Does Perceived Advertising Value Alleviate Advertising Avoidance in Mobile Social Media? Exploring Its Moderated Mediation Effects

Round 1

Reviewer 1 Report

This research attempt to verify the importance of perceived advertising value by investigating its moderated mediation effect on the links between endogenous value.  The authors need to address several points. 

1. the clarity of the text must be improved.
2. The literature review is rather short to prove the validity of this research (just listing the literature as a list)
3.  The testing hypothesis is still difficult to find within the manuscript.
4. There are a considerable amount of irrelevant expressions in the manuscript (such as 'we' in the abstract and the text found frequently)
5. The linguistic quality of the manuscript must be improved. 

Author Response

Reviewer 1

This research attempt to verify the importance of perceived advertising value by investigating its moderated mediation effect on the links between endogenous value.  The authors need to address several points. 

1. The clarity of the text must be improved.

• Overall we did proofreading and rewriting with the help of Alison Pearce, our co-author.

1. The literature review is rather short to prove the validity of this research (just listing the literature as a list)

• We did literature review and make some corrections on it.

1.  The testing hypothesis is still difficult to find within the manuscript.
2. There are a considerable amount of irrelevant expressions in the manuscript (such as 'we' in the abstract and the text found frequently)

• We revised the abstract.

1. The linguistic quality of the manuscript must be improved. 

• We revised the whole manuscript with corrections.

Author Response File: Author Response.pdf

Reviewer 2 Report

Lit review part should be improved.

Research questions is not well organized.

Practical part should be explained more 

Author Response

Reviewer 2

Lit review part should be improved.

• We rewrite the literature review

Research questions is not well organized.

• We add one more research question such as “ (1) Do ubiquity, personalization, customization, information richness, entertainment, social interaction, and social integration affect perceived intrusiveness and privacy concerns?”

Practical part should be explained more 

• We rewrite Contributions, Discussions, and Implications.

Author Response File: Author Response.pdf

Reviewer 3 Report

Overall comments

This study is an ambitious and timely attempt to examine in an objective way the challenges and prospects that arise when a company's marketing activities in the "non-contact" society that has emerged with COVID-19 are surrounded by a variety of user-generated content that surrounds the consumer in ever more adapted and closed relationships.

In particular, as companies seek to win the hearts and minds of their customers by diversifying and upgrading the content and quality of their advertising, using creative storytelling and metaphors more than ever before, the authors' research on this topic can be expected to provide useful signposts for practitioners.

COVID-19 will increase the interaction between consumers and companies in cyberspace, and the increasing opportunities for immersive consumer experiences in the virtual, through the implementation of advanced technologies, will also increase the importance of this issue.

Research methodology
1. In order to test the research model and hypotheses, an offline questionnaire was used. The reader would benefit from a more detailed explanation of the data collection methods and why it was possible to use non-online methods to collect information in the corona.

2. In relation to point 1 above, it is stated that 374 valid responses were used for the statistical analysis. 52.1% (n=195) of the questionnaire respondents were male and 47.9% (n=179) were female, which is almost balanced in terms of gender. Please explain in more details how you targeted the respondents.

3. The data stating that "KaKao Talk" (n=308) was the largest mobile social media application used, but this could perhaps use some explanation. As a reviewer, I happen to know about this app, but readers of this journal may be familiar with 'Face book' (n=112), but not with Kakao Talk. This comment is not meant as a criticism of the study, but just a suggestion that some explanation might be helpful when considering the profile of the journal's readers.

4. Given that 80% of the study's respondents were university students, i.e., the digital native generation (and younger), it would be better to explain and discuss the relevance or significance of considering data only from a generation with a high affinity for social media. We hope that you will find it useful.

Model design, scaling and analysis

Basic descriptive statistics and factor analysis have been carried out in a reasonable manner and the results are presented.

As for the analysis method, it is stated that the hierarchical regression analysis by OLS was conducted using SPSS24. However, given that the title of this study is "Moderated Mediation Effects", I think more words should be devoted to the discussion in section 5.3. section 5.3 is relatively weak, as the previous sections have extensively described the preparatory analysis. In addition, Figure 2 presents the results of the analysis incorporating the moderated mediation effect of perceived advertising value. By the way, why did the authors not conduct a comprehensive analysis using a structural equation model?

Also, in the course of the analysis, the authors state, "We leveraged a classic bootstrapping-based analytic procedure developed by Hayes (2017) to scrutinize the conditional indirect effect with 5,000 repeated samples. In the course of the analysis, the authors discuss the bootstrapping method as if it is the default method, but perhaps, it would be beneficial for the readers to understand why you followed the approach if you could provide some more explanations with justification on it.

Discussions and Conclusion

7. the Contributions, Discussions, and Implications section is too mixed.

At the very least, the discussion of the results should be moved to the last section of the analysis, rather than to the conclusion. For example, the discussion relating to Figure 1 of the comparison of men and women appears in line 494, but this discussion could be expanded to extend the discussion from the perspective of the comparison of men and women, and this should be discussed in the discussion section following the results of the analysis, citing the relevant academic sources. 

Contributions should be clearly presented, with separate subheadings for contributions to theory and contributions to practice.

Overall, there is a dearth of literature on the latest innovative consumer behaviour theories and digital marketing-related topics. It is necessary to enrich and extend the debate by including the latest literature.

Author Response

Reviwer 3

This study is an ambitious and timely attempt to examine in an objective way the challenges and prospects that arise when a company's marketing activities in the "non-contact" society that has emerged with COVID-19 are surrounded by a variety of user-generated content that surrounds the consumer in ever more adapted and closed relationships.

In particular, as companies seek to win the hearts and minds of their customers by diversifying and upgrading the content and quality of their advertising, using creative storytelling and metaphors more than ever before, the authors' research on this topic can be expected to provide useful signposts for practitioners.

COVID-19 will increase the interaction between consumers and companies in cyberspace, and the increasing opportunities for immersive consumer experiences in the virtual, through the implementation of advanced technologies, will also increase the importance of this issue.

Research methodology
1. In order to test the research model and hypotheses, an offline questionnaire was used. The reader would benefit from a more detailed explanation of the data collection methods and why it was possible to use non-online methods to collect information in the corona.

• We explained in the limitations section and made corrections.

2. In relation to point 1 above, it is stated that 374 valid responses were used for the statistical analysis. 52.1% (n=195) of the questionnaire respondents were male and 47.9% (n=179) were female, which is almost balanced in terms of gender. Please explain in more details how you targeted the respondents.

• We already explained in the Data Collection section. Most of respondents (81.2%) were undergraduate students majoring on Business and Economics. The half of the students in business school were usually male, and the other half were female.

3. The data stating that "KaKao Talk" (n=308) was the largest mobile social media application used, but this could perhaps use some explanation. As a reviewer, I happen to know about this app, but readers of this journal may be familiar with 'Face book' (n=112), but not with Kakao Talk. This comment is not meant as a criticism of the study, but just a suggestion that some explanation might be helpful when considering the profile of the journal's readers.

• We already explained in the limitations section and in Korea, the most popular Social Network has been “Kakao Talk”.
• Because of the respondents’ geographical locations, KaKao Talk were the main social networks they have used.

4. Given that 80% of the study's respondents were university students, i.e., the digital native generation (and younger), it would be better to explain and discuss the relevance or significance of considering data only from a generation with a high affinity for social media. We hope that you will find it useful.

• We already explained in the Data Collection section but will explain further

Regarding the above suggestions, we would like to give you more specific explanations on data collections.

Our questionnaires were distributed within a university located in South Korea Southwest. We culled survey data from students (including undergraduates, graduates and senior high school students) by distributing off-line questionnaires under the assistance of staffs during final examination span around the school campus and finally, 434 responses were collected. During Corona pandemic, students were strictly temperature-assurance before taking final examinations. In comparison with other methods (e.g., online surveys), off-line questionnaires increase the control of the researchers during the survey process which reduces the concern in lacking of truthfulness of respondents and guarantee a high response rate (Cook et al., 2000). After deletion of the cases with missing and serious repeated data, 374 valid responses were used for statistical analysis. The demographic statistics show that 52.1% of the participants were male (n = 195) and 47.9% were female (n = 179). In age, almost 93.3% were in their 20s, and 6.7% were of other ages. In educational background, almost 86.4% of the participants were undergraduate students. Among the participants, 85.8% reported using mobile social applications over one hour daily, which is a rather large use of mobile social applications. The main mobile social media applications that we investigated were KaKao Talk (n = 308 users) and Facebook (n = 112 users). This result is consistent with the report of “YONHAP NEWS AGENCY” (2021) which indicate that in South Korea, as social app, “Kakao Talk”(39%) is the most frequently used which followed by Facebook(28%) and You Tube (28%). The types of advertisements that appeared most often in mobile social media were messages or text ads (n = 133, 35.6%), banner ads (n = 125, 33.4%), brand app ads (n = 97, 25.9%), and video ads (n = 62, 16.6%). The degree of perceived advertising exposure was high: 75.9% of the participants reported that they unintentionally received ads in mobile social media more than three times a day. In Table 1, we summarize the participants’ demographic statistics. Approximately 80% of the participants were college students, who were the most active social network users which was an echo of Celebi (2015) and Okazaki (2004) who contended that “young individuals” or “unmarried young females” playing an important role in using social network (10, 47).

Cook, C., Heath, F. and Thompson, R. L. 2000. A meta-analysis of response rates in Web- or Internet-based surveys. Educational and Psychological Measurement, 60(6), pp. 821–836.

Model design, scaling and analysis

Basic descriptive statistics and factor analysis have been carried out in a reasonable manner and the results are presented.

As for the analysis method, it is stated that the hierarchical regression analysis by OLS was conducted using SPSS24. However, given that the title of this study is "Moderated Mediation Effects", I think more words should be devoted to the discussion in section 5.3. section 5.3 is relatively weak, as the previous sections have extensively described the preparatory analysis. In addition, Figure 2 presents the results of the analysis incorporating the moderated mediation effect of perceived advertising value. By the way, why did the authors not conduct a comprehensive analysis using a structural equation model?

• We already explained in the Data Collection section. We rewrite section 5.3 with additional sentences. We already did the structural equation model analysis and got very similar results.
• But we would like to stay in OLS results. We would like to provide the next comments.

 “SEM-survey approaches are increasingly controversial, and PLS-SEM analysis is particularly controversial, as it tends to generate inflated pairwise correlations. In general, we would like to stay using regression approaches. Due to rising controversy over p-hacking and replicability. In other full information SEM approaches such as LISREL and AMOS require specification of populations and sampling design, determination of adequacy of sample sizes, and clear reporting of goodness-of-fit and other performance metrics. Residuals may be analyzed further, or conclusions be reported on the appropriateness of modeling assumptions in light of residual analysis.”

Also, in the course of the analysis, the authors state, "We leveraged a classic bootstrapping-based analytic procedure developed by Hayes (2017) to scrutinize the conditional indirect effect with 5,000 repeated samples. In the course of the analysis, the authors discuss the bootstrapping method as if it is the default method, but perhaps, it would be beneficial for the readers to understand why you followed the approach if you could provide some more explanations with justification on it.

• We would like to explain with the following:

“We leveraged the classic bootstrapping-based analytic procedure developed by Hayes (2017) to scrutinize the conditional indirect effect with 5,000 repeated samples.”

Compared with other indirect effects estimating methods (e.g., Sobel Test: only if has no access to raw data and Confidence Interval) in which only work well in large samples, our data analysis conducted on the raw data which is much more recommended using bootstrapping method, can offer a much better alternative that imposes no distributional assumptions (Preacher and Hayes, 2004, 2008). And at least 5000 resamples for final reporting are recommended by Preacher and Hayes (2008).

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879-891.

Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36, 717-731.

Discussions and Conclusion

1. the Contributions, Discussions, and Implications section is too mixed.

• We rewrite the Contributions, Discussions, and Implications with some corrections.

At the very least, the discussion of the results should be moved to the last section of the analysis, rather than to the conclusion. For example, the discussion relating to Figure 1 of the comparison of men and women appears in line 494, but this discussion could be expanded to extend the discussion from the perspective of the comparison of men and women, and this should be discussed in the discussion section following the results of the analysis, citing the relevant academic sources. 

Contributions should be clearly presented, with separate subheadings for contributions to theory and contributions to practice.

Overall, there is a dearth of literature on the latest innovative consumer behaviour theories and digital marketing-related topics. It is necessary to enrich and extend the debate by including the latest literature.

• We rewrite Conclusions section adding the following sentence and made additional corrections. The results were also very similar to those of the studies insisting the importance of the demographic variables in affecting advertising effectiveness (1, 7, 9, 20, 46, 67, 69).

• Regarding the above suggestions, we revised the parts of “Conclusions, Contributions” with some corrections and we would like to explain theoretical implications and practical implications as follows:

Theoretical implications

As for academic contributions, the study takes several valuable theoretical implications. First, for the uses of general Internet (24) and social media (32, 49), inforamtiveness, entertainment, irritation, and advertising value were important antecedents of attitudes toward advertising. However, the important effects of perceived advertising value based on demographic characteristics have been underestimated (1, 7, 9, 10, 20, 46, 67, 69). The results of our empirical analysis show significantly negative moderating effects of perceived advertising value on the trend related to irritation to avoid advertising, which suggested that advertising avoidance was alleviated among the female group and the group with low exposure to advertising. These results confirmed the importance of considering demographic characteristics (gender, advertising exposure frequency, and media use time) in advertising studies again.

Second, irritation, as a negative attitude, plays a strong mediating role in how perceived intrusiveness and privacy concerns affect advertising avoidance, which is an empirical application of the TRA in mobile social advertising. With the proliferation of social media, privacy has become a major issue of contemporary debate in the ICT field (29, 50, 68). Perceived intrusiveness as the determinant for advertising avoidance appears in various advertising media platforms (conventional mass media, the Internet, mobile platforms, and SNS). Unlike other studies, we combined perceived intrusiveness and privacy concerns together to investigate advertising avoidance in mobile social media. We found that perceived intrusiveness affects privacy concerns and advertising avoidance as mediated by irritation, which is another contribution to research on advertising avoidance.

Third, given the features of mobility and socialization, our research findings offer meaningful implications for previous relevant research. The effects of personalization and customization on perceived intrusiveness contrast with each other, which represents a meaningful contribution to information tailored to advertising in the ICT field. Prior studies have suggested that personalization and customization affect user attitudes and behavior (33, 40, 64). Sundar and Marathe (2010) successfully established that personalization and customization had a close relationship to perceptions of privacy, which in turn differently affect attitudes toward news websites according to “usage power.” “Power users” preferred system-initiated personalization with high privacy, but preferred user-initiated customization under low privacy. In a study on location-based advertising experiments, Lee et al. (2015) empirically confirmed that user-tailored customization mediated by perceived intrusiveness was a more effective strategy than was system-tailored personalization in inducing positive attitudes about LBA. We found significant differences in the effects of personalization and customization on advertising in mobile social media. These research results are the extent of the previous research in addressing the distinct roles of customization and personalization playing for affecting advertising.

Last but not at least, clear effects of social interaction and social integration on privacy concerns indicate that socialization is much more associated with privacy concerns, much as in the insights of Toch et al. (2012). In prior advertising-related studies, social gratification was shown to affect attitudes toward advertising (32, 49, 76). However, with serious leaks of user information (29, 50, 68) in social media that threaten user privacy, a comprehensive discussion on the “social role” should be initiated. We found that social interaction increases privacy concerns, thereby confirming the results of Lewis et al. (2008).  

Practical implications

Regarding the practical implications, the first major recommendation is for marketing managers, depending on the importance of demographic characteristics in marketing segmentation of advertising. In particular, to reduce advertising avoidance and increase advertising effectiveness, implementable programs should be developed for the group with less exposure to advertising, and especially for young females in mobile social media.

A second recommendation concerns the role of ubiquity, which is negative for advertising. Specially, market managers and advertisers should weight the benefits of mobile convenience and challenges to privacy associated with advertising brought from ubiquity. Our finding provides some caveats that although the mobility of social media facilitates access to advertising information, given the sensitivity of information, privacy, and intrusiveness, advertising exposure based on gender differences will be a good strategy.

Finally, regarding another exogenous values for advertising in mobile social media, several suggestions are going to be offered for market managers and advertisers. In terms of privacy, system-initiated personalization in mobile social media should be a primary concern. However, considering the relevant service come from personalization, allowing for increasingly intensive personalization is a better way to alleviate advertising avoidance. Informativeness alleviating perceived intrusiveness with statistical significance suggests that having access to relevant and updated information in a timely fashion can decrease the perception of intrusiveness which in turn may minimizes advertising avoidance. We found that social interaction increases privacy concerns while social integration improves advertising effectiveness by reducing privacy concerns. Thus, when promoting advertising via mobile social media, any interaction between consumers via advertising elicits some privacy concerns. However, forming a virtual social community to improve “social integration” is an effective approach to reducing privacy concerns, which subsequently minimizes advertising avoidance.

• We would like to stay the flow of Contributions, Discussions, and Implication with small changes.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

It is fine.

Author Response

Our co-author, Native English Speaker, Alison Pearce, has revised the whole documents already.

Reviewer 3 Report

As online marketing activities become more active, and messages from companies are sent through a variety of media, the study confirmed that the perceived intrusiveness and privacy concerns mediated by consumers' irritation directly affect their avoidance of advertisements.

Although the verification of perceived advertising value, as this study aimed to do, is expected to become a more important research topic in the future, the behavioral principle that the irritation caused by "intrusiveness" and "privacy anxiety" has a positive mediating effect on "advertising avoidance", which this study revealed, has already been proven  empirically by several research. I feel that a little more effort should be made to claim the originality and contribution of this research as a novel finding.

In this sense, this paper has developed an interesting model and validated it with data, and the evidence is quantitatively presented, so it would be possible to emphasize the value of this research by stating the novelty and original contribution more clearly in the abstract and conclusion.

On the other hand, I would like to suggest you include some more recent articles to support your discussions. 

Ubiquity and privacy concerns, personalisation and customisation, and social interaction in cyberspace are very current topics, and in this sense the focus of this study is timely and commendable, but the literature cited is hardly up-to-date (2020-2021). This diminishes the value of the study as a developmental study that relies on the latest academic findings. The topic of this research has been of interest to researchers ever since Internet-based marketing strategies became commonplace. Moreover, the recent COVID disaster has accelerated people's online purchasing behaviour, and companies have been interested in the recent metaverse interactions between companies and consumers, for instance. In this context, it is essential to consider and critically discuss the key themes and topics with latest academic article discussions.

Author Response

We add 6 articles about mobility, privacy concerns, personalisation, customization, and social interaction in cyber space. We have already revised our literature review section. Our co-author, Native English Speaker, Alison Pearce, have revised the whole manuscripts.

1. Hartemo, M., (2021). Conversions on the rise–modernizing e-mail marketing practices by utilizing volunteered data. Journal of Research in Interactive Marketing.
2. Holiday, S., Hayes, J. L., Britt, B. C., & Lyu, Y. (2021). The cause effect: The impact of corporate social responsibility advertising on cause consumer engagement behavior after Brand affiliation ceases.International Journal of Advertising, 40(2), 199-224.
3. Li, X., Wang, C., & Zhang, Y. (2020). The dilemma of social commerce: Why customers avoid peer-generated advertisements in mobile social networks. Internet Research, 30(3),1059-1080.
4. Olsen, G. D., & Pracejus, J. W. (2020). Customized advertising: Allowing consumers to directly tailor messages leads to better outcomes for the brand. Journal of Business Research,116, 245-257.
5. Sreejesh, S., Paul, J., Strong, C., & Pius, J. (2020). Consumer response towards social media advertising: Effect of media interactivity, its conditions and the underlying mechanism. International Journal of Information Management, 54,
6. Zhang, Y., Li, X., & Hamari, J. (2020). How does mobility affect social media advertising effectiveness? A study in WeChat. Industrial Management & Data Systems.