2008, p. 11) population, but “is really about the robustness of these

experiments across different formulations” (p. 11).

This experiment employed a 2 (high economic aptitude x low economic aptitude)

x 2 (base-rate directionally positive x base-rate directionally negative) x 3 (exemplar

representative x exemplar non-representative x exemplar mixed) factorial design. In

addition, a 14-question economic aptitude test was provided, the 10-question quantitative

literacy exam, a five-item questionnaire regarding the accuracy in recall of the news story,

and 12 questions regarding overall perceptions of the economy, along with

seven basic demographic questions. The within-subjects design portion of the experiment

encompassed six different variations of how the material was circulated. Each version

incorporated one of the following distributions:


54

Figure 3.1: Within-Subjects Design

In sum, there were 36 different versions of the experiment requiring at least 360

participants to adequately power the study (Ellis, 2010). Participants in the experiment

always read one of the stimulus articles at the beginning of the study. After each

participant read the stimulus, a random rotation of block 2 that included economic

assessment, arithmetic ability, and economic perceptions was incorporated within the

questionnaire, controlling for order effects. Block 3, stimulus recall, always appeared in

the same order following block 2. The reason stimulus recall was not part of the

randomization of block 2 was to ensure that everyone answered those questions relatively

within the same time frame. If one group of participants were able to take it earlier than

another group, it would place the second group at a comparative disadvantage, adding a

contaminant to the experiment. Demographic questions always concluded the study.

Each base-rate story included three exemplar related manipulations. The first,

exemplar-representative, included three exemplars that accurately reflected either the

positive or negative base-rate information, with one exemplar not supporting the base-rate.

Therefore, the exemplars “exemplify” or provide support for an accurate account of what

Within-Subjects Design

Block 1 Block 2 Block 3 Block 4

Stimulus Economic Assessment

Arithmetic Ability

Economic Perceptions

Stim Recall Demographics

Stimulus Economic Assessment


Arithmetic Ability


Stimulus Arithmetic Ability


Economic Assessment


Stimulus Arithmetic Ability

Economic Assessment



Stimulus Economic Perceptions

Economic Assessment

Arithmetic Ability


Stimulus Economic Perceptions

Arithmetic Ability

Economic Assessment



55

is being reported. The second manipulation, exemplar non-representative used the same

base-rate information but consisted of three exemplars that contrasted and did not reflect

the numerical economic information, with one exemplar supporting the base-rate. In

effect, the exemplars create a dynamic in which it appeared either the base-rate

information or the exemplars were incorrect. The last manipulation, exemplar mixed, is

comprised of two representative and two non-representative exemplars. Including this

manipulation helped determine what effects might occur regarding overall perceptions of

the story and base-rate when exemplars contradicted one another.

Experimental Design (2x2x3)

High Assessment

Base-Rate Positive Base-Rate Negative Exemplar Representative Exemplar Non-representative Exemplar Mixed

Low Assessment

Base-Rate Positive Base-Rate Negative

Exemplar Representative Exemplar Non-representative Exemplar Mixed

Figure 3.2: Experimental Design

Procedure

While the use of a laboratory would offer greater control over participants and the

environment, a field or Internet experiment offers greater potential for sampling larger,

more diverse audiences. Because this study is interested in looking at how the average

American understands and interprets economic information, an online-based experiment

provides more robustness than a laboratory experiment. Even as ethical criticisms exist

stating that participants in field experiments are not aware they are being manipulated


56

(Wantchekon, 2003), survey experiments in political science have become more popular

and accepted among social scientists (Sniderman, Brody, & Tetlock, 1991; Sullivan,

Pierson, & Marcus, 1978).

While there are a variety of limitations in using online survey experiments, one

revolves around sampling issues of the population. Some (Dillman, 2000; Stanton, 1998)

have stated that little is known about the characteristics of a sample and generally cannot

be verified before or after a study begins. Much of the demographic information relies

heavily on self-reporting from the participants. Others (Andrews, Nonnecke, & Preece,

2003; Couper, 2000) say a true random sample cannot be generated because not all

eligible participants within a population have an active email address. One group of

researchers (Konstan, Rosser, Ross, Stanton, & Edwards, 2005) found that while this is a

legitimate concern, response rates from online survey experiments is generally as good, if

not better, than traditional mail surveys (Mehta & Sivadas, 1995; Stanton, 1998;

Thompson, Surface, Martin, Sanders, 2003). Self-selection is often another problem in the

use of online surveys (Stanton, 1998; Witmer, Colman, & Katzman, 1999). However, for

research that is less concerned with using probability samples and more focused on

establishing causation, then self-selection becomes less of an issue (Wright, 2005). As

stated earlier, this study is not concerned with looking at special populations or

demographics within the United States. It is designed to examine how the general

population at large processes economic information. Therefore, sampling issues that may

complicate or limit online survey experiments are less likely to pose a problem within the

findings of this study.


57

Using the sampling survey company, The Sample Network, participants were

randomly assigned one of the twelve versions of the experiment to complete and

compensated for their responses. After a participant received the email, he or she was

asked to read a news story regarding a municipality’s local economy. Following the story,

participants were asked to answer the 14-question economic assessment quiz, the 10-item

arithmetic aptitude test, and the six-item measure of economic perceptions. The order was

rotated depending on the version, but all participants were asked to complete each section.

Following the assessment and index, all participants were asked five questions regarding

the story in order to measure recall as well as demographic questions about themselves.

Participants were not told in advance they were taking part in an experiment, only that

they were helping researchers understand the relationship of the media and economy.

Upon completion, they were provided the contact information of the primary

investigator in order to learn more about the project. The questionnaire was not be timed,

but participants were not allowed to go back and reread any previous information. Their

answers were based on recall and perceptions of the story, so allowing respondents to go

back and review the material would defeat the purpose of the study.

Stimulus Material

The news stories were designed and written to look as if each were taken directly

from a daily newspaper. The conceptualization of exemplars for each story are based upon

studies (Burns, 1992; Mervis & Rosch, 1981; Zillmann, 1999) that have measured

exemplars as direct encounters or personal experience to an issue, event, or happening.

This definition is also in line with research (Horner, 2008; Lang & Lang, 1981; Zucker,

1978) stating that obtrusive issues such as the economy are highly salient in people’s


58

minds when those issues are derived from personal experience, observations, or events in

which a given population is affected by the issue. Therefore, exemplars used must reflect

how the issue of the economy has directly affected the stated individual’s life.

In the first version, the base-rate information was built upon the positive

favorability regarding a recent Consumer Confidence survey. For the first story, exemplar

representative, three exemplars in the form of direct quotes and paraphrased statements

were used to reflect positive viewpoints mirroring the base-rate information. Any of the

following exemplars could be positive or negative depending on the version. One

exemplar was from the standpoint of a local business owner, the second was from a recent

college graduate, next was a statement provided from a local commercial contractor, and

finally the fourth was a quote from a bank president regarding personal and commercial

finance. Three of the four exemplars in version one positively supported the base-rate

information while one did not (see Appendix A, p. 130).

In the second story, the base-rate information remained the same, but all three

exemplars were non-representative in a negative direction in their reporting. The three

quotes or statements used the exact same sources as before; however, the reporting of

exemplars in this story runs completely contrast to the base-rate information provided.

The type of information provided in each statement given was roughly identical to the

aforementioned story. The only difference was the information was no longer reflective of

the numbers or statistics provided. For example, in the exemplar representative story a

leader in the home-building industry might say growth in their industry has forced his

business to expand and to increase hiring, in essence supporting the base-rate. However,

in the non-supportive story that same business leader would state their industry is still


59

suffering from the poor economic environment, and he’s had to reduce the number of

employees on his payroll (see Appendix A, p. 138).

The third story was a mix of exemplars using both positive and negative

statements from the first two. The positive base-rate information remained the same;

however, now two of the statements were representative and two were non-representative.

This scenario seeks to achieve a balanced approach among exemplars, as well as helping

to determine if stories that provide two dissimilar sides are significantly different from

those that only offer one. It helps to determine if individuals when presented with

conflicting exemplars will consciously or subconsciously return to the original base-rate

information when asked about their perceptions of the economy (see Appendix A, p. 146).

The second version reversed the original economic climate in which the economic

indicator of unemployment is negatively affecting the economy. In the first story (fourth

overall), the focus is once again on exemplar-representative attributes. The base-rate

information is reported in a negative direction where three of the four exemplars reflected

the direction of those numbers and statistics. The exemplars used for this story were

completely identical to the exemplars used in version one-story two of the study. As those

statements originally went against the base-rate, they are now reflective of the base-rate

(see Appendix A, p. 154).

Overall, the fifth story is much like the fourth one. Whereas, the base-rate

information has now changed in a negative direction, the positive exemplars from the first

story were now non-representative in this one. The three exemplars were non-

representative in a positive direction in their reporting. While the exemplar narratives are


60

positive the strength of the economy, it was opposite of the base-rate making it non-

representative (see Appendix A, p. 162).

The sixth story was almost completely identical to the third story with the

exception of having the base-rate information changed. The quotes and statements are

identical, and they reflect exemplars in both a positive and negative direction of the base-

rate information, making the entire premise balanced (see Appendix A, p. 170).

Figure 3.3: Story Manipulation

Each story had four exemplars embedded following the initial base-rate figure.

The four exemplars categorized as 1) small business, 2) employment, 3) home sales, and

4) personal finance were either positive or negative, representative or non-representative

depending on the version. However, in order to avoid the possibility that any one of those

exemplars could affect someone’s perception of the economy, the positive (or negative)

exemplar was rotated throughout each story version. Therefore, each story manipulation

actually had four different versions in order to control for this potential confounding

variable. Once the rotation was completed there were 24 different versions of the stories.

75 percent optimistic on

Consumer Confidence Index

75 percent pessimistic on

Consumer Confidence Index

Representative

Positive

Non-

Representative

Negative

Mixed

Representative

Negative

Non-

Representative

Positive

Mixed

1 + - + 1 - + +

2 + - - 2 - + -

3 + - + 3 - + +

4 - + - 4 + - -


61

Measures

Story Version and Manipulation

The six manipulations of the economic news stories had an almost even

distribution of participants. Both the “Positive base-rate: Representative exemplars” and

“Positive base-rate: Mixed exemplars” stories had 79 participants or 16.3 percent of the

total sample. “Positive base-rate: Non-representative exemplars” and “Negative base-rate:

Non-representative exemplars” stories had 80 participants or 16.5 percent of the total

sample. “Negative base-rate: Representative exemplars” and “Negative base-rate: Mixed

exemplars” stories followed with 83 participants each or 17.1 percent.

These manipulations were then broken down into three separate categories in order

to better determine what role, if any, the base-rate and exemplars had about economic

perceptions. The first group, base-rate, incorporated all directionally positive stories into

one level (positive base-rate: representative: non-representative: mixed) and all

directionally negative stories (negative base-rate: representative: non-representative:

mixed) into another. The positive level had an n = 238 (49.2%) while the negative level

had an n = 246 (50.8%). See figure 3.4.

Base-rate Positive

Base-rate Negative 75 percent optimistic 75 percent pessimistic

n = 238 n = 246

Representative Positive

Non-Representative

Negative Mixed

Representative Negative

Non- Representative

Positive Mixed

Figure 3.4: Base-rate Manipulation

The second group, exemplars, was constructed by taking the stories and dividing

them into categories based upon the positive, negative, or mixed exemplars used. This

category incorporated all positive exemplars into a level (positive base-rate: representative


62

exemplars & negative base-rate: non-representative exemplars), all negative exemplars

(positive base-rate: non-representative exemplars & negative base-rate: representative

exemplars), and finally mixed exemplars (positive base-rate: mixed exemplars & negative

base-rate: mixed exemplars). The positive level had an n = 159 (32.9%); the negative level

had an n = 163 (33.7%); the mixed level had an n = 162 (33.5%). See figure 3.5.

Positive Exemplars

Negative Exemplars

Mixed Exemplars n = 159

n = 163

n = 162

Positive base-rate: Representative

exemplars

Negative base-rate: Non-

representative exemplars

Positive base-rate: Non-


Negative base-rate:

Representative exemplars

Positive base-rate:

Mixed exemplars

Negative base-rate:

Mixed exemplars

Figure 3.5: Exemplar Manipulation

The final category was based upon the representativeness of the exemplars in

regards to the base-rate, essentially, how reflective or non-reflective the exemplars were to

the base-rate information. Similar to the exemplars group, this category was broken down

into three levels: representative (positive base-rate: representative exemplars & negative

base-rate: representative exemplars), non-representative (positive base-rate: non-

representative & negative base-rate: non-representative), and mixed (positive base-rate:

mixed exemplars & negative base-rate: mixed exemplars). The representative level had an

n = 162 (33.5%), the negative level had an n = 160 (33.1%), and the mixed level had an n

= 162 (33.5%). See figure 3.6.

Representative Exemplars

Non-representative Exemplars

Mixed Exemplars

n = 162

n = 160

n = 162

Positive base-rate:


Negative base-rate:


Positive base-rate: Non-


Negative base-rate:

Non-representative

exemplars

Positive base-rate:

Mixed exemplars

Negative base-rate:

Mixed exemplars

Figure 3.6: Representative Manipulation


63

Economic Literacy

After reading each story, participants were given three sets of questions. The first

set measured a participant’s economic assessment and literacy. The Test of Economic

Literacy (TEL) is widely regarded as one of the most thoroughly examined and accurate

measures for economic literacy in the United States (Soper-Walstad, 1987; Walstad &

Rebeck, 2002; Walstad & Soper, 1988). The exam is a standardized test focused on four

economic concept clusters – fundamental, microeconomic, macroeconomic, and

international - that provides a comprehensive assessment of economic literacy (Walstad &

Soper, 1988, p. 251).

In 1999, the National Council on Economic Education (NCEE) commissioned

Harris Interactive to evaluate adult understanding of basic economics, and in 2005 the

NCEE asked Harris Interactive to conduct a follow up study (Harris Interactive, 2005).

The 2005 online study consisted of 3,512 U.S. adults and took on average 14 minutes to

complete. The average grade was a 70 while one-third of adults scored above average with

an A or B. Participants were given a 24-question quiz covering basic economics and

personal finance questions. The following is a breakdown of the TEL: five questions

regarding economics and the consumer; five questions focusing on factors pertaining to

production; four questions over money, interest rates, and inflation; six questions covering

government and trade in economics, with the final four questions examining personal

finance. According to the NCEE (2005),

The content standards covered by the quiz reflect the essential principles of

economics. The standards encompass concepts such as scarcity, allocation of


64

goods and services, role of competition, role of money, and specialization and

trade. In a global economy, understanding trade has particular resonance (p. 6).

The initial NCEE (2005) findings were reaffirmed by Tabesh & Schultz (2007),

indicating that gender plays a significant role in economic literacy. The NCEE study

(2005) in turn reaffirmed the findings of Walstad and Rebeck (2002) that showed age,

gender, ethnicity, and level of education were all significant factors in basic economic

understanding. Unfortunately, the NCEE study did not ask questions about participants’

political affiliation. Once again this variable and its role in the area of economic literacy

and economic perceptions remains unresolved (Walstad & Rebeck, 2002).

Five of the original questions within the NCEE study were eliminated as the

subject matter concerning international economic issues might introduce confounding

variables into the experiment. Because the focus of the experiment was strictly on the

overall U.S. economy, any variable that potentially introduced outside or international

economic factors might result in skewed results. In addition, five of the original questions

within the NCEE study were removed as results showed those questions were correctly

answered by more than 90 percent of the respondents, with of three of those questions

correctly answered by more than 95 percent. These were excluded to further prevent

survey/respondent fatigue as research (Krosnick, 1999) has shown excessively long and

complicated surveys lead to a deterioration of the quality of data.

Economic literacy was based upon a 14-item multiple choice test in which

participants’ answers were recoded from their original choice into a new variable where a

one meant the answer was correct or a zero for incorrect. These scores were then summed

together in a new variable, with a third variable reporting their overall average. Economic


65

literacy was almost evenly split with roughly half of the participants scoring a seven or

higher (M = 7.25, SD = 2.70). It was then divided into categories of high or low based

upon a person’s score. When recoding into a new variable, mean scores that were seven or

fewer were categorized into low economic literacy while participants with scores greater

than seven were considered high in economic literacy. An ANOVA was run to determine

if a difference existed between high and low economic literacy in order to validate

transforming participants’ scores into a categorical variable. The test indicated the two

levels were significantly different from each other, F1, 487 = 972.75, p < .01, ηp2 = .67.

Those in the low group represented 51.7 percent of the overall sample.

Arithmetic Aptitude

The arithmetic aptitude test was developed by Zillmann et al. (2009) in order to

measure how individuals' arithmetic ability influences their ability to process media

messages. According to the authors, the test “is the only one scale - sensitive to all aspects

of arithmetic competence, including the ability to convert frequencies to percentages and

ratios” (Callison, Gibson, & Zillmann, 2011, p. 5). It consists of 10 numerically-based

questions using a mix of percentages, ratios, currency, and narratives i.e. “less than a

quarter”. The test has been shown to affect individuals’ assessments of risk and retrieval

of statistical information based upon their placement of arithmetic ability (high or low).

Arithmetic aptitude followed the same structure and calculation that economic

literacy did. Arithmetic aptitude was based upon the 10-item multiple choice test in which

participants’ answers were recoded from their original choice into a new variable where a

one meant the answer was correct or a zero for incorrect. These scores were then summed

together into a new variable. Participants scored slightly higher in arithmetic aptitude than


66

they did in economic literacy. More than half of the participants scored a five or higher (M

= 5.80, SD = 2.45). Like economic literacy, arithmetic aptitude was then divided into

categories of high or low based upon a person’s average score. When recoding into a new

variable, mean scores that were six or fewer were categorized into low arithmetic aptitude

while participants with scores greater than six were considered high in arithmetic aptitude.

Once again an ANOVA was run on high and low arithmetic aptitude to determine if

differences existed once it was transformed into a categorical variable. The results from

the test showed the two levels were significantly different, F1, 487 = 1025.35, p < .01, ηp2 =

.68. Those in the low group represented 54.5 percent of the overall sample.

Political Ideology

Using the single-item measure political affiliation, i.e. Democrat, Republican, or

Independent, as an independent variable does not provide a reliable estimate of where

individuals truly lie along the political spectrum, nor does it offer any potential variance as

to where they feel they are on both broad and specific issues. Research (Jost, Federico, &

Napier, 2009) suggests party affiliation reflects political heuristics and short cuts that do

not truly reflect a person’s political views. A better measure for this study was to ask

participants where they felt they belonged along the political ideology spectrum.

According to Jost, Federico, and Napier (2009) “ideology predicts citizens’ general value

orientations” (p. 324) and is “among the strongest and most consistent predictors of

political preferences” (p. 324). Three seven-point measures were asked regarding their

political beliefs.

The first question focused on their general political ideology where a one was very

liberal, a two was liberal, a three was left-of-center, four was moderate, a five was right-


67

of-center, a six was conservative, and a seven was very conservative. On average, people

indicated they were fairly moderate (M = 4.19, SD = 1.62). Using the same scale for fiscal

and social ideology, most respondents indicated they were fiscally moderate (M = 4.34,

SD = 1.54) as well as socially moderate (M = 4.08, SD = 1.68). This study however only

examined ideology using the single-item measure of overall political ideology.


The Consumer Sentiment Index is considered to be one of the top consumer

economic indicators as it reliably gauges the personal experience of an individual with

their overall perception of the economy (Ladner & Wlezien, 2007). The authors found the

Index to be an accurate and relatively predictive tool because of its questions regarding

consumer opinions on unemployment, inflation, perceptions of the overall U.S. economy,

and the prices of goods and services, such as gas. One of the more important variables

they point out is how the index looks at both present conditions and future perceptions

twelve months out. While this study is not trying to determine consumer confidence, it is

attempting to ascertain participants’ current and future perceptions of the economy. The

Index uses a mix of Likert-type scale questions and “best-guess” percentage questions.

For this study only a few of the questions from the Consumer Sentiment Index will be

used in order to measure economic perceptions. See Appendix B, p. 178 for question

wording and response options.

In order to measure the economic perceptions of each individual, six questions

with Likert-type scale answer choices were asked regarding different aspects of the

economy. Each questions asked the individual to rank from one to seven a person’s

opinion of the economy. A one was considered to be bad or worse off where a seven was


68

considered to be excellent or better off. The highest a person could rank the economy was

a 42, with the lowest a seven. After calculating the overall average of the six questions,

the respondents indicated the economy was roughly unchanged or neutral (M = 3.30, SD =

1.15). Using the same scale measures, respondents were asked how confident they were in

their given answer. A one represented “not very confident” where a seven meant they

were “very confident.” While economic perceptions were slightly less than positive and

more in neutral territory, two-thirds (67 percent) of individuals when asked about the

certainty of their answer felt confident in their perception regarding the current state of the

economy (M = 4.71, SD = 1.34).

Base-rate Recall

Participants were asked to remember and then indicate along a sliding scale what

percent of the public in the article had positive perceptions of the economy and what

percent had negative perceptions. The rationale behind asking all participants both of

these questions was based upon the actual base-rate number. Half of the respondents were

in conditions where 75 percent of the population had positive perceptions; whereas, the

other half were in conditions where 75 percent of the population had negative perceptions.

Just asking what percent had positive perceptions would have created a host of problems,

introducing at least one confounding variable. In the positive situation half of the

respondents would only have to remember the actual base-rate number; whereas, the other

half would need to mentally convert the percentage and then flip the positive 75 percent

into a negative 25 percent. And while this could potentially be controlled by asking the

contrasting side of the question, the ultimate goal of this study is not to see if people are


69

able to accurately convert numbers but to see if they can accurately recall the base-rate

information.

The accuracy of recall was based upon two questions that were combined to form

a single-item measure. In order to create the singe-item measure, both questions were

asked, but only one question was used depending on the story version a participant was in.

The rationale behind this is that both scores could offset one another if taken as an

average. For example, if a participant in the positive base-rate manipulation (75%) stated

that 75 percent of the people were optimistic and then when asked what percent was

pessimistic and gave an answer of 25 percent then the average would be 50 percent,

indicating they were moderately accurate instead of 100 percent accurate. On the flipside

by asking the question to all participants we would create another troublesome issue.

Those in the positive base-rate manipulation when asked about optimism would be correct

if they said 75 percent. However, those in the negative base-rate manipulation would also

be correct if they said 25 percent. Therefore, it is necessary to divide participants into

separate groups in order to judge their accuracy.

Respondents in the positive base-rate versions were separated from the others and

scored based upon the percentage they gave. Those within 10 points in either direction of

75 percent, i.e. 65 to 85, were considered to be high in accuracy. Percentages at least 10

but no more than 25, i.e. 86 to 100 or 50-74, points off were considered to have moderate

accuracy. Anyone answering less than 50 was low in accuracy. Respondents in the

negative base-rate versions were calculated the same way; only their answers were taken

from the negative perception question. Combining these two questions into a one single-


70

item measure makes calculating the accuracy in recall more consistent and removes the

confounding variable.

Data Analysis

It should be noted that one limiting factor of the study may be due to effects of an

overpowered experimental design. The research called for 360 participants, but more than

480 individuals completed the study, potentially affecting interaction and main effect

results. However, Cohen’s (1988) seminal study regarding effect sizes stated that effect

sizes between .01 to .059, while considered small, were acceptable. Any effect size less

than .01 was viewed as trivial, whereas any number greater than .06 was deemed to have a

medium to large effect size. Therefore, while parts of the study may indeed be

overpowered, the effect sizes are still within an acceptable range.

To determine if the population means are different in order to satisfy the research

hypotheses ANOVAs were used primarily throughout the statistical analyses. In almost all

of the tests two or three independent variables were measured using economic perceptions

or recall as the dependent variable. The manipulation of economic news stories is

composed of four sets of independent variables: news story, base-rate, exemplars,

representativeness.

For hypotheses 1a and 1b, eight separate ANOVAs were run on each of these IVs,

with economic perceptions and recall of information as the dependent variables. More

specifically, each IV was analyzed independently of the other three IVs on each of the two

dependent variables. Each test looked at the relationship between one independent

variable and one dependent variable. For hypotheses 2a and 2b, a second independent

variable, economic literacy, was introduced and analyzed using the same method in H1a


71

and H1b. In these cases there were two independent variables examined on each of the

dependent variables respectively.

R1 employed the same measures as those in the first two hypotheses and once

again added another independent variable, arithmetic aptitude, to the analysis. Here an

ANOVA was run using the three IVs and each of the two dependent variables separately.

The last research question, R2, focused only on the group of independent variables found

in the economic news story set, the independent variable, political ideology, and the

dependent variable, economic perceptions. The dependent variable recall was not

examined as there is little literature indicating that political affiliation or ideology affects a

person’s ability to accurately recall or remember specific information. Figure 3.7

illustrates how the data was analyzed within the study.


72

Figure 3.7: Graphical representation of the method employed to measure the independent and dependent variables. Variables pointing directly to “overall news story”, “economic perceptions”, and “recall” are how they were categorized and measured. Independent variables pointing to the predictor lines were mediators


73

CHAPTER IV

RESULTS

H1a. Overall Economic Perceptions: Base-rate positive information, as well as positive

exemplars, will lead to perceptions that the overall U.S. economy is currently

strong/healthy or is in a strong recovery than will negative base-rate information or

negative exemplars.

Early exemplification studies have shown that when exemplars are representative

or reflective of the base-rate in a news article, then overall recall of the base-rate

information is greater (Zillmann, Perkins, & Sundar, 1992). However, there is little

research as to whether base-rate and exemplars potentially affect how individuals view a

broader topic touched upon within the story. The story manipulations in this study use the

economy as the main topic. However, the dependent variable economic perceptions is not

based upon how the participants perceive the economy of the town featured in the story

but upon how they see the overall health of the U.S. economy. Essentially, can factors

within a news story affect perceptions on a much broader scope?

Figure 3.7 (p. 71) provides a map that illustrates how many of the following

analyses were run. Participants in story one (M = 3.61, SD = 1.09), story three (M = 3.43,

SD = 1.28), and story five (M = 3.49, SD = 1.17) all perceived the economy as doing

better or was better off than those in story two (M = 3.10, SD = 1.06), story four (M =

3.05, SD = .98), and story six (M = 3.15, SD = 1.25). The main effect of story type on the

economic perceptions variable produced F5, 478 = 3.45, p < .05, ηp2 = .04, supporting

hypothesis one. The graph below illustrates how base-rate and exemplars affect those

perceptions.


74

Figure 4.1: Economic perceptions across story manipulations

A post hoc analysis across the different manipulations provided the first indication

that exemplars may be causing the significant difference among the stories and economic

perceptions. Figure 4.2 shows how the change of exemplars within the articles affects

overall economic perceptions.


75

Base-rate Positive Mean Difference Base-rate Negative Mean Difference

Article Representative

exemplars

Non-representative

exemplars

Mixed exemplars


Non-representative

exemplars

Mixed exemplars

Positive Ba

se-‐rate Representative

exemplars ___ 0.51** 0.18 0.56** 0.12 0.46**

Non-representative exemplars

-0.51** ___ -0.33 0.05 -0.39* -0.05

Mixed exemplars -0.18 0.33 ___ 0.38* -0.06 0.29

Negative Ba

se-‐rate Representative

exemplars -0.56** -0.05 -0.38* ___ -0.44* -0.09


-0.12 0.39* 0.06 0.44* ___ 0.35*

Mixed exemplars -0.46** 0.05 -0.29 0.09 -0.35* ___

*p < .05. **p < .01 Figure 4.2: Economic Perceptions by Story Version

The independent variable “exemplars,” created from the “story” variable, helped in

reinforcing this initial find. Exemplars produced a main effect of F2, 481 = 7.10, p < .01, ηp2

= .03 on economic perceptions (see figure 4.3). A Student-Newman-Keuls post hoc

revealed those in the positive exemplars (M = 3.55, SD = 1.13) condition held more

favorable economic perceptions among participants while those in the negative (M = 3.08,

SD = 1.01) and mixed (M = 3.29, SD = 1.25) exemplars condition tended to possess more

pessimistic outlooks on the economy. These results support hypothesis one.


76

Figure 4.3: Economic perceptions across exemplar conditions

The next analysis lends credence to the strong effect of exemplars on participants’

perceptions but negates the effects of base-rate data. There was no main effect of the

variable “base-rate” on overall economic views (F1, 482 = 2.12, p > .05, ηp2 = .004). Neither

of the two levels, base-rate positive (M = 3.38, SD = 1.15) or base-rate negative (M =

3.23, SD = 1.15), gave any indication they affected perceptions positively or negatively.

While the first analysis of “story version” on economic perceptions indicated

promising results in support of the first hypothesis, a deeper examination showed the

hypothesis was only partially supported. One half of the hypothesis was supported as the

variable “exemplars” did have a significant effect on perceptions of the economy.

However, the other half of the hypothesis showed that neither positive nor negative “base-

rate” information had any effect on economic perceptions.


77

H1b: Accuracy in Recall: Representative exemplars will lead to greater accuracy in

recall of base-rate information than will non-supportive or mixed exemplars.

The second hypothesis stated that representative exemplars, those that support or

accurately reflect the base-rate information, would lead to greater accuracy in the recall of

the base-rate information than non-representative or mixed. This hypothesis was

supported as there was a main effect (F2, 481 = 3.98, p < .05, ηp2 = .02) produced by the

variable “representative” on recall. Participants in the representative conditions (M = 1.88,

SD = .85) tended to have higher recall than those in the non-representative (M = 1.64, SD

= .82) and mixed (M = 1.67, SD = .79) conditions. Unfortunately, while those in the

representative condition possessed greater accuracy than the others, on average they were

just slightly below being moderately accurate in their recall.

Another sequence of analyses was run to ascertain if any of the other independent

variables potentially affected recall. In order to test the variables of “story version,”

“exemplars,” and “base-rate” information on recall, a series of ANOVAs were used. The

results for “story version” were F5, 478 = 1.64, p > .05, ηp2 = .02, “exemplars” was F2, 481 =

.62, p > .05, ηp2 = .003, followed by “base-rate” with F1, 482 = .20, p > .05, ηp

2 < .001. In

each analysis there was no main effect on the dependent variable recall, indicating that

only representativeness was significant.

H2a. Economic Literacy and Economic Perceptions: The effect of H1 is mediated by

economic knowledge such that people with high economic literacy will have perceptions

that more accurately reflect the base-rate information showing the overall U.S. economy

is currently getting stronger (recovering) or weaker (remaining stagnate or sliding back),


78

while those with low economic literacy will tend to be overly influenced by positive and

negative exemplars.

While studies have shown that arithmetic aptitude can play a role in mediating the

effects of exemplification (Zillmann, Callison, & Gibson, 2009), there has been little

research into what effect, if any, knowledge of a particular subject has. When solely

applied to economic perceptions, those with higher economic literacy perceived the

economy to be stronger (M = 3.43, SD = 1.15) than those with lower economic literacy (M

= 3.19, SD = 1.21) with a main effect of F1, 482 = 5.23, p < .05, ηp2 = .01.

There was a significant interaction between a person’s economic literacy, the story

condition they were in, and their economic perceptions, F5, 472 = 2.70, p > .05, ηp2 = .03

(see Figure 4.4). As the first analysis showed, those with high literacy tended to have

stronger views of the economy than those with lower literacy. This remained relatively

true throughout most of the conditions. The interaction between high and low literacy

occurred in the “base-rate positive: non-representative exemplar” condition, F1, 78 = 15.53,

p < .01, ηp2 = .17, as those with low economic literacy had largely different perceptions of

the economy based upon the story manipulation they were in. Those with high economic

literacy were consistent in their perceptions and were less swayed by the different stories.


79

Figure 4.4: Economic Perceptions, Economic Literacy, & Story Version

While the perceptions of those in high economic literacy stayed fairly consistent

and with little variance across all conditions, there were quite a few differences among

those possessing low economic literacy, F5, 244 = 4.37, p < .01, ηp2 = .08. Looking at the

chart from a macro-level, these individuals tended to be more affected by exemplars than

base-rate information. According to Figure 4.5, perceptions regarding the strength of the

economy jumped up or down based upon a person's condition.


80

Base-rate Positive Mean Difference Base-rate Negative Mean Difference

Article Representative

exemplars

Non-representative

exemplars

Mixed exemplars


Non-representative

exemplars

Mixed exemplars

Posi

tive

Bas

e-ra

te Representative

exemplars ___ 0.82** 0.13 0.55* -0.25 0.37


-0.82** ___ -0.69** -0.27 -1.07** -0.45

Mixed exemplars -0.13 0.69** ___ 0.42 -0.38 0.24

Neg

ativ

e B

ase-

rate

Representative exemplars -0.55* 0.27 -0.42 ___ -0.80* -0.18


0.25 1.07** 0.38 0.80** ___ 0.62*

Mixed exemplars -0.37 .045 -0.24 0.18 -0.62* ___

*p < .05. **p < .01 Figure 4.5: Low Literacy, Economic Perceptions Across Story Versions

An examination of economic literacy and exemplars revealed the interaction

between the two significantly reduced economic perceptions on those with low literacy

when negative exemplars were introduced, F2, 478 = 3.13, p < .05, ηp2 = .01. Participants

possessing low literacy tended to be much more affected by exemplars than those in the

high level. Low-level respondents who viewed positive exemplars saw the economy as

stronger (M = 3.56, SD = 1.15); negative exemplars created weaker perceptions (M = 2.77,

SD = 1.00); mixed exemplars found room in the middle (M = 3.20, SD = 1.34). The

discrepancy among those in the high level was much less. Individuals who saw positive

exemplars had slightly higher perceptions (M = 3.54, SD = 1.12) than those in the negative

conditions (M = 3.37, SD = .94) or mixed (M = 3.38, SD = 1.14). Figure 4.6 illustrates

how exemplars have little to no effect among those high in literacy. However, it does

show that those with low literacy are significantly affected by the exemplars within the

articles.


81

Figure 4.6: Economic Literacy, Economic Perceptions, & Exemplars

When it comes to economic literacy and base-rate information, an opposite effect

occurs between the levels of high and low literacy than was seen in the exemplar

condition. Here those with high economic literacy were more affected by base-rate than

those with low literacy. High literacy within positive base-rate conditions (M = 3.62, SD =

.96) lead to significantly greater economic perceptions than those in the negative condition

(M = 3.24, SD = 1.13). The same cannot be said about low literacy individuals. When

moving from positive base-rate (M = 3.16, SD = 1.25) to negative base-rate (M = 3.21, SD

= 1.17), there was no significant difference. However, there was an interaction effect of

economic literacy and base-rate information on economic perceptions, F1, 480 = 4.43, p <

.05, ηp2 = .01.


82

Figure 4.7: Economic Literacy, Economic Perceptions, &Base-rate Information

The results exploring the role of economic literacy across the manipulations of

story version, exemplars, and base-rate information support the second hypothesis. Low

economic literacy individuals are more likely to reflect the exemplars within story,

affecting their perceptions regarding the strength of the economy than high economic

literacy participants. However, as low literacy individuals are affected by exemplars, the

same cannot be said of base-rate information. As hypothesized, those with high economic

literacy are much more likely to reflect base-rate indicators within a story while focusing

less on the exemplars. Therefore, subject matter knowledge is an important factor when it

comes to exemplification.


83

H2b. Accuracy of Recall and Economic Literacy: The effect of H1 is mediated by

economic knowledge such that people with high economic literacy will have better recall

of factual data regardless of the representativeness of exemplars, while those with low

economic literacy will tend to be overly influenced by exemplars.

The role of subject matter knowledge on a person’s ability to recall information

has not been previously explored. As discussed earlier, the representativeness of

exemplars to base-rate information was a significant factor in the accuracy of recall. In

this case however, economic literacy did not play a significant role in respondents’ ability

to recall the base-rate. There were no main effects of literacy on recall as F1, 482 = .01, p >

.05, ηp2 = .01. Both low (M = 1.73, SD = .83) and high levels (M = 1.74, SD = .83) were

almost identical across the recall variable. The same issue occurred when the variable

“representativeness” was added to the analysis. Once again there was no interaction effect,

F2, 478 = 1.33, p > .05, ηp2 = .01, as there were few differences between the levels.

Therefore, hypothesis 2b was not supported as economic literacy does not have an effect

on accuracy of recall.

Economic Literacy

Low High

Representativeness M SD M SD Representative 1.89 0.86 1.88 0.85 Non-representative 1.55 0.77 1.73 0.86 Mixed 1.73 0.83 1.62 0.76 Figure 4.8: Economic Literacy, Recall, & Representativeness

R1. Arithmetic Aptitude, Economic Literacy, and Perceptions of the Economy: Under

varying conditions, does either arithmetic aptitude or economic literacy have a more


84

dominant influence over an individual’s evaluations as to the health/state of the economy

or ability to accurately recall economic base-rate information?

Incorporating arithmetic aptitude into the study helps to identify if individuals'

capability to calculate numbers affects their ability to understand and recall numeric

information. The two variables, arithmetic aptitude and economic literacy, were

moderately correlated, r (482) = .35, p < .01. The results from the analysis provide little

support that arithmetic aptitude affects individual’s perceptions regarding the strength of

the economy, F1, 482 = .57, p > .05, ηp2 = .001. Further analysis shows the interactions

across the different manipulations, story version (F5, 472 = .17, p > .05, ηp2 = .001),

exemplars (F2, 478 = .22, p > .05, ηp2 = .001), and base-rate (F1, 480 = .01, p > .05, ηp

2 = .001)

were not significant.

Figure 4.9: Arithmetic Aptitude and Economic Perceptions


85

The initial three-way interaction between arithmetic aptitude, economic literacy,

and story version did not produce a significant interaction effect, F5, 460 = 1.19, p > .05, ηp2

= .01, nor was there an interaction when story version was replaced with the exemplar

variable, F2, 472 = .59, p > .05, ηp2 = .002. Upon closer inspection, an interesting

phenomenon occurred. In the case of exemplars when aptitude was low, there was no

difference, F2, 258 = .63, p > .05, ηp2 = .01, in the levels of literacy. Exemplars in this case

did not have an effect on people’s perceptions of the economy regardless of their

economic literacy.

Low Arithmetic Aptitude Economic Literacy

Low High

Exemplars M SD M SD

Positive 3.61 1.23 3.69 1.25

Negative 2.90 0.99 3.38 0.88

Mixed 3.27 1.35 3.48 1.13 Figure 4.10 Low Aptitude, Economic Literacy, & Economic Perceptions

However, when arithmetic aptitude was high, there was an interaction affect

between exemplars and literacy, F2, 214 = 3.28, p < .05, ηp2 = .001.

High Arithmetic Aptitude Economic Literacy

Low High

Exemplars M SD M SD

Positive 3.44 0.98 3.46 1.04

Negative 2.37 0.95 3.36 1.00

Mixed 2.98 1.31 3.34 1.16 Figure 4.11 High Aptitude, Economic Literacy, & Economic Perceptions


86

The same phenomena occurred in the story manipulations as well. When

arithmetic aptitude was low there was no interaction effect between low and high

economic literacy F5, 252 = 1.42, p > .05, ηp2 = .03.

Figure 4.12: Low arithmetic aptitude, economic literacy, and story version

However, when arithmetic aptitude was high there was an interaction effect

between the two levels of economic literacy, story version and perceptions of the

economy, F5, 208 = 2.71, p < .05, ηp2 = .06.


87

Figure 4.13: High arithmetic aptitude, economic literacy, and story version

This indicates that economic literacy may have a more dominant influence over

individuals' perceptions than arithmetic aptitude while continuing to show that individuals

with high literacy are less affected by exemplars than those with low literacy.


88

Figure 4.14 High Aptitude, Economic Literacy, & Exemplars In comparison with the base-rate variable, there was no interaction effect between

literacy, aptitude, and a person’s perception of the economy, F1, 476 = 012, p > .05, ηp2 =

.001. Unlike exemplars, no difference was found between the different levels of literacy

and aptitude.


89

Low Arithmetic Aptitude High Arithmetic Aptitude Economic Literacy Economic Literacy

Low High Low High Base-Rate M SD M SD M SD M SD

Positive 3.24 1.28 3.72 1.01 2.96 1.16 3.55 0.92

Negative 3.29 1.18 3.22 1.08 3.04 1.15 3.25 1.16 Figure 4.15: Arithmetic Aptitude, Economic Literacy, & Base-rate Information

When it comes to accuracy of recall, there was no main effect involving arithmetic

aptitude, F1, 482 = .82, p > .05, ηp2 = .002, nor was there an interaction effect between

aptitude and the representativeness of exemplars, F2, 478 = .26, p > .05, ηp2 = .001.

Furthermore, the combination of aptitude and literacy on recall, F1, 480 = .03, p > .05, ηp2 <

.001, was not significant, and when combined with the representativeness of exemplars,

F2, 472 = .18, p > .05, ηp2 = .001, there was still no interaction.

There was a significant difference due to arithmetic aptitude when recall was

modified from the categorical variable and back into the original open-ended percentage

variable. A main effect was present between arithmetic aptitude and accuracy of recall

when this happened, F1, 482 = 4.00, p < .05, ηp2 = .01. Under the same conditions a main

effect did not occur when economic literacy was tested, F1, 482 = .95, p > .05, ηp2 = .002.

In answering the first research question, there is little doubt that economic literacy

had a more dominant influence over participants’ views on the strength of the economy

than did arithmetic aptitude. However, results showed that when it came to the accuracy

of recall, neither literacy nor aptitude played a dominant role. Only when the categorical

variable was replaced with the original percent did arithmetic aptitude show significance.


90

R2. Political Ideology and Perceptions of the Economy: Under varying conditions, what

role does political ideology have over an individual’s perception of the overall strength of

the economy?

The initial results of political ideology and perceptions of the economy emerged as

both common sense and research would suggest. Examining the single-item measure of

general political ideology, there was a main effect, F2, 481 = 13.61, p < .01, ηp2 = .05 (see

figure 4.14), as liberals (M = 3.72, SD = 1.15) perceived the economy to be slightly

stronger than moderates (M = 3.33, SD = 1.08) and substantially stronger than

conservatives (M = 2.93, SD = 1.18). However, across all story versions there was no

interaction effect, F10, 466 = .69, p > .05, ηp2 = .02.

Figure 4.16: Political Ideology & Economic Perceptions


91

Upon close examination of each individual story, significant differences were

found within the articles (see figure 4.17). Interaction effects occurred in the conditions:

positive base-rate: non-representative exemplars, F2, 77 = 7.1, p < .01, ηp2 = .16; negative

base-rate: representative exemplars, F2, 80 = 4.19, p < .05, ηp2 = .10; and negative base-

rate: mixed exemplars, F2, 80 = 4.25, p < .05, ηp2 = .10. Figure 4.17 provides a quick

overview that exemplars tend to have less influence over liberals than over moderates or

conservatives. It also shows support for Figure 4.16 in that liberals are generally going to

have more positive perceptions of the economy than moderates or conservatives

regardless of base-rate information or exemplars.

Figure 4.17: Political Ideology, Economic Perceptions, & Story Version


92

To further test the impact of these stories on political ideology, the conditions of

base-rate and exemplars will be analyzed as well. In the case of exemplars, we once again

do not have an overall interaction effect, F4, 475 = .29, p > .05, ηp2 = .002, as seen in Figure

4.16. Political ideology differences do occur in the conditions of negative exemplars, F2,

160 = 10.02, p < .01, ηp2 = .11, and mixed exemplars, F2, 159 = 4.3, p < .05, ηp

2 = .001.

Positive exemplars had no significant effect. This reaffirms the effect of exemplars on

moderates and conservatives while observing liberals consistent beliefs regarding the

economy.

Figure 4.18: Political Ideology, Economic Perceptions, & Exemplars


93

Again there was no interaction effect across the base-rate variable, F2, 478 = .57, p >

.05, ηp2 = .002 (see figure 4.19), but there were significant differences within each

condition. In base-rate positive, liberals (M = 3.73, SD =1.21) perceived the economy as

much stronger than conservatives (M = 2.96, SD = 1.06) and slightly stronger than

moderates (M = 3.46, SD = 1.06), F2, 235 = 7.14, p < .01, ηp2 = .06. When the base-rate

flipped to negative, liberals held an almost identical view regarding the economy’s

strength (M = 3.71, SD = 1.11); there was a tiny drop in the perceptions of moderates (M =

3.21, SD = 1.09); while conservatives stayed fairly consistent in their views as well (M =

2.90, SD = 1.21), F2, 243 = 7.17, p < .01, ηp2 = .06.

Figure 4.19: Political Ideology, Economic Perceptions, & Base-rate


94

CHAPTER V

DISCUSSION

Research into exemplification theory has often found that exemplars play a more

influential and decisive role on an individual’s perceptions of an event than does base-rate

information (Brosius & Bathelt, 1994). Further, some scholars even downplay the

prominence of base-rate information on how individuals form certain judgments (Koehler,

1996). While there is little doubt that exemplars play an important role within the

persuasiveness of information, there may be other unexplored internal factors within

personal characteristics of an individual that make base-rate data just as informative.

The primary purpose of the research was to build upon Zillmann, Callison, &

Gibson’s (2009) research into the mediation of personal characteristics within

exemplification studies by examining if subject matter knowledge, also referred to

throughout the study as aptitude or literacy, affects not only how an individual considers

and recalls information within a news article but also how they might perceive that subject

on a much broader scale. In essence could a person’s high level of knowledge regarding a

specific subject disrupt the influence of exemplars?

In this case economics was chosen for a number of reasons. First, it is an area that

is both relevant and also affects the lives of almost everyone in today’s society whether

the individual truly understands it or not. Among all news stories in 2010, the economy

was the most covered topic, even more so than the midterm elections (The Pew Research

Center, 2011). Next, it is a subject that has a natural fit within the basic fundamentals

required of exemplification: base-rate information and exemplars. Many times numbers

and percentages are incorporated into news stories that represent attributes of the


95

economy, e.g. unemployment, the housing market, personal finance, etc., and are often

followed with exemplars of those events. Ever more so, the fundamental nature of

experience through exemplars revolves around the idea that individuals make quantitative

assessments constructed not specifically on numerical estimates but in proportional terms

(Zillmann, 2000). Similarly, facets of economics in part revolve around people relying on

their “day-to-day personal experiences with the economy and observations of real-world

economic indicators when making assessments of current economic conditions” (Hester &

Gibson, 2003, p. 85).

Subsequently, an individual’s political ideology is another aspect of personal

characteristics that was examined. Politically both credit and blame go hand-in-hand when

discussing the state of the economy (Edwards, Mitchell, & Welch, 1995; Patterson, 1993),

and consequently the party of the sitting president and the party in control of Congress

play substantial roles in how Americans form economic opinions (Gerber & Huber, 2010).

It was therefore necessary that if this paper covered economic perceptions, political

ideology would also be taken into account.

Analysis of basic exemplification theory and economic perceptions

Before moving directly into the effects surrounding the aforementioned

characteristics of an individual, a test of basic exemplification hypotheses was run. Initial

results suggested exemplars had a more profound effect on economic perceptions than

base-rate, supporting the findings of many of the early exemplification studies (Aust &

Zillmann, 1996; Brosius, 1995; Gibson & Zillmann, 1994; Zillmann, 1999) Participants in

the positive condition with representative exemplars found the strength of the economy to

be generally higher, and as one might expect, the opposite occurred under the negative


96

condition with representative exemplars. In this scenario we can’t say with certainty

whether it was the base-rate or exemplars that caused the initial perceptions, but moving

forward the picture becomes a little clearer.

When the exemplars did not reflect the overall direction indicated in the story,

either positive or negative, perceptions began to change significantly. Those participants

in the positive base-rate: non-representative exemplar condition tended to believe the

overall strength of the economy was weaker, indicating that while the base-rate remained

positive, the exemplars had a more polarizing effect on their perceptions. Reinforcing this,

the same effect occurred within the negative base-rate: non-representative exemplars

condition. Zillmann (1999) discussed this phenomenon stating that when accurate base-

rate information is paired or reported along with exemplars that distort or are atypical

cases then the result often leads to less influential base-rate data. Therefore, a strong

argument could be made that exemplars were the driving force behind the profound and

significant changes within the experiment.

One of the more interesting findings focuses on perhaps the least telling of

variables: mixed exemplars. Mixed exemplars, while not intended to be a control

variable, do offer the opportunity for a reader to base their perceptions on conflicting

statements. Essentially, there is the potential for a individual to be influenced solely by the

negative exemplar set, the positive exemplar set, find the middle ground, or revert back to

the base-rate. If the base-rate is 75 percent directionally positive, the mixed exemplar

condition, because it is 50-50, can be seen as either half supportive with a 25 percent

difference between it and 75 or half non-supportive with a 25 percent difference toward

the negative side with an identical manifestation created when the direction is reversed.


97

The question is then asked, “What happens to perceptions when exemplars are neither

representative nor non-representative?” The results from this study suggest that

participants in conditions with mixed exemplars reverted back to the original base-rate

information. One possible answer for this is that the exemplars had an overriding effect on

one another, canceling one another in individuals' minds, leaving them with only the base-

rate. A more probable explanation is the base-rate primed the respondent to “see” the

economy moving in a certain direction, with the conflicting exemplars lacking any

dominance to overwhelm the supporting ones. Participants may have then been

subconsciously looking for evidence to support the base-rate information, and without

anything substantial to refute it, the base-rate remained.

In this scenario, Brosius’ (1995) explanation that base-rate is often an inferior

instrument of influence falters. It does however provide support to priming research

(Hetherington & Rudolph, 2008) that shows exposure to certain stimuli produce future

evaluations. In the case of representative exemplars the audience saw a base-rate number

regarding a population of people and exemplars that reflected that number. Whether base-

rate priming or exemplar influence occurred here is difficult to tell. In the case of non-

representative exemplars it seems likely the exemplars had a significant influence over the

base-rate, potentially negating any priming that may have occurred. But coming back to

mixed exemplars, it is possible that viewers were primed by the base-rate, and without

overriding exemplars to counter that phenomenon, the priming effect held. This may

coincide with Holbrook and Hill’s (2005) idea of activation stating that as information is

activated within memory the likelihood of similar issues being activated increases as well.


98

The problem here is determining whether or not this occurs only when exemplars are

mixed.

Further reinforcing the first hypothesis and providing additional support to the

“powerful” exemplar notion was the analysis of both base-rate and exemplar variables.

The findings showed that base-rate information, in either direction, had no significant

effect among the respondents. The difference, however, among positive, negative and

mixed exemplars was significant, indicating their overall impact with viewers. Whereas it

was hypothesized that base-rate positive stories as well as stories with positive exemplars

would contribute to stronger perceptions of the economy, only positive exemplars

supported this idea. This finding supports prior research that shows the dominant

influence exemplars often have over base-rate information (Brosius & Bethelt, 1994;

Gibson & Zillmann, 1994; Zillmann, Gibson, Sundar, & Perkins, 1996; Zillmann, Perkins,

& Sundar, 1992).

Analysis of the Effect of Recall

When evaluating participants’ accuracy of recall, only one of the variables,

representativeness, had any effect. The findings fully supported the second part of the first

hypothesis, demonstrating that when base-rate information is reinforced by representative

exemplars, individuals are more likely to recollect the information with a higher degree of

accuracy. Conventional wisdom would even indicate that most numerical data

accompanied with descriptive passages, quotes, or statements that reflect that information

would be more memorable within an individual’s memory (Zillmann & Brosius, 2000).

What could be said then is that it is more important for an individual to have an

understanding of the subject matter of the story than have the ability to accurately recall


99

and process numbers. An area like economics is not an easy field to pursue, even for the

most educated. This shows that accuracy in reporting economic information is critical to

the conveyance of what the information means and how people interpret it. As was seen in

this study, disproportionate exemplars affect not only the way readers view the story but

also their perception of the overall national economy. The implication goes deeper if

media outlets, through the use of representative exemplars, are able to increase the

accuracy of recall, potentially enhancing people’s remedial understanding of the subject,

while also strengthening the message.

The findings for both parts of the first hypothesis are by no means ground breaking

within exemplification research. Multiple studies (Berger, 2000; Koehler, 1996) have

shown the dominant role of exemplars over base-rate information as typical base-rate

figures lack the colorful narratives and often tend to be less vivid within the imagination.

What the results do provide is the basic groundwork from which to evaluate how personal

characteristics can potentially shift the influence away from exemplars toward the base-

rate.

Analysis of Economic Literacy

One such characteristic, arithmetic aptitude, has been shown to mediate the effects

of exemplars and base-rate within exemplification research (Zillmann, Callison, &

Gibson, 2009). This research pioneered some of the first steps in recognizing the

limitations of exemplars while acknowledging that certain traits reduce and even nullify

the effects. One area that has not been examined, however, is how one’s knowledge of a

subject might also mediate the effects of exemplification. Simple common sense would

say that a doctor reading a story on medicine or health would have a different perception


100

on the subject than the average layman. For example, if the media ran a story about a

prescription drug progressing through clinical trials, stating it had a 95 percent success

rate, and then derived the interviewees from the 5 percent in which it failed, then non-

healthcare professionals might see the drug as a failure based upon the interviewees’

experiences while doctors hailed its success based upon the base-rate.

And while that example focuses on a specific profession that requires a high level

of knowledge, another consideration might be in areas where elements of a story are left

out because they are either deemed irrelevant or the authors themselves are not aware of

the facts. The difference in these cases focuses on readers' knowledge of the subject in

which they might be aware of information not provided that changes their perception of

the story. For instance, in 2001 Ford Explorers and Expeditions were the target of a media

firestorm that focused on a high rate of deadly accidents surrounding the vehicles. It is

conceivable that a story would report that less than five percent of Explorer owners might

be affected followed by a string of interviews from owners who described near death

experiences because of the defective design. The average car owner with little knowledge

of how a vehicle works may worry that all Ford Explorers and Expeditions are dangerous,

warning anyone who will listen to avoid them. What the story may fail to mention is that

anyone with a high level of interest or knowledge of automobiles may know that it was

not the car but tread separation and air-pressure within the tires. So education of an issue,

not level of intelligence, may be the most important determining factor in how individuals

interpret the base-rate and exemplars within the media.

Until now, this hypothesis had not been tested under exemplification. The results

showed that subject matter knowledge does play a role in how effective exemplars and


101

base-rate are in forming our perceptions. Overall, those with high economic literacy

believed the economic conditions in the U.S. were stronger than those with low literacy,

but more importantly those individuals’ perceptions were more closely related to the base-

rate information and less affected by exemplars. The real interaction and differences

occurred across the story manipulations. Higher economic literacy participants were not

influenced by the exemplars within the articles, which could indicate that they read the

articles as they were intended to be viewed. Representative, non-representative, and mixed

exemplars played almost no part in their understanding of the article. Directionally

positive stories tended to be viewed as indicators of a stronger economy while

directionally negative stories were viewed as indicators of a weaker economy. These

findings are similar to Walstad and Rebeck’s (2002) research that showed individual’s

possessing higher economic aptitude had opinions that were more reflective of real-world

economic conditions.

It could be conjectured that those with low economic literacy were oblivious to the

base-rate direction within the stories. When the exemplars were representative, there was

virtually no difference between high and low economic literacy, occurring almost

identically in the mixed exemplar conditions as well. The situation quickly changed when

non-representative exemplars were introduced. Whereas respondents with high economic

literacy stayed consistent throughout the conditions, those with low literacy were

significantly influenced by the exemplars. Negative exemplars were overly influential in

the directionally positive base-rate condition with the exact opposite occurring between

positive exemplars within the directionally negative base-rate condition. This shows that

economic perceptions of those with low economic literacy were formed around statements


102

given by those interviewed in the stories rather than the preceding base-rate. Therefore, it

is very likely that those with low economic literacy are almost solely focused on the

exemplars within a story, making them less likely to recall substantive information

(Brosius, 2003; Kazoleas, 1993).

By eliminating the base-rate variable and only focusing on the effect of exemplars,

the findings help support the argument advanced in the previous discussion. There was no

effect of exemplars, whether positive, negative, or mixed, on individuals with high

literacy. Their economic perceptions once again held steady across the conditions.

However, the exemplar variable does highlight an area that was not apparent in the story

variable. Participants with low literacy had essentially identical views on the economy as

those with high literacy under positive and mixed conditions. Once the negative condition

was introduced, their perceptions dropped substantially. A few assumptions could be

made from this. First, that as audiences view recent economic coverage as

overwhelmingly negative (The Pew Research Center 2011), they have been primed or

tend to focus more on negative exemplars within news stories. And secondly, that

negative exemplars are more persuasive than positive, reinforcing research (Soroka, 2006)

that shows reporters believe that viewers value negative information more than positive

information. However, without further evidence this cannot be said with any certainty.

Another area within the research targets the base-rate variable while eliminating the effect

of exemplars. Here we see an opposite outcome from the previous analysis in that high

literacy is affected while low literacy remains unchanged. Individuals possessing high

economic literacy perceived stronger economic conditions when they viewed stories under

the positive base-rate manipulation with those perceptions decreasing when the base-rate


103

switched to negative. Low literacy individuals were unaffected, maintaining an almost

even opinion across base-rate conditions. According to Kazoleas’ (2004) findings, this

may indicate that higher levels of economic literacy result in increased critical thinking,

while reducing the use of heuristics and shortcuts.

The evidence from the first part of the second hypothesis confirms that an

individual’s level of knowledge regarding a particular subject does have a mediating

effect within exemplification. As was expected, those with little knowledge of the subject

were more susceptible to the effects of exemplars, essentially relying on the testimony of

others to form their opinion. What was uncertain, however, was the role high literacy

would play. The research established exemplars are less influential over those with high

literacy as they were found to be more affected by the base-rate information provided.

This gives additional support to Zillmann, Callison, and Gibson’s (2009) research that

showed how personal characteristics do indeed affect the processing of base-rate and

exemplars under exemplification. This leads into the second part of the hypothesis that

seeks to determine if high literacy also translates into the ability to recall information

more accurately.

Unfortunately, the results indicated there were no differences between high and

low levels of literacy in gauging a person’s ability to recall information. Across all

conditions there was never an interaction effect suggesting that literacy did not play a role

in recall. Examining the issue further indicates that literacy would probably not be an

influential factor on one’s memory. Elements within the story, such as the

representativeness of exemplars, are more likely to make the information more accessible

than any character trait of the individual reading it. In some situations, level of knowledge


104

may make a difference in a person’s ability to remember facts and figures. However, this

study found no evidence to support that hypothesis.

Analysis of Arithmetic Aptitude and Economic Literacy

Recently, the most studied personal aspect under exemplification has been the

ability to understand and process numerical data. The test of arithmetic aptitude has

shown that as the levels of knowledge increase, so does the ability to recall quantities and

proportions (Zillmann, Callison, & Gibson, 2009). The effect arithmetic aptitude might

have on economic perceptions is something entirely new. Across story conditions,

aptitude was not significant as those with both high and low scores saw the strength of the

economy relatively the same. The same results occurred throughout the exemplar, base-

rate, and representative conditions as well, indicating that mathematical understanding

does not impact one's perceptions of the economy.

The comparison between arithmetic aptitude and economic literacy provided little

evidence that there was a relationship between the two. Once again across the conditions,

there were no interaction effects between the two on economic perceptions. Two areas

where the interactions can be seen are between the levels of low and high arithmetic

aptitude and economic literacy in the conditions of story version and exemplars. When

aptitude was low, those with both high and low economic literacy tended to be more

influenced by exemplars. However, when arithmetic aptitude was high, there was a

difference between the levels of literacy. Once again low literacy levels were easily

influenced by the exemplars, but in this case high levels were not. Individuals who

possess low aptitude, regardless of high or low economic literacy, will undoubtedly then


105

be affected by exemplars within a story while those with high aptitude will vary

depending on their economic literacy.

Previous research (Zillmann, Callison, & Gibson, 2009) states that in some cases

arithmetic aptitude should play a role in participants’ accuracy of recall. Unfortunately,

none of the findings suggested that aptitude was a foolproof predictor of recall. Another

analysis changed the nature of these findings when the measure regarding the accuracy of

recall was altered from a categorical variable and reverted back to the open-ended

percentage the respondents gave. In this case instead of low, moderate, and high accuracy,

the findings focused on how close each level of aptitude moved toward the correct base-

rate percentage. Individuals with high arithmetic aptitude were significantly closer to the

original base-rate data than those with low aptitude, signifying that high aptitude

participants either had better memories or were less affected by exemplars than originally

believed. When the same measurements were applied to economic literacy, there was no

significant difference between high and low levels.

The interesting results from this research question show that economic literacy has

a stronger influence on how people perceive the economy and how those perceptions are

more affected by exemplars and base-rate information while also demonstrating the

nominal impact arithmetic aptitude has. On the other side, arithmetic aptitude is a much

better measurement and indicator of actual base-rate recall than literacy was.

Analysis of Political Ideology within Exemplification Theory

Earlier results have shown that individual knowledge of an issue, as well as the

ability to calculate and estimate numerical data, can affect the way a person processes

information viewed in the media. These characteristics focus more on an internal


106

understanding and inherent aptitude than on personal opinion and assumptions. The next

set of personal characteristics moves away from how a person thinks about specific

information to how a person applies that information to individual philosophical political

beliefs and how that applies to the world around them.

The findings showed, as one might expect, that liberals viewed the economy as

stronger than moderates and conservatives. And while no questions were asked to

determine how participants felt about President Obama or the U.S. Congress, it would be

easy to speculate based upon both previous research (Gerber & Huber, 2010) and common

sense that those who most identify with the party in the White House will see the country

as doing better than those ideologically opposed to those in control. If this study had been

conducted in 2008, the results, as far as political ideology is concerned, might logically be

reversed with conservatives declaring a stronger economy than liberals.

The manipulation of economic stories provided an interesting jumping-off point to

see if a relationship between political ideology and exemplification existed. As was

expected, liberals and conservatives were not nearly as affected by exemplars to the extent

that moderates were. Throughout the different versions, liberals stayed relatively

consistent in their views of the economy; on its face moderates appeared somewhat

affected by both base-rate and exemplars while conservatives throughout each condition

consistently ranked the economy fairly weak.

A question that arises from the analysis is why were liberals so consistent across

all versions while conservatives tended to fluctuate more? Conservatives did rank the

economy as weaker than liberals and moderates, but the strength of those views does not

appear to be based solely on political ideology. Unfortunately, the study did not seek to


107

ascertain the strength of ideological conviction, only what relationship it might have. The

time period the experiment was conducted also fell close to a heavily debated and watched

presidential election, making these issues much more salient to audiences. As the 2012

political election cycle gears up, President Obama (D) has been persistently criticized for

his economic policies and the state of unemployment, which is consistent with historical

analyses that the tone of news coverage often is highly correlated with a president’s

approval rating (Graber, 1993) and that incumbent president’s often receive more

economic coverage than their opponents (Shah et al., 1999). Perhaps due to the campaign

attacks on his economic record, liberals feel the media is too harsh in its tone or is

providing too much negative reporting regarding the president. If this is indeed the case,

liberals may be so headstrong in their beliefs that it is possible their perceptions are

unaffected regardless of positive or negative base-rate or exemplars. If that hypothesis is

true, then it would seem sensible that conservatives would have an equal, albeit opposing

strength in their belief. The findings show there is no significant difference between

conservative ideology and the story versions even as Figure 4.15 seems to illustrate more

movement along the different story manipulations. Therefore, liberals and conservatives

do indeed have consistent, polarizing views of the economy. The most interesting and yet

conceivable finding of the three came from moderates as their perceptions significantly

differed across the stories. This suggests that moderates may not be as resilient or rely as

heavily upon their ideological views when reading or viewing economic news stories.

Political pundits and pollsters generally focus on independents and moderates in political

races, stating they often don’t make concrete decisions about parties and candidates until

right up till the election date (Miller, 1980). This finding furthers the notion that


108

moderates are more influenced by facets of base-rate and exemplars, perhaps indicating

that their minds are not yet made up.

Through the examination of both exemplar and base-rate variables, we are able to

delve deeper into these findings, illustrating the effect these variables have on economic

perceptions when combined with political ideology. The role of exemplars emphasizes

what was seen in the earlier analysis in that positive exemplars generally create stronger

perceptions of the economy across ideology; negative exemplars, with the exception of

liberals, significantly weakens those perceptions; and mixed exemplars tend to force

individuals to find middle ground between the two. Further reinforcing this are the

differences between liberal, moderates, and conservatives with the base-rate condition.

Once again liberals and conservatives were reliably consistent across both conditions. The

economy, in the eyes of liberals, was stronger regardless of the base-rate while

conservatives were almost identical in their belief that the economy remains weak.

Moderates, however, were significantly affected by positive and negative base-rate

information, lending support to the idea that both the media and campaigns may have a

prominent influence over their perceptions.

So what role do political ideologies play in exemplification and overall perceptions

of the economy? Unfortunately, because of the lack of significance among liberals and

conservatives with respect to exemplars and base-rate, it could be said these two groups in

effect have blinders on when it boils down to how they see they news. Neither factual

information presented in numerical form nor the opinions and statements offered by those

within the story affect what seems to be pre-existing ideas of that which is occurring

within the economy. The polarization of the parties and the growing extremism among


109

political ideologists almost creates a bubble or vacuum in which conflicting views are met

with hostility or derision. Sadly, the state of politics in our country no longer tends to lend

itself to open-mindedness or acceptance of information that does not correspond with our

beliefs.

However, for all the zealotry within the political arena, the emphasis on the power

of moderates and their ability to shape elections remains essential. The results showed that

while they may be more influenced by base-rate information and exemplars, it could also

be interpreted as moderates’ capacity to remain flexible, unbiased, and more open to

conflicting information. The results also show the danger of how the media could

manipulate economic perceptions through the reporting of news stories and more

specifically the exemplars used within those stories. If they overly exaggerate those

exemplars in either direction, then perceptions of that event, issue, or occurrence may be

distorted, and the true reality of the story is lost. Consequently, the media has a

responsibility to be as representative with their exemplars when discussing an event

population.

Theoretical Implications

The findings discussed in this paper continue to build upon early exemplification

research put forth by Gibson and Zillmann (1994). As Brosius (1995) once pointed out

that base-rate information was often regarded as an inferior instrument of influence this

study shows that his statement is both right and wrong. However, this study was not the

beginning or origin that first explored this phenomenon, it only sought to build upon

recent findings. Zillmann, Callison, and Gibson (2009) provided some of the first clues

and analysis as to what factors might short-circuit the influence of exemplars. When they


110

found that an individual’s arithmetic aptitude played a significant role in the processing of

base-rate and exemplars a new area of potential research opened up in exemplification.

According to Zillmann, Callison, and Gibson (2009), as an individual’s aptitude

increases, their understanding and view of a story becomes less influenced by exemplars,

while the inverse occurs with lower aptitude. This helped highlight how certain personal

characteristics that individuals possess help break down how we understand

exemplification. This study reinforces their findings and moves the theory forward by

applying subject-matter knowledge to the personal characteristics literature. Complex

topics such as the economy are often difficult to fully comprehend without some degree of

education gained from academia or experience. The results found here indicate that as one

becomes more knowledgeable of a topic then the effect of exemplars diminishes, whereas

similar to the arithmetic aptitude variable those with lower aptitude or literacy are more

susceptible to a greater influence exerted by exemplars. These findings partially contradict

early exemplification research (Brosius & Bethelt, 1994; Gibson & Zillmann, 1994;

Zillmann, Gibson, Sundar, & Perkins, 1996; Zillmann, Perkins, & Sundar, 1992) that puts

forth the notion of powerful exemplar effects.

There is little doubt those authors are still correct in their assessment regarding the

vividness and influence of exemplars. When aptitude and literacy were excluded from the

analysis there were significant differences throughout the story manipulations. Base-rate

information, both positive and negative, had little to no effect on the participants.

However, the direction or tone of exemplars, positive and negative, did affect individual’s

perceptions of the economy, further supporting early exemplification research.


111

Just recently, Gibson, Callison, and Zillmann (2011) examined how “individual

differences in arithmetic competence have notable implications for persons’ assessment of

situations that have relevant quantitative characteristics and also instigate a degree of

emotional reactivity” (p. 24). The results found in this research, coupled with previous

findings (Callison, Gibson, & Zillmann, 2011; Zillmann, Callison, & Gibson, 2009), push

forward the idea that future exemplification research should not only focus on criteria and

facets of the message, but on characteristics of the receiver as well.

Practical Implications

The results illustrated the dual roles of base-rate information and exemplars and

their effect on audience’s perceptions of the economy. Less knowledgeable audiences tend

to be more influenced by exemplars, whereas when a participant is more knowledgeable

about a subject we see a diminishing effect on that influence. The application of these

results indicates that in many situations the primary focus should first be in understanding

audience characteristics for which the message will be developed.

Niche areas and fields such as trade groups, professional organizations, and

technical journals may be encouraged to use more base-rate numbers and data as their

members and readers should be able to extract the facts more efficiently. The addition of

representative exemplars regarding these events and populations have the ability to

strengthen the recall of the message, but overall the substance may not be the most

important aspect of the message. In these cases public relations practitioners, journalists,

and executives may not have to worry about how much additional detail is needed to tell

the story. As the research showed, those with high literacy had consistent views across the

conditions as exemplars had little influence on their perceptions. This is not to say that


112

everyone reads articles the same way and all share similar viewpoints, but it does suggest

they are able to process the information comparably.

Not all media and organizations have the luxury of niche audiences. In these cases

those outlets must consider the proper use and implementation of base-rate information

and exemplars. Individuals with greater subject-matter comprehension will hopefully

understand the information as it was intended. The issue at hand becomes how to present

that information to those with little or no understanding of the subject. Here, when

constructed appropriately exemplars can make the message more impactful. Some

organizations should be able to do this with little difficulty. For instance, a financial

officer issuing an annual report to stockholders can and should use exemplars that are

representative of the information being provided. In the event that a stockholder does not

understand what the financial information (base-rate) means, the exemplars can be used to

more vividly paint a picture, providing greater meaning. For news outlets this may be a

more difficult task. Journalists cannot, within reason, be expected to seek out quotes and

statements that accurately portray the exact proportion of the population “exemplified”

within the story. In these moments a journalist needs to ask himself or herself if the

disproportionate use of exemplars detracts and takes away from the true meaning of the

story or has the ability to distort audience’s perceptions of the issue. If there exists a risk

the journalist may need to provide additional background information regarding the topic

in an effort to make the reader more knowledgeable, or at the very least to minimize any

misrepresentation or misconceptions.

When exploring the practical applications of exemplification theory it is important

to note that the theory can only be applied in limited ways. Much of this is due in part to


113

the complicated nature of how it is used. First, base-rate information must pertain to or

explain a specific event or population. When discussing the economy, GDP base-rate

information, while informative, cannot be “experienced” in a way that exemplars could

explain. Which isn’t to say that the economy can’t be used as unemployment, inflation,

and consumer confidence can all be reinforced with statements and quotes from people

experiencing those events. Second, the ability to proportionally present exemplars in

relation to base-rate information may not always exist. In some cases there may be an

inadequate population to draw from, limiting communicators’ options to make the

exemplars representative. This isn’t to say exemplars must be 100 percent proportional to

the base-rate, but as discussed earlier a person runs the risk of miscommunicating the

message.

Limitations

No study, no matter how well-thought out and planned, is without its limitations.

Perhaps one of the greatest strengths and biggest concerns focuses on the sample used.

Research (Walstad & Rebeck, 2002) has shown that individuals with only one college

economics course are more likely to have different economic views than individuals

without any economics course above high school. And while one of the objectives of an

experiment is not always to generalize to a specific population, the use of college students

as a sample did not seem appropriate to the overall goal. Therefore, a non-probability

sample of the general population of the United States was used in order to collect more

robust data. This allowed more variation in age, gender, ethnicity, education, and political

ideology, creating a more diverse group of participants; and while the use of a non-


114

probability sample, more specifically one taken from a sampling firm, could potentially

pose its own issues, the reward of having such a diverse sample far outweighed the risks.

The second limitation, unfortunately, is considerably less easier to control. The

current economic and political environment could be described as hostile or at the very

least uneasy. Any study with a focus on the economy conducted during a time of

economic uncertainty is not going to be without its critics. Adding to that existing

confounding variable is the circumstance under which the study was conducted. The

economy becomes a more sensitive issue, covered extensively by the media, during a

presidential election year. Some might argue the heightened coverage and tone given by

the media would skew the results in one direction or the other, or that individuals have

been so bombarded and exposed to messages that any semblance of being unbiased is

futile. If this study was conducted during a non-presidential election cycle, when the

economy was sound, or when the president and congress were not viewed so poorly, then

perhaps economic perceptions, especially when measured with political ideology, might

be more easily influenced. The results from the study did, however, support the

hypotheses with the research questions providing clear answers. Research (Horner, 2008;

Zucker, 1978) also shows that because the economy is an obtrusive issue, there will

always be, regardless of timing, on-going, evolving perceptions regarding the subject.

Conclusion

The focus of this study was designed to determine if certain characteristics of

individuals potentially affect the way they view news stories. Empirical research (Berger,

2000) has identified the overwhelming and influential nature of exemplars over base-rate

information, providing evidence to the strength of their persuasive nature. If the media,


115

when reporting the news, choose not to use interviews or statements that proportionately

reflect the event taking place, there is little doubt based upon exemplification research that

many people will be overly influenced by those exemplars. The reality of the story can be

distorted as the exemplars present an inaccurate depiction of actual events. Only recently

have studies (Zillmann, Callison, & Gibson, 2009) begun to show that specific aspects of

an individual can override and short circuit the effects of exemplars. Arithmetic aptitude is

one of those traits, and from this study we now know that subject matter knowledge, more

specifically economic literacy, is another.

The findings highlight the substantial persuasiveness of exemplars and the

susceptibility of individuals with lower aptitude and literacy to them. The consequences of

irresponsible or lazy reporting could range from mild to severe. If the subject of a news

article focuses on amusement parks and the probability of people who get sick on rides,

most people might read it, and based upon the exemplars, they might also think that

almost everyone becomes ill or that the rides really do not cause sickness. The article

probably is not going to change a individuals' perceptions on rides, change amusement

park behavior, or hurt the amusement park industry.

Taking it another step we can use the current economic-exemplification

experiment as an example. The results showed that exemplars did have an overall effect

on participants’ perception as to the strength of the economy. This is important because

many economic studies (Blood & Phillips, 1995; Deboef & Kellstedt, 2004; Fan, 1993;

Ladner & Wlezien, 2007) agree that consumer confidence is one of the major components

in judging the health of the economy. Therefore, if news stories are negatively affecting

those perceptions, we could see a decrease in consumer spending, job hiring, or investing,


116

in effect harming the economy. The media may not even be aware of the influence its

choice of exemplars is having on the public. Perhaps the most severe ramifications of

using non-representative exemplars would be in healthcare related articles. If a story

discussed the high-rate of cancer among teens using tanning booths followed by

exemplars of teens who have been tanning for years without problems, this account may

lead to attitudes that tanning beds really are not all that dangerous. The consequences of

that attitude would be far more reaching than incorrect perceptions of going to the

amusement park. Another example would be a story focused on the possible side effects

of a new pharmaceutical drug. If the article stated that of those who took the drug less

than 2 percent experienced any problems while everyone quoted within the story said they

did experience a negative side effect, then people who need the medicine may have a

grossly unbalanced perception of how it works and may avoid taking it. So not only do

exemplars affect how viewers see a particular news story or article, but they also affect

how people see the issue or event in its entirety, potentially changing their behavior in

accordance.

Fortunately, this effect can be countered when individuals are more knowledgeable

regarding the subject. In the case of the amusement park rides, the issue may be so

inconsequential that it may not matter if someone has high or low knowledge of the

subject. However, as we saw in this study, those with high economic literacy were more

likely to have economic perceptions that mirrored the base-rate and were less affected by

the exemplars. If they have a better understanding of how the economy works, then

inaccurate exemplars within the news should be less likely to affect their judgments and

attitude toward the subject. The same reasoning should apply to those more


117

knowledgeable about healthcare. One doesn’t necessarily need to be a doctor or nurse to

filter out the irrelevant aspects of a story to understand what the reporter is trying to say.

The importance of these studies highlights the necessity for people to be educated

about critical issues. The media does have influence over audience’s perceptions,

attitudes, and behavior, but that influence can be diminished and reduced when people

take the time to better understand the issue. In some cases, as we saw with political

ideology, people will always have blinders on, and no matter what they see or hear those

beliefs are not going to change. But that may also be limited to certain issues or events,

and perhaps through education and the advancement of knowledge, they too will be able

to see past the misrepresentation of information.


118

BIBLIOGRAPHY

Abate, T. (2008, December 12). It's official: U.S. in recession all of 2008. San Francisco

Chronicle, p. A1.

Alsem, K. J., Brakman, S., Hoogduin, L., & Kuper, G. (2008). The Impact of newspapers on

consumer confidence: Does spin bias exist? Applied Economics, 40(4-6), 531–539.

Altheide, D. L. (1997). The news media, the problem frame, and the production of fear.

Sociological Quarterly, 38(4), 647-668.

The American National Election Studies (ANES; www.electionstudies.org). The 2006 ANES

pilot study full release dataset. Stanford University and the University of Michigan.

Ansolabehere, S., Marc, M., & Snowberg, E. (2011). Sociotropic voting and the media. In The

ANES Book of Ideas, ed. John H. Aldrich and Kathleen McGraw. Princeton University

Press.

Ansolabehere, S., Rodden, J., & Snyder, J. M. (2008). The strength of issues: Using multiple

measures to gauge preference stability, ideological constraint, and issue voting. The

American Political Science Review, 102(2), 215–232.

Ansolabehere, S., Snowberg, E. C., & Snyder Jr., J. M. (2005). Unrepresentative information.

Public Opinion Quarterly, 69(2), 213–231. doi:Article

Associated Press (2008) Bush: U.S. is not headed for recession. MSNBC, Retrieved from

http://www.msnbc.msn.com/id/23389508/ns/business-stocks_and_economy/t/bush-us-

not-headed-recession/

Aust, C. F., & Zillmann, D. (1996). Effects of victim exemplification in television news on

viewer perception of social issues. Journalism & Mass Communication Quarterly, 73, 787–803.


119

Bates, E.M. (2010). Public relations via new media: Influence of blog postings and comments on

organizational perception. (Doctoral Dissertation). Retrieved from Texas Tech University

Libraries Theses and Dissertations.

Blanchard, O. (1993), Consumption and the recession of 1990-1991. American Economic Review,

83(2), 270-274.

Blendon, R.J., and others (1997). Bridging the gap between the public’s and economists’ views of

the economy. Journal of Economic Perspectives, 11(3), 105–118.

Blinder, A. (1997). Is government too political? Foreign Affairs, 76, 115–126.

Blinder, A. S., & Krueger, A. B. (2004). What does the public know about economic policy, and

how does it know it? Brookings Papers on Economic Activity, (1), 327–387.

Blood, D. J., & Phillips, P. C. B. (1995). Recession headline news, consumer sentiment, the state

of the economy and presidential popularity: A time series analysis 1989-1993.

International Journal of Public Opinion Research, 7(1), 2 –22. doi:10.1093/ijpor/7.1.2

Bloom, H. S., & Price, H. D. (1975). Voter response to short-run economic conditions: The

asymmetric effect of prosperity and recession. The American Political Science Review,

69(4), 1240–1254.

Breen, G. M., & Matusitz, J. (2008). Communicating the negative aspects of pack journalism to

media reporters. Global Media Journal, 7(12), 1-10.

Brewer, M.B., & Kramer, R.M. (1986). Choice behavior in social dilemmas: Effects of social

identity, group size and decision framing. Journal of Personality and Social Psychology,

3, 543-549.

Brosius, H., & Bathelt, A. (1994). The utility of exemplars in persuasive communications.

Communication Research, 21, 48-78.


120

Burns, B. (1992). Percepts, concepts and categories: The representation and processing of

information. Amsterdam: Elsevier Science Publishers.

Callison, C., Gibson, R., & Zillmann, D. (in press). Effects of differences in numeric ability on

the perception of adversity risk to others and self. Journal of Media Psychology.

Callison, C., Gibson, R., & Zillmann, D. (2009). How to report quantitative information in news

stories. Newspaper Research Journal, 30(2), 43–55. doi:Article

Citizenship and critical economic literacy: Vitalizing economics standards for system change.

(2010). Social Advocacy & Systems Change, 2(1), 11–43. doi:Article

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New Jersey:

Lawrence Erlbaum.

Cook, F. L., Tyler, T. R., Goetz, E. G., Gordon, M. T., Protess, D., Leff, D. R., & Molotch, H. L.

(1983). Media and agenda setting: Effects on the public, interest group leaders, policy

makers, and policy. Public Opinion Quarterly, 47(1), 16 –35. doi:10.1086/268764

Croteau, D. (1998). Challenging the "liberal media" claim. Extra!, 11(4), 4-‐9.

Daly, C. B. (2011, August 5). Ex-Obama adviser thinks U.S. double-dip recession chances are 50-

50. CBS News, Retrieved from http://www.cbsnews.com/8301-503544_162-20088580-

503544.html

DeBoef, S., & Kellstedt, P.M. (2004). The political (and economic) origins of consumer

confidence. American Journal of Political Science, 48(4), 633–649.

Diemer, T. (2010). Joe Biden: It's still the 'Bush recession.'

Retrieved from http://www.politicsdaily.com/2010/07/29/joe-biden-its-still-the-bush-

recession/


121

Domke, D., Fan, D.P., Fibison, M., Shah, D.V., Smith, S.S., & Watts, M.D. (1997). News media,

candidates and issues, and public opinion in the 1996 presidential campaign. Journalism

and Mass Communication Quarterly, 74(4), 718-737.

Dua, P., & Smyth, D. J. (1993). Survey evidence on excessive public pessimism about the future

behavior of unemployment. Public Opinion Quarterly, 57(4), 566–574. doi:Article

Eccles, R. G., & Nohria, N. (1992). Beyond the hype: Rediscovering the essence of management.

Boston: Harvard Business School Press.

Economy dominates public's agenda, dims hopes for the future. (2011, January 20). Pew

Research Center. Retrieved from http://www.people-press.org/2011/01/20/economy-

dominates-publics-agenda-dims-hopes-for-the-future/

Edwards III, G. C., Mitchell, W., & Welch, R. (1995). Explaining presidential approval: The

significance of issue salience. American Journal of Political Science, 39(1), 108.

Elliott, C.S., & Hayward, D.M. (1998). The expanding definition of framing and its particular

impact on economic experimentation. Journal of Socio-Economics, 27(2), 229–243

Ellis, P.D. (2010). The Essential Guide to Effect Sizes: An Introduction to Statistical Power,

Meta-Analysis and the Interpretation of Research Results. United Kingdom: Cambridge

University Press.

Entman, R. M. (1993). Framing: Toward clarification of a fractured paradigm. Journal of

Communication,43(4), 51-58.

Erikson, R. S., MacKuen, M. B., & Stimson, J. A. (2000). Bankers or peasants revisited:

Economic expectations and presidential approval. Electoral Studies, 19, 296-312.

Fan, D. (1993). Predictions of consumer confidence/sentiment from the press. Paper presented at

1993 annual meeting of American Association of Public Opinion Research.


122

Federal Reserve Bank of Minneapolis (FED). (1998, December). The Minneapolis Fed’s national

economic literacy survey. The Region, 12(4), 12–15.

Fogarty, B. J. (2005). Determining economic news coverage. International Journal of Public

Opinion Research, 17(2), 149 –172. doi:10.1093/ijpor/edh051

Gerber, A. S., & Huber, G. A. (2010). Partisanship, political control, and economic assessments.

American Journal of Political Science, 54(1), 153–173. doi:Article

Gibson, R., Callison, C., & Zillmann, D. (2011). Quantitative literacy and affective reactivity in

processing statistical information and case histories in the news. Media Psychology, 14,

96-120.

Gibson, R., & Zillmann, D. (1994). Exaggerated versus representative exemplification in news

reports: Perception of issues and personal consequences. Communication Research, 21(5),

603-624.

Gibson, R. & Zillmann, D. (1998). Effects of citation in exemplifying testimony on issue

perception. Journalism & Mass Communication Quarterly, 75(1), 167-176.

The global slumpometer. (2006, November 06). The Economist, Retrieved from

http://www.economist.com/node/12553076

Goidel, K., Procopio, S., Terrell, D., & Wu, H. D. (2010). Sources of economic news and

economic expectations. American Politics Research. doi:10.1177/1532673X09355671

Goidel, R. K., & Langley, R. E. (1995). Media coverage of the economy and aggregate economic

evaluations: Uncovering evidence of indirect media effects. Political Research Quarterly,

48(2), 313–328.

Graber, D.A. (1993). Mass Media and American Politics (4th ed.), Congressional Quarterly Press.


123

Haiman, F. S. (1999). The voices of extremism revisited. Critical Studies in Mass

Communication, 16, 119-135.

Haller, H. B., & Norpoth, H. (1997). Reality bites: News exposure and economic opinion. Public

Opinion Quarterly, 61, 555-575.

Haller, H. B., & Norpoth, H. (1994). Let the good times roll: The economic expectations of US

voters. American Journal of Political Science, 38(3),625-650.

Harrington, D. E. (1989). Economic news on television: The determinants of coverage. Public

Opinion Quarterly, 53(1), 17-40. doi:Article

Harris, L., and Associates (1999).The standards in economics survey. New York: National

Council on Economic Education.

Hatzinikolaou, D. (2010). How to turn a recession into a depression: The role of the media, of the

politicians, and of the political analysts. Modern Economy, 1, 144-148.

Hester, J. B., & Gibson, R. (2007). The agenda-setting function of national versus local: A time-

series analysis for the issue of same-sex marriage. Mass Communication & Society, 10(3),

299–317. doi:Article

Hetherington, M. J. (1996). The media’s role in forming voters’ National Economic ... American

Journal of Political Science, 40(2), 372. doi:Article

Hetherington, M. J., & Rudolph, T. J. (2008). Priming, performance, and the dynamics of

political trust. The Journal of Politics, 70(2), 498-512.

Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and salience. In E.T.

Higgins & A. W. Kruglanski (Eds.), Social Psychology: Handbook of Basic Principles

(pp. 133-168). New York: Guilford Press.


124

Holbrook, R. A., & Hill, T. G. (2005). Agenda-setting and priming in prime time television:

Crime dramas as political cues. Political Communication, 22(3), 277–295. doi:Article

Hopkins, D.J. (2011). Whose economy? Perceptions of national economic performance during

unequal growth. Georgetown. mimeo.

Isidore, C. (2010, October 01). Risk of double-dip recession: Unlikely, but rising. CNNMoney.

Iyengar, S. (1990). The accessibility bias in politics: Television news and public opinion.

International Journal of Public Opinion Research, 2, 1-15.

Iyengar, S. & Kinder, D. (1984). The evening news and presidential evaluations, Journal of

Personality and Social Psychology, 46, 778-87.

Iyengar, S., & Reeves, R. (1997). Do the Media Govern?, Retrieved from www.scopus.com

Iyengar, S., & Simon, A. (1993). News coverage of the gulf crisis and public opinion: A study of

agenda-setting, priming, and framing. Communication Research, 20(3), 365-383.

Iyengar, S., & Simon, A. F. (2011). New perspectives and evidence on political communication

and campaign effects. Annu. Rev. Psychol., 51(1), 149–169. doi:doi:

10.1146/annurev.psych.51.1.149

Jost, J. T., Federico, C. M., & Napier, J. L. (2009). Political ideology: Its structure,

functions, and elective affinities. Annual Review of Psychology, 60, 339–367.

Kahneman, D. (2003). A perspective on judgment and choice: Mapping bounded rationality.

American Psychologist, 58(9), 697–720.

Kahneman, D., Slovic, P., & Tversky, A. (1982). Judgment under uncertainty: Heuristics and

biases. Cambridge, England: Cambridge University Press.

Kahneman, D., & Tversky, A. (1972). On the psychology of prediction. Psychological

Review, 80, 237-251.


125

Kahneman, D., & Tversky, A. (1974). Judgment under uncertainty: Heuristics and Biases.

Science, 185(4157), 1124-1131.

Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk.

Econometrica, 47(2), 263-291.

Kaniss, P. C. (1991). Making local news. Chicago: University of Chicago Press.

Kazoleas, D. (1993), The impact of argumentativeness on resistance to persuasion. Human

Communication Research,20, 118-137.

Kiewiet, D. R. (1983). Macroeconomics and Micropolitics: The Electoral Effects of Economic

Issues, Retrieved from www.scopus.com

Kinder, D.R., Adams, G.D., & Gronke, P.W. (1989). Economic and politics in the 1984

presidential election. American Journal of Political Science, 33, 491-515.

Kinder, D. R., & Kiewiet, D. R. (1979). Economic discontent and political behavior: The role of

personal grievances and collective economic judgments in congressional voting. American


Kinder, D. R., & Kiewiet, D. R. (1981). Sociotropic politics: The American case. British Journal

of Political Science, 11(2), 129–161.

Koehler, J.J. (1996). The base rate fallacy reconsidered: Descriptive, normative, and

methodological challenges. Behavioral and Brain Sciences, 19(1), 1-54.

Kollmeyer, C. J. (2004). Corporate interests: How the news media portray the economy. Social

Problems, 51(3), 432-452.

Kramer, G.H. (1971). Short-term fluctuations in US voting behavior, 1896-1964. The American

Political Science Review, 65(1), 131-143.


126

Kramer, G. H. (1983). The ecological fallacy revisited: Aggregate- versus individual-level

findings on economics and elections, and sociotropic voting. The American Political

Science Review, 77(1), 92–111.

Krause, G. A., & Granato, J. (1998). Fooling some of the public some of the time? A test for

weak rationality with heterogeneous information levels. Public Opinion Quarterly, 62,

135-151.

Krosnick, J.A. (1999). Survey research. Annual Review of Psychology, 50, 537-567.

Krugman, P. (2011, July 21). The lesser depression. The New York Times. Retrieved from

http://www.nytimes.com/2011/07/22/opinion/22krugman.html

Kuklinski, J. H., & N. L. Hurley. (1994). On hearing and interpreting political messages: A

cautionary tale of citizen cue-taking. Journal of Politics, 56, 729-51.

Kurtz, H. (1990). Is the economy suffering from media malady? Washington Post, October 28, p.

H1.

Ladner, M., & Wlezien, C. (2007). Partisan preferences, electoral prospects, and economic

expectations. Comparative Political Studies, 40(5), 571 –596.

doi:10.1177/0010414005285758

Lang, G. E., & Lang, K. (1981). Watergate: An exploration of the agenda-building process. In G.

C. Wilhoit & H. de Bock (Eds.), Mass Communication Review Yearbook. Beverly Hills:

Sage.

Lerner, J. & Tetlock, P.E. (1999). Accounting for the effects of accountability. Psychological

Bulletin, 125, 255-275.

Lewis-Beck, M. S. (1985). Pocketbook Voting in U.S. National Election Studies: Fact or Artifact.

American Journal of Political Science, 29(2), 348. doi:Article


127

Lewis-Beck, M. S. (1988). Economics and Elections: The Major Western Democracies,

Retrieved from www.scopus.com

Limon, M. S., & Kazoleas, D. C. (2004). A comparison of exemplar and statistical evidence in

reducing counter-arguments and responses to a message. Communication Research

Reports, 21(3), 291-298.

MacKuen, M. B., Erikson, R. S., & Stimson, J. A. (1992). Peasants or bankers? The American

electorate and the U.S. economy. The American Political Science Review, 86(3), 597–611.

Martin, P. (2008). The mass media as sentinel: Why bad news about issues is good news for

participation. Political Communication, 25(2), 180–193.

Martin, J., & Powers, M. (1980). Skepticism and the true believer: The effects of case and/or base

rate information on belief and commitment. Paper presented at the meeting of the Western

Psychological Association. May 1980, Honolulu.

Matusitz, J., & Breen, G. M. (2007). Unethical consequences of pack journalism. Global Media

Journal, 6(11), 1-10.

McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public

Opinion Quarterly, 36(2), 176-187.

McCombs, M. E., & Shaw, D. L. (1993). The evolution of agenda-setting research: Twenty-five

years in the marketplace of ideas. Journal of Communication, 43(2), 58–67.

McQuail, D. (1994). Mass communication theory: An introduction. (3rd ed.). London: Sage.

Media figures referring to "Obama bear market" did not refer to "Bush bear market". (2009,

March 10). Media Matters for America, Retrieved from

http://mediamatters.org/research/200903100036


128

Miller, W.E. (1980). Disinterest, Disaffection, and Participation in Presidential Politics. Political

Behavior, 2, 7-32.

Mondak, J. (1993). Source cues and policy approval: The cognitive dynamics of public support

for the Reagan agenda. American Journal of Political Science, 37, 186-212.

Mullainathan, S., & Shleifer, A. (2005). The market for news. The American Economic

Review, 95(4), 1031-1053.

Mutz, D. C. (1992). Mass media and the depoliticization of personal experience. American


Mutz, D. C., & Mondak, J. J. (1997). Dimensions of sociotropic behavior: Group-based

judgments of fairness and well-being. American Journal of Political Science, 41(1), 284–

308.

Nadeau, R., Niemi, R. G., Fan, D. P., & Amato, T. (1999). Elite economic forecasts, economic

news, mass economic judgments, and presidential approval. Journal of Politics, 61, 109-

135.

Neikirk, W. (1991, Fall). Needed: Commitment. Nieman Reports.

Nisbett, R. E., Krantz, D. H., Jepson, C., & Kunda, Z. (1983). The use of statistical heuristics in

everyday inductive reasoning. Psychological Review, 90, 339-363.

Nisbett, R. E., & Ross, L. (1980). Human inference: Strategies and shortcomings of social

judgment. Englewood Cliffs, NJ: Prentice Hall.

Norpoth, H. (1996). Presidents and the prospective voter. Journal of Politics, 58, 776-792.

Number hearing mostly bad economic news soars (2011, August 10). Pew Research Center,

Retrieved from http://www.people-press.org/2011/08/10/number-hearing-mostly-bad-

economic-news-soars/


129

Pan, Z., & Kosicki, G. M. (1993). Framing analysis: An approach to news discrouse. Political

Communication, 10, 55-75.

Patterson, T. E. (1997). The news media: An effective political actor? Political Communication,

14(4), 445-455.

Patterson, T.E. (1993). Out of order. New York: Vintage Books.

Paul, R. (2008, September 25). The creation of the second Great Depression. Retrieved from

http://www.lewrockwell.com/paul/paul479.html

Ratner, R. K., & Miller, D. T. (2001). The norm of self-interest and its effects on social action.

Journal of Personality and Social Psychology. 81, 5-16

Reeves, A., & Gimpel, J. G. (2001). Ecologies of unease: Geographic context and national

economic evaluations. Political Behavior, doi: 10.1007/s11109-011-9167-8

Roos, M. W. M. (2007). Nonexpert beliefs about the macroeconomic consequences of economic

and noneconomic events. Public Choice, 132, 291-304.

Selfish and Prospective: Theory and Evidence of Pocketbook Voting (2008). 1–30.

Severin, W. & Tankard, J.W. (I992). Communication Theories: Origins, Methods, and Uses in

the Mass Media (3rd ed.), New York, Longman.

Shah, D. V., Watts, M. D., Domke, D., Fan, D. P., & Fibison, M. (1999). News coverage,

economic cues, and the Public’s Presidential Preferences, 1984–1996. The Journal of

Politics, 61(4), 914–943. doi:10.2307/2647548

Shen, F. (2009). An economic theory of political communication effects: How the economy

conditions political learning. Communication Theory, 19(4), 374–396.

Shiller, R. J. (1995). Conversation, information, and herd behavior. American Economic Review,

85(2), 181–185.


130

Sniderman, P. M., R. A. Brody, & P. E. Tetlock (1991). Reasoning and Choice: Explorations in

Political Psychology. New York: Cambridge University Press.

Soroka, S. N. (2006). Good news and bad news: Asymmetric responses to economic

information. The Journal of Politics, 68(2), 372-385.

Soroka, S. N. (2003). Media, public opinion, and foreign policy. Harvard International Journal

of Press/Politics, 8(1), 27-48.

Southwell, P. L. (1996, April). Economic salience and differential abstention in presidential

elections. Retrieved from

http://search.ebscohost.com/login.aspx?direct=true&db=21h&AN=9604175380&site=eho

st-live

Starr, M. A. (2010). Consumption, sentiment, and economic news. Economic Inquiry, doi:

10.1111/j.1465-7295.2010.00346.x

Stevenson, R., Gonzenbach, W., and Prabu, D. (1991). Economic recession and the news.

Presented to the Communication Theory and Methodology Division, Association for

Education in Journalism and Mass Communication, meeting in Boston, MA, August,

1991.

Taylor, S. E., & Fiske, S. T. (1978). Salience, attention, and attribution: Top of the head

phenomena. In L. Berkowitz (Ed.), Advances in experimental social psychology, . New

York, NY: Academic Press.

Tims, A. R., Fan, D. P., & Freeman, J. R. (1989). The cultivation of consumer confidence: A

longitudinal analysis of news media influence on consumer sentiment. Advances in

Consumer Research, 16, 758-770.


131

Tseng, N. H. (2010, August 06). Did the recession actually end? CNN Money, Retrieved from

http://money.cnn.com/2010/08/06/news/economy/NBER_recession.fortune/index.htm

Tversky, B., & Hemenway, K. (1984). Objects, parts, and categories. Journal of Experimental

Psychology: General, 113, 169-193.

Tversky, A., & Kahneman, D. (1973). Availability: A heuristic for judging frequency and

probability. Cognitive Psychology, 5, 207-232.

Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases.

Science, 185, 1124-1131.

Tversky, A., & Kahneman, D. (1982). Judgment under uncertainty: Heuristics and biases.

In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics

and biases (pp. 3-22). Cambridge, England: Cambridge University Press.

Walstad, W. B. (1997). The effect of economic knowledge on public opinion of economic issues.

Journal of Economic Education, 28(3), 195–205.

Walstad, W. B., & Rebeck, K. (2002). Assessing the economic knowledge and economic

opinions of adults. The Quarterly Review of Economics and Finance, 42(5), 921–935.

doi:10.1016/S1062-9769(01)00120-X

Wessel, D. (2010, February 27). Academics on what caused the financial crisis. The Wall Street

Journal. Retrieved from http://blogs.wsj.com/economics/2010/02/27/academics-on-what-

caused-the-financial-crisis/

When did it end? (2010, April 15). The Economist, Retrieved from

http://www.economist.com/node/15911334


132

Wiseman, P. (2011, October 23). Economists see no 2nd recession, no real recovery. Associated

Press, Retrieved from http://www.msnbc.msn.com/id/44234721/ns/business/t/economists-

see-no-nd-recession-no-real-recovery/

Worstall, T. (2011, August 11). When journalists meet numbers: Not a pretty sight. Forbes,

Retrieved from http://www.forbes.com/sites/timworstall/2011/08/27/when-journalists-

meet-numbers-not-a-pretty-sight/

Wu, H. D., & Day, A. (2005). The dominance of bearish news? Comparing coverage against the

state of the economy. Paper presented at the 2005 convention of the Association for

Education in Journalism & Mass Communication, San Antonio, TX.

Wu, H. D., Stevenson, R. L., Chen, H., & Güner, Z. N. (2002). The conditioned impact of

recession news: A time-‐series analysis of economic communication in the United States,

1987–1996. International Journal of Public Opinion Research, 14(1), 19 –36.

doi:10.1093/ijpor/14.1.19

Zillmann, D. (1999). Exemplification theory: Judging the whole by some of its parts. Media

Psychology, 1(1), 69. doi:Article

Zillmann, D. (2002). Exemplification theory of media influence. Media Effects: Advances in

Theory and Research, 2nd ed. Mahwah, NJ: Lawrence Erlbaum Associates.

Zillmann, D., & Brosius, H. (2000). Exemplification in communication: The influence of case

reports on the perception of issues. Mahwah, NJ: Lawrence Erlbaum Associates.

Zillmann, D., Callison, C., & Gibson, R. (2009). Quantitative media literacy: Individual

differences in dealing with numbers in the news. Media Psychology, 12(4), 394–416.

doi:10.1080/15213260903287275


133

Zillmann, D., & Gan, S. (1996). Effects of threatening images in news programs on the

perception of risk to others and self. Media Psychology, 8, 288–305, 317–318.

Zillmann, D., Gibson, R., Sundar, S. S., & Perkins, J. W. (1996). Effects of exemplification in

news reports on the perception of social issues. Journalism & Mass Communication

Quarterly, 73(2), 427-444.

Zillmann, D., Perkins, J. W., & Sundar, S. S. (1992). Impression-formation effects of printed

news varying in descriptive precision and exemplifications. Media Psychology, 4, 168-

185.


134

APPENDICES A

Positive Base-Rate: Exemplar Representative Economic News Articles

Consumer confidence points to economic growth By John Gibbs There are signs of growth in the economy, as the Consumer Confidence Index released Friday for the state and local area seems to confirm that forecast by area economists.

Consumer confidence has fluctuated since the end of January as U.S. job growth numbers have seemingly influenced personal finances and spending. Almost 75 percent of the 5,000 consumers who were surveyed stated they felt more confident about the stabilization of their local economy and the direction it was headed.

Small Businesses (positive)

While many small business owners in the area have a number of specific concerns they have expressed constructive assessments about the future of their own businesses.

Anne Billingsley, a local business owner, said that number is no surprise. Billingsley has become more optimistic about her public relations business, Imagine Now.

“I’ve begun hiring full-time employees and no longer have to depend on freelance writers, graphic artists and administrative help. I am getting more long-term projects, which means my income is more predictable. I don't think there is any question that the economy is getting better," Billingsley said.

Employment (positive)

Monthly employment figures continue to inch up and down after a few years of turbulence and uncertainty. The report showed experienced workers, who are defined as workers ages 30 to 54 years old, tended to fare better in most fields, while recent college graduates continue to look for jobs.

When Linh Sasser graduated two years ago with a degree in advertising, she had big ideas about where her career might take her. Although she now admits she entered “a pretty dreadful job market,” she thought she had a strong resume, but was not getting interviews. She had almost given up when six months ago she was hired by an ad agency.

“I’m encouraged by what I have seen recently. As the national economy continues to improve, employers are returning to campuses in greater numbers, looking for the next generation of employees, Sasser said. “Fortunately, a lot of my friends are beginning to


135

find jobs, and many are finding jobs in fields they’re excited about. Everyone seems to be really hopeful about the economy.”

Home Sales (positive)

The housing market is another potential sign of economic strength. Consumer confidence in the housing industry can give a boost to or limit local residential and commercial construction projects.

Brett Underwood, the owner of a construction management company, said he has plans to hire more people in the upcoming months. He said his business has almost twice the amount of work it can handle because the building industry is now recovering from the recession.

"Things are looking up in the construction business," he said. “I no longer have the mixed feelings and emotions that I had just a few years ago. I definitely feel the economy has improved in our area.”

Personal Finances (negative)

Consumers' expectations for their personal financial situations are typically tied to broad economic indicators. Americans are optimistic about their own wealth when the economy is doing well, and pessimistic in times of economic downturn.

Anthony Lewis, president of R&T Bank and a board member of the local Chamber of Commerce, stated his bank and other businesses are seeing the impact of an uncertain economic atmosphere. Personal loans at his bank are down and they haven’t been able to extend larger lines of credit to local businesses.

“Consumers, whether looking back, forward, or when taking stock of today, are more cautious than they’ve been in nearly five years,” Lewis said. “The decrease in demand for loans from small business as well as from larger businesses makes me believe the economy is still in a state of fluctuation.”


136










“I’m encouraged by what I have seen recently. As the national economy continues to improve, employers are returning to campuses in greater numbers, looking for the next generation of employees, Sasser said. “Fortunately, a lot of my friends are beginning to find jobs, and many are finding jobs in fields they’re excited about. Everyone seems to be really hopeful about the economy.”


137

Home Sales (negative)


Brett Underwood, the owner of a construction management company, said he has no plans to hire more people in the upcoming months. He said his business has only about half of the amount of work it can handle because the building industry is still hurting from the recession.

"Things are looking down in the construction business," he said. “I still have the mixed feelings and emotions that I had just a few years ago. I definitely feel the economy has not improved in our area.”

Personal Finances (positive)


Anthony Lewis, president of R&T Bank and a board member of the local Chamber of Commerce, stated his bank and other businesses are seeing the impact of a steady economic atmosphere. Personal loans at his bank are up and they’ve begun to extend larger lines of credit to local businesses.

“Consumers, whether looking back, forward, or when taking stock of today, are more upbeat than they’ve been in nearly five years,” Lewis said. “The increase in demand for loans from small business as well as from larger businesses makes me believe the economy is moving in the right direction.”


138







Employment (negative)


When Linh Sasser graduated two years ago with a degree in advertising, she had big ideas about where her career might take her. Although she now admits she entered “a pretty dreadful job market,” she thought she had a strong resume, but was not getting interviews. Six months ago she gave up looking for an ad agency job.

“I’m discouraged by what I have seen recently. As the national economy continues to stagnate, employers are returning to campuses in fewer numbers, often overlooking the next generation of employees,” Sasser said. “Unfortunately, many of my friends are still looking for jobs, and those who are finding jobs aren’t finding them in fields they’re excited about. Everyone seems to be really down about the economy.”


139










140



Small Businesses (negative)


Anne Billingsley, a local business owner, said that number is no surprise. Billingsley has become more hesitant about her public relations business, Imagine Now.

“I’ve begun relying less on full-time employees and have to depend on freelance writers, graphic artists and administrative help. I am getting more short-term projects, which means my income is less predictable. I don't think there is any question that the economy is still hurting," Billingsley said.






141










142

Positive Base-Rate: Exemplar Non-representative Economic News Articles












143










144












145










146












147










148












149










150

Positive Base-Rate: Exemplar Mixed Economic News Articles Consumer confidence points to economic growth By John Gibbs There are signs of growth in the economy, as the Consumer Confidence Index released Friday for the state and local area seems to confirm that forecast by area economists.











151










152












153










154












155










156












157










158

Negative Base-Rate: Exemplar Representative Economic News Articles

Consumer confidence points to a downturn in the economy By John Gibbs There are signs of decline in the economy, as the Consumer Confidence Index released Friday for the state and local area seems to confirm that forecast by area economists.

Consumer confidence has fluctuated since the end of January as U.S. job growth numbers have seemingly influenced personal finances and spending. Almost 75 percent of the 5,000 consumers who were surveyed stated they felt less confident about the stabilization of their local economy and the direction it was headed.










159










160












161










162












163










164












165










166

Negative Base-Rate: Exemplar Non-representative Economic News Articles












167










168












169










170












171










172












173










174

Negative Base-Rate: Exemplar Mixed Economic News Articles












175










176



Small Businesses(negative)









177










178












179










180












181










182

APPENDICES B

Measures

Intro

Hello, this is an online research project designed to help us understand individual’s perceptions of the economy. To better understand this we ask for your participation. The online questionnaire will take approximately 20 minutes of your time. Your participation is completely voluntary, the information shared with us will remain confidential, and your responses will only be used for research purposes. The Sample Network, an online marketing company, has been hired on behalf of the researchers to help us collect this data. Your email address and answers to the survey will be kept completely separated. The persons responsible for this project are Dr. Trent Seltzer (faculty member in the College of Mass Communications) and Austin Sims (doctoral student in the College of Mass Communications). Please feel free to contact, [email protected] if you have any questions or if you wish to receive a copy of the results of this study. Thank you so much for your willingness to participate in our research study. This study has been approved by the Protection of Human Subjects Committee at Texas Tech University. Texas Tech Human Resources Protection Program (HRPP) Drane Hall, Room 147 Box 41075 Lubbock, TX 79409-1075 (806) 742-2064

Study Info

You will now be asked to read a news article that has been taken directly from a local newspaper. Following the article you will be asked a series of questions regarding a range of topics. Please read each question carefully and answer each of them to the best of your ability.

ECONOMIC LITERACY

EL-Intro We are studying individual's understanding and knowledge of economic issues. For the next set of questions please answer each one to the best of your ability.

EL1 When a person rents an apartment, who benefits from the transaction?

m Only the person renting the apartment

m Only the landlord

m Both the person renting the apartment and the landlord

m Don't know

EL2 In the United States, who determines what goods and services should be produced?

m Producers and government

m Consumers and government

m Producers, consumers, and government

m Don't know


183

EL3 Since the resources used in the production of goods and services are limited, society must:

m Reduce their use of resources

m Try to obtain resources

m Make choices about how to use resources

m Don't know

EL4 An increase from 5percent to 8percent in the interest rates charged by banks would most likely encourage:

m Businesses to invest

m People to purchase housing

m People to save money

m Don't know

EL5 The stock market is an example of an institution within our economy that exists to help people achieve their economic goals. The existence of this institution:

m Results in an increase in the price of stocks

m Brings people who want to buy stocks together with those who want to sell stocks

m Helps predict stock earnings

m Don't know

EL6 Which of the following are most likely to be helped by inflation?

m People living on a fixed income

m Banks that loaned money at a fixed rate of interest

m People who borrowed money at a fixed rate of interest

m Don't know


184

EL7 Assume that one U.S. dollar is equivalent to 10 Japanese yen. If the value of the dollar appreciates so that one dollar is equivalent to 15 Japanese yen, which of the following occurs?

m The price of all goods produced in Japan but sold in the U.S. will increase

m The Japanese people will import more goods made in the U.S.

m American traveling to Japan find their dollars buy more goods and services

m Don't know

EL8 If your city government sets a maximum amount landlords can charge in rent, what is the most likely result?

m There will be more apartments available than people want to rent

m There will be fewer apartments available than people want to rent

m The number of apartments available will be equal to the number of people that want to rent

m Don't know

EL9 Which of the following methods for reducing pollution would most economists support?

m Increase regulation on all industries that create pollution in the production process or all products that pollute

m Eliminate all pollution since clean air and water are so important

m Reduce pollution until the additional cost of the further reduction is greater than the additional benefit

m Don't know

m

EL10 If the real gross domestic product of the United States has increased, which of the following has also definitely increased?

m The amount of resources used to produce final goods and services

m The prices of final goods and services produced

m The amount of final goods and services produced

m Don't know


185

EL11 When the federal government's expenditures for a year are greater than its revenue for that year, the difference is known as:

m The national debt

m A budget deficit

m A budget surplus

m Don't know

EL12 On average, how much more do adults who are college graduates earn per year than adults who are high school graduates only?

m About 10percent more



m Don't know

EL13 Which of the following types of investments has the greatest risk of losing value due to inflation?

m Buying stocks in the stock market

m Investing money market mutual funds

m Keeping your savings as cash hidden in a mattress or in a piggy bank

m Don't know

m

EL14 Some people prefer to buy mutual funds rather than stocks in a few individual companies because generally mutual funds:

m Guarantee a steadier income than individual stocks

m Provide a higher rate or return than individual stocks

m Provide more diversification than individual stocks

m Don't know


186

ARITHMETIC APTITUDE

AA-Intro We are studying how people make sense of the complicated and often confusing facts and figures that are reported in the news media. Could you please give us your estimate or "educated guess" for the set of examples that we have selected from news magazines? PLEASE DO NOT USE A CALCULATOR.

AA1 Example 1: The freeze destroyed 5/16 of the orange crop. Is this less or more than a quarter?

m Less than a quarter

m More than a quarter

m Same as a quarter

AA2 Example 2: In this small community high school, only seven of the 20 seniors passed the exit exam. What percentage of passing students are they talking about?

m 7percent

m 20percent

m 27percent

m 35percent

m 63percent

AA3 Example 3: Doug McKinney got by on a $7,500 income. He is still in disbelief about the promised 20percent pay raise. How much will he make in the future?

m $7,750

m $8,500

m $8,750

m $9,000

m $9,500


187

AA4 Example 4: The storm (a tornado) damaged 80percent of all buildings in this small community. What portion of the buildings was damaged?

m 3/4

m 4/5

m 8/12

m 10/18

m 20/30

AA5 Example 5: Last Sunday, 30percent of the 1200 patients at Merion Memorial had visitors. How many hospital patients enjoyed visitations on that day?

m 180

m 300

m 360

m 420

m 680

AA6 Example 6: Two of the eight underweight piglets survived the first two weeks. What is the percentage of surviving piglets?

m 10percent

m 12percent

m 18percent

m 20percent

m 25percent


188

AA7 Example 7: This poison proved to be surprisingly effective. Of the 20,000 brown rats tested, 18,000 died within 24 hours. What is the percentage of dying rats?

m 90percent

m 80percent

m 70percent

m 60percent

m 50percent

AA8 Example 8: Only 10percent of those exposed to this virus contract the disease, and only 5percent of those infected die from it. What does this mean for the population of people? Of a million people, how many will die?

m 5

m 50

m 500

m 5,000

m 50,000

AA9 Example 9: According to research conducted by the CDC, the risk of developing this disease from the bite is .00003. Of a million bite victims, how many are at risk of illness?

m 3

m 30

m 300

m 3,000

m 30,000


189

AA10 Example 10: Of the 2.5 million annual deaths (in the U.S.), 11,212 are the result of drug overdose. For anyone overdosing on drugs, what is the likelihood of dying from it (in terms of a probability number between 0 and 1)?

m .045

m .0045

m .00045

m .000045

m .0000045

ECONOMIC PERCEPTIONS

P-Intro The next set of questions are designed to help us better understand your perceptions regarding the overall general U.S. economy.

P1 We are interested in your perceptions regarding the overall general state of the U.S. economy at this moment. Would you say the economy is getting better, getting worse, or about the same?

m Getting worse 1

m 2

m 3

m About the same 4

m 5

m 6

m Getting better 7

P1.a Relating back to the previous question, how confident are you in your answer that things will get better, get worse, or stay the same?

m Not very confident 1

m 2

m 3

m Neutral 4

m 5

m 6

m Very Confident 7


190

P2 We are interested in your perceptions regarding the state of the economy at this moment. Do you think the overall state of the economy is:

m Weak 1

m 2

m 3

m Neutral 4

m 5

m 6

m Strong 7

P2.a Relating back to the previous question, how confident are you in your answer in the overall health of the economy?


m 2

m 3

m Neutral 4

m 5

m 6

m Very Confident 7

m

P3 Now turning to business conditions --do you think overall business conditions are:

m Weak 1

m 2

m 3

m Neutral 4

m 5

m 6

m Strong 7


191

P3.a Relating back to the previous question, how confident are you in your answer regarding business conditions?


m 2

m 3

m Neutral 4

m 5

m 6

m Very Confident 7

P4 With regards to the housing market -- do you think the overall housing market is:

m Weak 1

m 2

m 3

m Neutral 4

m 5

m 6

m Strong 7

P4.a Relating back to the previous question, how confident are you in your answer regarding the housing market?


m 2

m 3

m Neutral 4

m 5

m 6

m Very Confident 7


192

P5 Now turning to unemployment -- do you think overall unemployment is:

m Weak 1

m 2

m 3

m Neutral 4

m 5

m 6

m Strong 7

P5.a Relating back to the previous question, how confident are you in your answer regarding unemployment?


m 2

m 3

m Neutral 4

m 5

m 6

m Very Confident 7

P6 With regards to personal finance -- do you think overall personal financial conditions are:

m Weak 1

m 2

m 3

m Neutral 4

m 5

m 6

m Strong 7


193

P6.a Relating back to the previous question, how confident are you in your answer regarding personal financial conditions?


m 2

m 3

m Neutral 4

m 5

m 6

m Very Confident 7

ECONOMIC RECALL

R-Intro Reports about economic issues have become an important component of the daily news. We are studying how people make sense of these reports and also how important and informative they find the various reported economic issues. Please answer the following questions based upon what you remember from the article you read earlier.

R1 Regarding the Consumer Confidence Index, what percent of the public had...

______ positive perceptions of the economy?

R2 Of the 5,000 participants in the study, how many would you estimate were more...

______ optimistic about the economy?

R3 According to the Consumer Confidence Index, would you say those surveyed feel the economy is worse off, about the same, or better off now?

m Worse off 1

m 2

m 3

m About the same 4

m 5

m 6

m Better off 7


194

R4 Regarding the Consumer Confidence Index, what percent of the public had...

______ negative perceptions of the economy?

R5 Of the 5,000 participants in the study, how many would you estimate were more...

______ pessimistic about the economy?

DEMOGRAPHICS

D1 How old are you?

m 18-25

m 26-34

m 35-54

m 55-64

m 65 or over

D2 What is your gender?

m Male

m Female

D3 What is your race?

m White/Caucasian

m African American

m Hispanic

m Asian

m Native American

m Pacific Islander

m Other


195

D4 What is the highest level of education you have completed?

m Less than High School

m High School / GED

m Some College

m 2-year College Degree

m 4-year College Degree

m Some graduate

m Graduate Degree

D5 In which state do you currently reside?

m Alabama

m Alaska

m Arizona

m Arkansas

m California

m Colorado

m Connecticut

m Delaware

m District of Columbia

m Florida

m Georgia

m Hawaii

m Idaho

m Illinois

m Indiana

m Iowa

m Kansas


196

m Kentucky

m Louisiana

m Maine

m Maryland

m Massachusetts

m Michigan

m Minnesota

m Mississippi

m Missouri

m Montana

m Nebraska

m Nevada

m New Hampshire

m New Jersey

m New Mexico

m New York

m North Carolina

m North Dakota

m Ohio

m Oklahoma

m Oregon

m Pennsylvania

m Puerto Rico

m Rhode Island

m South Carolina

m South Dakota

m Tennessee


197

m Texas

m Utah

m Vermont

m Virginia

m Washington

m West Virginia

m Wisconsin

m Wyoming

m I do not reside in the United States

D6 Which political party do you feel you most identify with?

m Democratic

m Republican

m Libertarian

m Independent

m Other (please specify) ____________________

D7 Please indicate on the scale below where you believe your political ideology lies.

m Very liberal

m Liberal

m Left of center

m Moderate

m Right of center

m Conservative

m Very conservative


198

D8 Please indicate on the scale below how where you believe you are fiscally.

m Very liberal

m Liberal

m Left of center

m Moderate

m Right of center

m Conservative

m Very conservative

D9 Please indicate on the scale below how where you believe you are socially.

m Very liberal

m Liberal

m Left of center

m Moderate

m Right of center

m Conservative

m Very conservative

D10 We value your interest in this research project and would like to know your opinions and input regarding the study. Please let us know what you liked, what you didn't like, or anything else you feel is important.

End

Thank you for participating in this research study. The goal of this project is to better understand how individuals process numerical data versus narrative information. Economic news stories are often full of numbers and percentages, as well as statements and quotes from a variety of sources. Understanding what news factors influence individual's perceptions of the economy helps us better understand how the media potentially affects the overall economy. Once again thank you for your participation and please feel free to contact, [email protected] if you have any questions or if you wish to receive a copy of the results of this study.

Copyright 2012, Austin Sims

Documents