Does online consumer generated media influence attitudes towards brands? A study of the credibility of recipe blogs and their effect on consumers’ attitudes towards food brands in Turkey This dissertation is submitted in part of the fulfilment of the MA Interactive Marketing in Bournemouth University I, Sevil Özer, declare that this dissertation is the result of my own independent investigation and that all sources are duly acknowledged ……………………….. Sevil Özer Supervisor: (September, 2005) Mr. Mike Molesworth
106
Embed
Does online consumer generated media influence attitudes towards brands?
This is my full copy of my dissertation reflecting an early review on social media and its impact on brand perception. Happy to share as after all these years the theory is still true, in fact the impact of social media on brand perception is now more prominent.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Does online consumer generated media
influence attitudes towards brands?
A study of the credibility of recipe blogs and their effect on
consumers’ attitudes towards food brands in Turkey
This dissertation is submitted in part of the fulfilment of the MA
Interactive Marketing in Bournemouth University
I, Sevil Özer, declare that this dissertation is the result of my own independent
investigation and that all sources are duly acknowledged
1.1. BACKGROUND OF THE STUDY.................................................................................................................. 5 1.2. STUDY FOCUS: WHY BLOGS? ................................................................................................................ 6 1.3. PURPOSE OF THE STUDY ........................................................................................................................ 7 1.4. STATE OF THE BLOGOSPHERE ................................................................................................................ 8 1.5. MORE ON GENERATING CONTENT .......................................................................................................... 9 1.6. FOCUS OF THE STUDY ........................................................................................................................... 10
CHAPTER 2 - LITERATURE REVIEW ....................................................................................................... 12
2.1 OVERVIEW .............................................................................................................................................. 12 THE MODEL FOR PROCESSING INFORMAL BRAND INFORMATION ONLINE: FRAMEWORK FOR BLOG’S
2.2.1. Persuasion and the Online Consumer ...................................................................................... 14 General View: Persuasion .................................................................................................................................. 14 Low-Involvement Processing Model of Heath ................................................................................................. 16 Product Categories and Involvement ............................................................................................................... 16
2.2.2. Blog’s Effects on Brand Attitude Change .................................................................................. 17 Attitude towards brand ........................................................................................................................................ 17 Online Word of Mouth Communication ............................................................................................................ 18
2.2.3. Credibility of a Blog ...................................................................................................................... 21 General View: What is Credibility?.................................................................................................................... 21 A General View on Web Credibility ................................................................................................................... 23 Blog Credibility ..................................................................................................................................................... 24
PATH MODEL OF PREDICTORS OF BLOG CREDIBILITY ................................................................................... 24 The Blog Author’s Credibility ......................................................................................................................... 25 Visitor Related Factors ................................................................................................................................... 26 Design and Content Related Factors ........................................................................................................... 28
CHAPTER 3- RESEARCH METHODOLOGY ........................................................................................... 30
3.1. OBJECTIVES AND MAIN RESEARCH QUESTIONS .................................................................................... 30 3.2. OVERVIEW OF THE RESEARCH PROCESS ............................................................................................ 30 3.3. THE SIMPLIFIED FRAMEWORK FOR BLOG’S EFFECTS ON BRAND ATTITUDE CHANGE ........................ 31
Hypothesis to be tested ......................................................................................................................... 31 3.4. PRIMARY RESEARCH .............................................................................................................................. 32
3.4.1. The methodology and methods adopted .................................................................................. 32 3.5. EVALUATION OF RESEARCH DESIGN ..................................................................................................... 33
3.5.1. Sample definition and rationale .................................................................................................. 33 3.5.2. Method of sampling and rationale ............................................................................................. 34 3.5.3. Method of data collection and rationale .................................................................................... 35 3.5.4. The Questionnaire Design .......................................................................................................... 36 3.5.5. Research Process ....................................................................................................................... 39
3.6. LIMITATIONS FOR THE STUDY ................................................................................................................ 40 3.6.1 Methods used to achieve reliability and validity in the findings. ............................................. 42 3.6.2 Post Research Findings ............................................................................................................... 43
3.7. SECONDARY RESEARCH ....................................................................................................................... 43
4.1 OVERVIEW OF THE STUDY ...................................................................................................................... 44 4.1.1. Descriptive Statistics ................................................................................................................... 44
Motivation * Have you visited this site before? Crosstabulation .............................................................. 45 Site and Visitors ............................................................................................................................................... 46 Previous Purchase Reports ........................................................................................................................... 48 Overall Perceptions of Credibility .................................................................................................................. 48
Visitor related factors .......................................................................................................................................... 51 Source reliance ................................................................................................................................................ 51 Internet experience ......................................................................................................................................... 51 Convenience .................................................................................................................................................... 52 Site Familiarity ................................................................................................................................................. 52
Design and Content Related Factors ............................................................................................................... 53 Usability and Design ....................................................................................................................................... 53 Content ............................................................................................................................................................. 53 Additional Factor - Interactivity ...................................................................................................................... 53
Author Related Factors ....................................................................................................................................... 53 4.2.3 Measurement Results for Blog Credibility Factor Analysis ..................................................... 54
Primary Results .................................................................................................................................................... 54 Visitor Related Factors and Blog Credibility ................................................................................................ 54 Experience level of the visitor and Blog Credibility .................................................................................... 55 Convenience Perception of Visitor and Blog Credibility ............................................................................ 55 Site Familiarity and Blog Credibility .............................................................................................................. 56 Site Design and Content and Blog Credibility ............................................................................................. 56 Interactivity Factor and Blog Credibility ....................................................................................................... 56 Perception of Author Characteristics and Blog Credibility ........................................................................ 56
4.2.4. Conclusion .................................................................................................................................... 57 Limitations for factor analysis ............................................................................................................................ 58
Blog Credibility: Definition of “Is the Source Credible?” ............................................................................ 59 Motivation: Browse or Search ....................................................................................................................... 60
5.1. CONCLUSION ......................................................................................................................................... 66 5.2. RECOMMENDATIONS FOR THE INDUSTRY.............................................................................................. 66 5.3. RECOMMENDATIONS FOR FUTURE RESEARCH ...................................................................................... 67
APPENDIX 1 – THE QUESTIONNAIRE ............................................................................................................ 71 APPENDIX 2 – THE SCREENSHOTS OF ONLINE QUESTIONNAIRE ........................................................ 72 APPENDIX 3 – SOME SCREENSHOTS OF THE BLOGS WHO SUPPORTED THE SURVEY STUDY 74 APPENDİX 4 – BLOG CREDİBİLİTY FACTOR ANALYSİS .................................................................................. 77 APPENDİX 5 – BLOG CREDİBİLİTY FACTOR ANALYSİS .................................................................................. 81 APPENDIX 6 – BLOG CREDİBİLİTY FACTOR ANALYSİS CORRELATİON MATRİXES ...................................... 1 APPENDİX 7 – BLOG CREDİBİLİTY FACTOR ANALYSİS CONTRİBUTİON FROM COMPONENTS ....................... 1 APPENDİX 8 – HYPOTHESİS TEST SPSS OUTPUTS ...................................................................................... 1
CASE 1 – Credible Blog * Search (High-involvement) ........................................................................ 1 CASE 2 – Credible Blog * Browse (Low-involvement) ........................................................................ 3 CASE 3 – Non-Credible Blog * Search (High-involvement)................................................................ 5 CASE 4 – Non-Credible Blog * Browse (Low-involvement) ................................................................ 7
– 4 –
LIST OF TABLES
Chart 1- Framework for Blog’s Effects on Brand Attitude Change, Pg 11
Table 1 - Examples of factors influencing credibility (print and interpersonal media), Pg21
The data for author characteristics is multidimensional. The correlation between variables
were significant at 0.01 level however the Cronbach’s alpha value for reliability were below
0.8 level (α= .645). So that generation of a reliable single variable for author related
measure were not possible. Therefore final analysis has to be executed on each single
item in order to achieve a representative outcome.
4.2.3 Measurement Results for Blog Credibility Factor Analysis
In order to estimate relationship between Blog credibility (dependent variables) and site
design content, visitor and author characteristics correlation analysis was used in this
study. While this case deals with one ordinal and one numerical characteristic the
Spearman rank correlation has been used as suggested by Cramer (2003). Test of
significance was 2-tailed and given the large sample size almost all tests reached
statistical significance (0.000). It must be stated at this stage, correlation does not mean a
direct cause and effect between factors but a sign of linear relationship (Fink, 2003b).
PRIMARY RESULTS Visitor Related Factors and Blog Credibility A correlation analysis was conducted between Blog Credibility Factor, Author Credibility
Factor and four variables specified for visitor’s internet reliability (Please see Appendix 6
for SPSS output). Close examination of Appendix 6 - Table 1 indicate that there is an
association between the blog credibility and visitor’s internet reliability. Using the Internet
regularly to decide what to cook and learn recipes appears to be the most effective
component (Spearman’s rho=.228, N=694). Both correlations between Blog Credibility and
variables “I use the Internet as an information resource (news, email, search engines)”, “I
use the Internet to learn about new products/developments in food category” are positive
and significant at 99% confidence level; however as the correlation coefficients are less
than 0.2 confidence level thus accepted as negligible. The correlation between blog
credibility and “I use the Internet only for work purposes“ is negative as expected, while
cooking is accepted as a hobby and people who use the Internet only for work purposes
tend to have less activities for leisure. So where respondent was closer to strongly agree
rating their perception of blog credibility decreases. In other words, it suggests that if
visitor’s number of online activities increases the perceived level of blog credibility might
– 55 –
increase. However the correlation coefficient between these two variables are very weak
for this study (Spearman’s rho=-.016, N=731).
Please note that, it has been taken into consideration that a significant correlation
between factors does not require a cause and effect relation as stated above. And
between variables there might even be two-way relations such as the “credibility
perception of the site” might influence “the usage of the Internet regularly decide what to
cook” positively. Whereas using the Internet to decide what to cook appears to have effect
on the credibility of the blog. Therefore, outcomes for this part of the study must be
interpreted depending on this particular warning.
Experience level of the visitor and Blog Credibility Variables “How long have you been an Internet user”, “How frequent do you use the
Internet”, “Do you have your own web site or blog” are tested for correlation with blog
credibility factors (Please see Appendix 6 – Table 2 for SPSS output). None of the factors
were significant enough to represent a correlation. However, only number of years being
an Internet user appears to have a negative effect on blog credibility with -0.097
Spearman’s rho correlation coefficient, the finding is also not significant.
Convenience Perception of Visitor and Blog Credibility Apparently convenience factors are highly associated with Blog and Author credibility
factors (Please see Appendix 6 – Table 3 for SPSS output). Preparing the shopping list
exactly on what the author recommends related highly with author credibility (Spearman’s
Rho=0.770, N=732, Sig.=0.000) and blog credibility (Spearman’s rho=0.687, N=702,
Sig.=0.000). Cooking recipes from the site follows has a moderate level correlation with
blog credibility (Spearman’s rho=0.43, N=695). Perceptions of Internet convenience as a
medium for information has a weaker but positive effect on the blog credibility
(Spearman’s rho=0.263, N=691). Personal relevance to the topic cooking has positive
correlation with credibility factors however as the correlation coefficients are less than 0.2
confidence level thus accepted as negligible. On the other hand personal relevance
factors correlated with “cooking recipes from the site” (Spearman’s rho=0.271) and
“Internet is a convenience source” (Spearman’s rho=0.288) at high statistical significance
(0.00) therefore might be accepted as a factor of blog credibility.
– 56 –
Site Familiarity and Blog Credibility According to bivariate correlation analysis between visit frequency and blog credibility,
there is a negative correlation (Spearman’s rho=-0.204, N=666). As frequency value
defined as reverse order smaller values indicate higher frequency, it can be suggested
that as visit frequency increases blog credibility might influence positively (Please see
Appendix 6 – Table 4 for SPSS output).. However the correlation between two variables is
weak for this sample group. The variable “Have you visited this site before” (Dichotomous
– Yes/No) tested using Pearson correlation and appear to be insignificant (-0.078).
Site Design and Content and Blog Credibility As stated above in the independent measurement section, there was one factor generated
for Usability, Design and Content variables. Correlation between “site related” factors and
blog credibility factors tested using Pearson’s correlation. Close examination of Appendix
6 - Table 5 indicate that there is a moderate level association between the blog credibility
and site design usability and content factors (Pearson coefficient=0.558, N=675). As in
other correlation analyses significant level were at the 0.01 and statistical significance
were high (0.000).
Interactivity Factor and Blog Credibility According to correlation analysis (Please see Appendix 6 – Table 5 for SPSS output)
user’s interaction with the blog has positive associations with the blog’s credibility
perception. However, the correlation is just close to borders of moderate level but still in
accepted weak correlation coefficient limits as it is 0.398 (N=684).
Perception of Author Characteristics and Blog Credibility According to correlation analysis (Please see Appendix 6 – Table 6 for SPSS output)
user’s perception about author’s expertise in the field has the highest correlation with blog
credibility factor. As suggested by previous studies, more the author perceived an expert
more the source perceived as credible (Spearman’s rho=0.494, N=699). As we would
expect, this is a two-way relationship whereas if the blog is credible surely the author must
be as the main source of messages.
The Perception of independence of the author also positively correlated with the blog
credibility as expected (Spearman’s rho=0.413, N=684). The variable “Messages given in
this web site are personal” appears to be less correlated with the credibility factor
(Spearman’s rho=0.294, N=690) however the inter-correlation between other author’s
independence characteristic is moderately strong (Spearman’s rho=0.508, N=724). For
– 57 –
this analysis confidence level was 99% and statistical significance was 0.000 for all of the
correlations.
4.2.4. Conclusion
According to findings based on correlation analysis, a final component analysis executed
in order to examine contributions from variables to a blog’s credibility (Please see
Appendix 7 for SPSS output). Below variables are selected as they are representing the
highest correlation with the credibility factor in the factor analysis. As can be seen final
output below Table 5 rotated component matrix for significant blog credibility variables 3
main factors appear to be influential. Site usability and content related variables which are
grouped under Site and Content Related Factor in the proposed model, convenience
related variables which are grouped under Visitor Related Factors in the proposed model
and trustability + believability of the blog including author characteristics of being an expert
and independent which are mainly grouped under Author Related Factors in the proposed
model.
Table 3 - Rotated component matrix for significant blog credibility variables
– 58 –
Therefore, we can conclude that the proposed model “The Path model of predictors of
blog credibility” proposes a correct approach for suggesting factors which influences a
blogs credibility. However the model is not presenting a complete picture of the relations
between these factors as can be observed in the factor analysis for credibility. It can be
stated that researcher is aware of the flows to this model. Resulting on the complexity of
the issue, learning outcomes might be criticized as unsatisfactory. However, it must also
be considered that sensible analysis of this data is beyond 20000 word dissertation. In
summary, future researchers must be aware of the complexity issue of the credibility
concept when taking this model as a base for their studies.
LIMITATIONS FOR FACTOR ANALYSIS As stated by Duncan (2003) “interpreting the results of a factor analysis solely based on
the items thought to make up a single index such as credibility is problematic”. Whereas it
is not possible to know without further information whether the final factor specifically
reflects author credibility or represents more general factor such as site usability and
design. Alternatively, when these factor grouped into three separate factors as in this
case, it would be harder to know whether a general factor such as “I cook recipes from
this site” had been broken down into two or more specific sub-factors (Adopted from
Duncan, 2003).
Therefore this factor analysis study must be interpreted as a way to illustrate the most
important relationship characteristics of factors influencing blog’s credibility.
4.3. BLOG’S EFFECTS ON BRAND ATTITUDE CHANGE ANALYSIS
The primary purpose of this part of the study is to test proposed hypothesis for blog’s
effects on brand attitude change according to proposed simplified “Framework for Blog’s
Effects on Brand Attitude Change” (See Chart 3 in Chapter 3). A series of factor analysis
and following that descriptive statistics and correlation analyses conducted in order to
explore the differences on changes in brand attitude according to independent variables,
blog credibility and user motivation.
4.3.1. Dependent Measures
For the purpose of this study, a multidimensional variable created to conduct the test.
Respondents were asked to report the possible brand attitude changes by rating their
– 59 –
intentions for following statements using a 5 point Likert scale 1 (Extremely unlikely) and 5
(Extremely likely) in total of six variables.
In order to produce factors for assessing reported attitude change, a component analysis
facilitating single component analysis using varimax rotation executed. Result of this
analysis two component was extracted, explaining 74,6% of the variance. Cronbach’s
alpha based on standardized items was computed on the resultant factor in order to
assess the scale reliability, which was found to be acceptable (α= .828). (Please see
Appendix – 8 for SPSS output). Factor 1 defined as Purchase Intention Factor with 4
variables “How likely is it to buy a new brand different than your regular brand for a
product (e.g. Soft cheese) recommended in a recipe by author?”, “If site author
recommends a brand's product (such as butter, flour, cheese) as a good product, how
likely you to buy and try that brand on your next purchase even it is different than your
regular brand”, “How likely is it to buy a new brand different than your regular brand for a
product (e.g. Soft cheese) recommended by another visitor in comments” and “If another
visitor comments a brand's product (such as butter, flour, cheese) as a good product, how
likely you to buy and try that brand on your next purchase even it is different than your
regular brand”. Factor 2 on the other hand defined as Memory Factor and with 2 variables:
“How likely will you remember recommendations by the author about a product when you
are shopping”, “How likely will you remember comments by other visitors about a product
when you are shopping”.
As a result of this factor analysis, in order to compare different result sets easier two
summated index variables are generated in order to measure reported attitude change,
Purchase Intention Summed and Memory Summed items.
4.3.2. Independent Measures
Blog Credibility: Definition of “Is the Source Credible?” As stated in previous factor analysis section, in order to represent blog credibility
perception a summed rating scale generated. As can be seen below table, this variable
has 5 as minimum value and 25 as the maximum value, which is a 5 times more of base
Likert scaled variables thus 5 for Strongly Agree and 25 for Strongly Disagree. Overall
perception of credibility of blogs is close to 19 which show a good level of credibility
(Please see Exhibit 8). The standard deviation for the summed blog credibility value is
close to 4 which show that the variance is not that high between the respondents
– 60 –
according to credibility perception. Please note that 55 of the respondents were accepted
as missing for this study while the rating for blog credibility were non-specified (0) values.
Summed Blog Credibility
Mean 18,39
Median 19
Std.Deviation 3,702
Frequency Percent
Strongly disagree 4 0,528401585
9 7 0,924702774
10 12 1,585204756
11 9 1,188903567
12 11 1,453104359
13 35 4,623513871
14 27 3,5667107
15 39 5,151915456
16 41 5,416116248
17 75 9,907529723
18 71 9,379128137
19 87 11,49273448
20 74 9,775429326
21 73 9,64332893
22 49 6,472919419
23 40 5,284015852
24 16 2,113606341
Strongly agree 32 4,227212682
Total 702 92,7344782
Missing 55 7,265521797
Total 757 100
Percentiles 25 16
50 19
75 21
Exhibit 8 – Descriptive statistics for Summed Blog Credibility Value
In order to test general perception of credibility, central value 15 (3 in 5 point scale) is
accepted as the limit. All summed blog credibility scores under 15 handled as No for this
study. On the other hand scores over and equal to 16 accepted as a positive answer to the
question “Is the source credible”. The answer valued 15 accepted as in between yes and
no, therefore total group of 39 respondents excluded from this analysis
Motivation: Browse or Search Second critical value for the hypothesis test was motivation of the visitor. Highest case
among respondents was Credible blogs * Browsing group of users with 68% of all
respondents. Credible blogs * Searching group was 38 respondents which is
– 61 –
approximately 5% of all respondents. For nonCredible blogs, Browsing group of users
were 12% and Searching group was approximately 2% of all respondents consequently.
The proposed the “Simplified Framework for Blog’s Effects on Brand Attitude Change” and
the distribution of respondents according to motivation value are as in below chart 5.
Chart 5 – Simplified Framework for Blog’s Effects on Brand Attitude Change and respondents
4.3.4. Hypothesis Test Results
OVERVIEW As stated in section 4.3.1 dependent measures for this test were Purchase Intention
Summed and Memory Summed values as indicators of reported attitude change. A
descriptive statistics analysis and correlation analysis between Summed blog credibility
variable and Purchase Intention Summed and Memory Summed variables has been
executed in order to provide empirical data to prove or reject the study hypotheses. Please
note that maximum value for a Purchase Intention Summed variable were 20 and for a
Memory summed variable were 10, which indicates that respondent was extremely likely
to change his/her attitude towards a brand.
Brand/Product
related message
received from a Blog
Is source credible?
Attitude Change
No
No
Summed credibility <15
N=105
Was consumer
Searching for
information?
No
N=92
Yes
N=13
No change
Was consumer
Searching for
information?
Yes
Summed credibility score>15
N=558
No
N=520
Attitude Change
Attitude Change
Yes
N=38
Summed credibility score=15
N=39 Excluded from the analysis
CASE 1
CASE 2
CASE 3
CASE 4
– 62 –
RESULTS
H1 – If the respondents were searching for information (high-involvement), they
more likely to report that their attitudes toward the brand mentioned on the
“credible blog” likely to change.
Among 38 respondents, 76.4% have indicated that they are likely to make a purchase
decision according to information they read on the blog. Values greater than 10 are
accepted as a positive attitude, whereas approximately 8% of the respondents reported an
extremely likely change in the attitude. For the same group of respondents, 52.6% of the
respondents reported that they are likely to remember comments they read on the blog.
Values greater than 5 are accepted as a positive attitude for memory indicator variable,
whereas 26% of the respondents reported an extremely likely attitude in the remembering
a comment.
According to correlation analysis between purchase intention summed, memory summed
and blog credibility summed variables, a positive relationship with low significance
detected. The highest correlation were between memory and blog credibility with 0.328
(Sig.=0,44, N=38), the correlation was significant at the 0.05 level. Please note that lower
number of respondents in this case effects the significance of the results negatively.
(Please see Appendix 8 – Case 1 for SPSS output for H1).
According to above stated empirical support H1 was accepted.
H2 – If the respondents were browsing (low-involvement), they more likely to report
that their attitudes toward the brand mentioned on the “credible blog” likely to
change. However, the reported change in the attitude expected to be lower than
high-involvement credible source case.
Among 514 respondents, 66.1% have indicated that they are likely to make a purchase
decision according to information they read on the blog and 5.6% of the respondents
reported an extremely likely change in the attitude. For the same group of respondents,
52.5% of the respondents reported that they are likely to remember comments they read
on the blog whereas 18.3% of the respondents reported an extremely likely attitude in the
remembering a comment.
– 63 –
According to correlation analysis between purchase intention summed, memory summed
and blog credibility summed variables, a positive relationship between variables at the
0.01 significance level has been detected. The highest correlation were between purchase
intention and blog credibility with 0.288 (Sig.=0,00, N=38). For this case the relation
between two attitude change indicators were more clear, while there were a positive
correlation with 0.333 correlation coefficient (Sig.=000, N=494). (Please see Appendix 8 –
Case 2 for SPSS output for H2)
Therefore, we can conclude that searching for information influences the involvement level
of the consumer for a credible blog, and high involvement influences consumer positively
to change their attitudes and remember comments they read on a blog. The percentage of
consumers reported to change their attitudes were lower for case 2 than case 1.
Therefore, According to above stated empirical support H2 was accepted.
H3 - If the respondents were searching for information (high-involvement), they
more likely to report that their attitudes toward the brand mentioned on the “non-
credible blog” likely to change. However, the reported change in the attitude
expected to be lower than high-involvement credible source case and low-
involvement credible source case.
Among 13 respondents, 30.8% have indicated that they are likely to make a purchase
decision according to information they read on the blog whereas none of the respondents
reported an extremely likely change in the attitude. Only 7.7% of the respondents were
reported a 15 likeliness to make a purchase decision in a 1-20 Likert scale. For the same
group of respondents, 38.5% of the respondents reported that they are likely to remember
comments they read on the blog whereas 15.4% of the respondents reported an extremely
likely attitude in the remembering a comment. It must be stated that for Case 3 least
number of respondents achieved, therefore resulting this narrow, inadequate sample for
the case the universality for this case might accepted as low.
According to correlation analysis between purchase intention summed, memory summed
and blog credibility summed variables, a high positive relationship between variables at
the 0.05 significance level has been detected. The highest correlation were between
purchase intention and blog credibility with 0.679 (Sig.=0,22, N=11). Please note that
– 64 –
lower number of respondents in this case effects the significance of the results negatively.
(Please see Appendix 8 – Case 3 for SPSS output for H3)
As correlation coefficient between credibility and purchase intention suggests, for a non-
credible blog, the little changes in credibility perception results bigger differences in the
purchase intention therefore attitude change as a result of blog readership. Mean values
for both attitude change indicators were also smaller than case 1 and case 2. This
indicates that there is likely a hierarchical relationship between them. Therefore, according
to above stated empirical support H3 was accepted.
H4 – If the respondents were browsing (low-involvement), they more likely to report
no change in their attitudes towards the brand mentioned on the “non-credible
blog”.
Among 92 respondents, 33.8% have indicated that they are likely to make a purchase
decision according to information they read on the blog whereas 1.1% of the respondents
reported an extremely likely change in the attitude. For the same group of respondents,
40.2% of the respondents reported that they are likely to remember comments they read
on the blog whereas 3.3% of the respondents reported an extremely likely attitude in the
remembering a comment.
According to correlation analysis between purchase intention summed, memory summed
and blog credibility summed variables, a positive relationship between variables at the
0.05 significance level has been detected. The highest correlation were between purchase
intention and blog credibility with 0.224 (Sig.=0,40, N=84). (Please see Appendix 8 – Case
4 for SPSS output for H4)
As general trend towards attitude change appear to be higher than case 3, it can be stated
that there is no hierarchical relationship between Case1-Case2-Case3 and Case 4.
Therefore, we can conclude that searching for information does not appear to be a major
influencer for a non-credible blog. According to empirical data stated above consumers
are reported that they are likely to change their attitudes and remember comments they
read on a blog even their credibility perception of the blog is low and they were not
searching for particular information. The percentage of consumers reported to change
their attitudes were higher for case 4 than case 3. Therefore, H4 was rejected.
– 65 –
4.3.4. Conclusion
According to Heath (2001) consumer’s attitude change process starts within the memory
as learning progress. Therefore, finding of two main factors representing purchase
intention changes and memory changes are found to be important outcomes at this level
of the study.
Browsing and searching appears to be an important indicator of the involvement of the
visitor. However, the rejection of H4 also suggests that regular usage of a blog likely to
increase the level of the involvement of the visitor. Even their overall credibility perception
is low, approximately 58% of the respondents in this group, reported that they are visiting
the particular blog almost everyday. In spite they are not confident about author’s
expertise (59.8% rated the author as 3 and below), majority of the population reported that
they are satisfied with the content (73.9% rated the content 4 and 5). Therefore, familiarity
and loyalty to the site itself appears to be influential on a possible attitude change as a
result of a blog readership.
It must be stated that the complexity of the issue explained in previous section also
appears in this case. In addition, to complexity of credibility perception, there are several
factors affecting consumer’s attitudes towards a brand and analyzing only but only one
factor’s effects on these changes are not possible using this research methodology.
While this study proposes a new approach to understand how blogs might have an effect
on consumer’s attitudes towards brands, the proposed framework for informal brand
information processing appears to present an incomplete picture of the current situation.
Therefore based on the outcomes of this study the framework must be restructured.
– 66 –
CHAPTER 5 – CONCLUSIONS
5.1. CONCLUSION
The most important outcome of this study is probably the detection of the level of influence
of blog readership to brand attitude change. As stated in Chapter 4 findings, in all four
different cases according to motivation and blog credibility perception respondents
reported their likelihood of attitude change ranging between 30% and 76%. Especially for
food category over 30% rates to purchase intention as a result of any advertising activity
are highly unimaginable. Even with the bias as a result of sampling method taken into
consideration, achieved values for attitude change reports highlights the potential of blogs
therefore the potential of consumer generated media (CGM).
As blogs are increasingly becoming main stream the influence of CGM will likely to
increase in the near future. According to ComScore research “Six of the top 10 blog-
hosting services have seen their traffic numbers grow by more than 100% from the first
quarter of 2004 to the first quarter of 2005. For example, Blogspot.com now draws more
traffic than NYTimes.com, USAToday.com, or WashingtonPost.com” (Claburn, 2005).
Many celebrity blogs are already became a point of attention and engaged with many
opinion leaders within online community.
As this study highlights blogs are capable of creating high loyalty among their readers.
Therefore, as a long term friend, blogs are remaining on the consumer’s side. Leveraging
the independence and expert characteristics and advantages of interpersonal
communication blogs appear to continue to influence consumer’s attitudes towards
brands.
5.2. RECOMMENDATIONS FOR THE INDUSTRY
I believe the value of this study for the Industry is lies under the analysis of credibility
factors of blogs. Past cases for commercial blogs showed that paid advertorials on blogs
or spoof blog sites does not capable to create same effect like a consumer created,
independent blog. However, instead of creating copycat blogs and try to fool consumers,
mimicking good characteristics of a blog might generate positive results for a commercial
blog.
– 67 –
5.3. RECOMMENDATIONS FOR FUTURE RESEARCH
As stated in the findings proposed models must be updated according to findings of this
study. Future researchers must be taken into consideration that, site visit frequency –it
might be defined in loyalty terms- is an important factor influencing involvement level of
the online consumer. There might be other motivational factors according to market and/or
sector. Therefore the complexity of the study would tend to increase. It must be taken into
consideration that overly complex data becomes an important block in front of the
researcher to make a sensible analysis of the data especially in a 20000 word limit if this is
a dissertation study. Please note that blog credibility is a complex multidimensional
concept which is slightly hard to simplify in order to generate a perfect explanation of the
issue.
Word count (20.766 including references)
– 68 –
REFERENCES
Bir A. Ali, 2005, Hürriyet Newspaper Online Edition, [Online] http://www.hurriyetim.com.tr/archive_articledisplay/0,,authorid~74@sid~9@nvid~571023,00.asp [accessed at: 02.05.2005]
Bloglines (2005) Home page, [Online] www.bloglines.com [Accessed at 26.08.05]
Branscombe M, (2005) The Business of blogs, The Independent Life 25.08.2005
Bryman A. (2001) Social Science Research Methods, OUP
Brown G. (1991) Big Stable Brands and a effects Freag thoughts about why, perhaps, consistent promotion keeps them big, Admap, 1991
Cho Chang-Hoan (1997) How Advertising works on the WWW: Modified Elaboration Likelihood Model, [Online] www.ciaadvertising.org/studies/reports/info_process/jcira.html [Accessed at] 02.08.2005
Claburn T. (2005) BLOGS ARE MAINSTREAM, InformationWeek, 8/15/2005, Issue 1052
Cramer D. (2003) Advanced Quantitative Data Analysis, Open University Press
Dahlen M. (2002) Thinking and feeling on the World Wide Web: the impact of product type and time on World Wide Web advertising effectiveness, Journal of Marketing Communications, Vol (8), pg. 115–125
Dutta-Bergman M. (2004) The Impact of Completeness and Web Use Motivation on the Credibility of e-Health Information, Journal of Communication, Vol(54) 2, pg. 253
Fink A. (2003b) How to analyze Survey Data, Sage Publications
Flanagin A., Metzger M. (2000) Perceptions of Internet Information Credibility, Journalism & Mass Communication Quarterly, Vol.77 No.3
Foxall G., Goldsmith R., Brown S. (1998) Consumer Psychology for Marketing, Thompson Business Press, 2nd Edition
Gotlieb J., Schlacter J., Louis R. (1992) Consumer Decision Making: A Model of the Effects of Involvement, Source Credibility and Location , Psychology & Marketing (1986-1998); May/Jun 1992, Vol (9) 3, pg. 191
Hansen E. (2004), CNET News.com, [Online] http://news.com.com/Yahoo+360+takes+spin+through+blogosphere/2100-1038_3-5619309.html [Accessed at:] 30.08.2005
Hawkins, S; Hoch, S. (1992), Low-Involvement Learning: Memory Without Evaluation, Journal of Consumer Research; Sep 1992; Vol (19) 2, pg. 212
Hewson C. et.al. (2003) Internet Research Methods, Sage publications
Ha Hong-Youl (2004), Factors influencing consumer perceptions of brand trust online, Journal of Product and Brand Management, Vol.13 No.5
Heath R. (2001) The Hidden Power of Advertising – How low involvement processing influences the way we choose brands, Admap Monograph 7 Admap Publications
Internetworldstats.com (2005) Middle East Internet Usage and Population Statistics http://www.internetworldstats.com/stats5.htm#me [Accessed at:] 05.09.2005
Jarboe G. (2005) The Search Engine Strategies conference report: Blogs, Boards, and Posts Capturing Consumer Buzz Online, April 7, 2005, SearchEngineWatch.com [Online] http://searchenginewatch.com/searchday/article.php/3495851 [Accessed at:] 30.08.2005
Johnson T., Kaye B. (2004), Wag the blog: How reliance on traditional media and the Internet influence credibility perceptions among blog users, Journalism & Mass Communication Quarterly, Vol.81 No.3
Johnson T., Kaye B. (2003), Webelievability: A Path model examining how convenience and reliance predict online credibility, Journalism & Mass Communication Quarterly, Vol.79 No.3
Johnson T., Kaye B. (2000), Using is beleiving: The influence of reliance on the credibility of online political information among politically interested Internet users, Journalism & Mass Communication Quarterly, Vol.77 No.4
Lyon P. (2004) Global Trends: Changes in Customer Journeys, Published on 08/10/05, Ogilvy Truffles Extranet
Mandase J. (2005), It’s Time to Think Small – Micro Actually, Admap, Issue 460
Mandase J. (2005-b) The Rise, and Fall of Institutional Media, Admap, Issue 459
Mattinson Deborah, Trayner Graeme (2004) Connecting with voters: How business can learn from new thinking in political research, Opinion Leader Research, Market Research Society Conferences
Mintel (2004) Leisure on the Internet - UK - June 2004. Mintel International Group Limited, London
Mintel-a (2005), Blog related reports, [Online] http://reports.mintel.com/sinatra/reports/search_results/show&type=NSItem&class=NewsAndMgr&sort=recent [Accessed at: 9.6.2005]
Neff J (2005) Taking package goods TO THE NET, Advertising Age. (Midwest region edition, Jul 11, 2005.Vol.76, Iss. 28; pg. S1, 3 pgs
Poh H., Adam S. (2002) An Exploratory Investigation of Attitude Toward the Website and the Advertising Hierarchy of Effects, Bowater School of Management and Marketing
Petty R., Cacioppo J., Schumann D. (1983) Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement, Journal of Consumer Research (pre-1986); Sep 1983; Vol (10) 2; pg. 135
Pew Internet and American Life Project (2004) Content Creation online [Online] http://www.pewinternet.org/pdfs/PIP_Content_Creation_Report.pdf [Accessed at: 04.06.2005]
Pew Internet and American Life Project (2005) The State of Blogging [Online] http://www.pewinternet.org/pdfs/PIP_blogging_data.pdf [Accessed at: 24.08.2005]
Riley D. (2005), Number of blogs now exceeds 50 million worldwide, [Online] http://www.blogherald.com/2005/04/14/number-of-blogs-now-exceeds-50-million-worldwide/ [Accessed at:] 27.08.2005
Oser K. (2004) More marketers test blogs to build buzz, Advertising Age, 9/13/2004, Vol. 75, Issue 37
Schiffman L., Kanuk L. Lazar (2004) Consumer Behavior, Pearson Prentice Hall, International 8
th Edition
Sifry D. (2005) State of the Blogosphere, August 2005, Part 1: Blog Growth, [Online] http://www.technorati.com/weblog/2005/08/34.html [Accessed at:] 26.08.2005
Sifry D. (2005b) State of the Blogosphere, August 2005, Part 2: Posting Volume, [Online] http://www.technorati.com/weblog/2005/08/35.html, [Accessed at:] 26.08.2005
Sifry D. (2005c) State of the Blogosphere, August 2005, Part 5: The A-List and the Long Tail [Online] http://www.technorati.com/weblog/2005/08/39.html, [Accessed at:] 26.08.2005
Silayoi P., Speece M. (2004) Packaging and purchase decisions: An exploratory study on the impact of involvement, British Food Journal, Vol(106) 8/9, pg. 607
Sudhaman, Arun (2005) Double-edged blog power puts marketers on guard, Media Asia, 7/1/2005
Technorati (2005) Home Page, Current Tracking Status, [Online] http://www.technorati.com [Accessed at:] 26.08.05
Viegas F. (2004), The survey on bloggers' expectations of privacy and liability, [Online] http://web.media.mit.edu/~fviegas/survey/blog/results.htm [Accessed at:] 27.08.2005
Wanten N., Burkell J. (2002) Believe it or not: Factors Influencing Credibility on the web, Journal of American Society for Information Science and Technology, Vol. 53/2
Warnick B. (2004) Online Ethos: Source Credibility in an “Authorless” Environment, American Behavioral Scientist, Vol.48/2