A review of research methods

A review of research methods

COMM 420.8

Fall 2007

Nan Yu

Quantitative research methods

SurveysAdvantages Disadvantages

Surveys show correlational evidence

(i.e., patterns of variables in a large population).

Survey instruments rely solely on self-reported

measures. Low reliability.

Surveys are very good at discerning attitudes, beliefs

and perceptions

They are not good for studying behavior,

emotions, etc.

Have high external validity(real-world situation, large

samples)

They can hardly prove causation.

Random error is high.

Survey design

Cross-sectional studies observation at one point of time

Longitudinal studies observation over time

establish patterns and shifts over time

Types Longitudinal Study

Trend Cohort Panel

different samplesSamples are

subgroups from a same population.

same samples

A cohort is a group of people who share a common characteristic or experience within a defined time period. (e.g. class of 1997, people that were born between 1990 to 1995. )

Sample size might be different over time (people moved, died…).

Require probability sampling from the same population, otherwise you can not compare the data collected at different times.

Types of surveys

Watt, J. H., & van den Berg, S. A. (2002). Research methods for communication science. Needham Heights, MA: Allyn and Bacon, p.278, p. 290

Sampling error and sample size

Sample size

The greater the sample size, the less the sampling error.

Problematic questions

What are problems with the following questions? “What is your income?”

“Do you jog regularly?”

“ Does your college offer master or doctoral programs?”

What is your total family’s household income before taxes during last year?

How many times do you job per week?1 2 3 4 5 6 7

Double-barreled questions

Techniques of asking sensitive questions

Goal: to increase truthful answers to sensitive questions:

“Foot in the door”

The researcher builds some trust and respondents are more comfortable towards the end with the researcher and the procedures.

Place sensitive questions (e.g. demographic questions) at the end.

ExperimentsA true experiment has 4 characteristics:

Manipulated independent variables

Random assignment

Experimental conditions (other than manipulated IVs) are held constant or as similar as possible.

Dependant variabless are systematically measured.

Internal vs. External Validity of Experimental Studies

When a true experiment (with full control by the researcher) is carried out properly, the study has high internal validity.

Changes in the DV were probably caused by (not merely related to or correlated with) the treatment.

However, experiments have lower external validity. What is the reason for this?

Factorial design

DV:Liking toward the picture

A 2X2 factorial design4 conditions

Random Assignment (cont.)

Why we need random assignment?

Because we want to make sure that

the differences between groups after

they were exposed to different levels of

the IV are due to the IV effect on the

DV, and not to the differences between

groups.

Random Assignment (cont.)

Treatment A Treatment B

experimenter

• Flipping a coin• Pull names from a thoroughly shaken box.• Random numbers table

Participants come in random orders, and experimenter randomly assign them to treatment A or treatment B.

Steps in a true (full) experiment study

Formulate a hypothesis or a research question.

Manipulate IV(s) (stimulus).

Pretest the stimulus

Recruit participants

Present to the participants a cover story

Randomly assign participants to a treatment group or to the control group.

Introduce treatment or intervention

Measure the dependent variable (s)

Debriefing

Main effect and interaction

Content Analysis

Systematic study of communication contents in an objective and quantitative manner.

The researcher uses objective and systematic counting and recording procedures to produce a quantitative description of the symbolic content in a text.

Why do we need content analysis?

Applications of content analysis

Describing communication content

Testing hypothesis of message characteristics

Assessing the image of particular groups in society.

Comparing media content to “real world”

Establishing starting point for media effects research, (e.g. cultivation and agenda setting)

Content Analysis

The content refers to words, meanings, pictures, symbols, ideas, themes, any message that can be communicated, etc.

The text refers to anything written, visual, or spoken that serves as a medium for communication (e.g., books, newspapers and newspaper articles, advertisements, speeches, official documents, movies, musical lyrics, photographs, etc.)

Coding and measurement

Measurement (coding) in content analysis uses structured observation: systematic, careful observation based on written rules.

The rules explain how to categorize and classify the observations (i.e., units).

Written rules are important in content analysis, as they improve the reliability and make the replication possible.

Intercoder reliability

In order to improve the reliability and eliminate differences in judgments, researchers train and use more than one coder.

If they agree most of the time on what unit should be placed in which category, the reliability is high.

This type of reliability is called intercoder or interrater reliability.

Qualitative Research Methods

When are qualitative methods most useful?

When the research objective is:To interpret, illuminate, illustrateTo understand how or why To describe previously unstudied processes or

situationsTo get in-depth, detailed informationTo understand contextTo learn about subjects who are few or hard to

reachTo capture experience

http://www.eotu.uiuc.edu/live/index.html

University of Illinois, Champaign-Urbana

Research Questions

Research questions that seek to describe lived experience, explain meaning or offer rich, thick description should be answered using qualitative methods.

Sampling

Generally always purposive Definition: A goal of reaching a pre-defined group

Different types of purposive sampling:Snowball, extreme case, stratified and opportunistic

No need for random sampling: the goal of qualitative research is not to generalize but to explainExample: We need people who have attended a Penn

State football game in order to study the experiences at a Penn State football game

Two types of qualitative methods

Focus groups (p.206)Researcher (facilitator, moderator) is part of the

communication with a selected group of participants, but in control of questions being asked and the environment.

In-depth interviewOne-on-one interviews,Researcher observe the communication verbally and

nonverballyMore freedom in constructing questions

Advantages and limitations of focus group

Richness in the dataExplanatory and descriptive

Inexpensive and quick

Quality of the research largely depends on the facilitator’s ability to direct the discussion.

Some participants may monopolizing the talk.Low external validity

Advantages and limitations In-depth interviews

Story-tellingRichness of the data

Take long time to completePersonal perspectivesLimited ability to generalize

Experiment SurveyContent Analysis

Focus Group

In-depth interview

Rationale

Sampling

Instrument

Study procedure

Analysis /conclusion

Overview

Control

In-depth interview focus group survey experiment

Less rigorous Most rigorous