The Dual Effects of Consumer Satisfaction on Brand Switching Intention of Sharing Apparel

Round 1

Reviewer 1 Report

Good work

Author Response

Thank you so much for your positive evaluation. You will never know how much your encouragement meant to us. Even though you didn’t propose any further revisions, it will be a great honor and help to us that if you would review the revised manuscript and comment about whether the revisions are appropriate. Thank you again for your time and effort.

Reviewer 2 Report

The study examines the impact of consumer satisfaction on brand switching intention, moderated by variety seeking and mediated by brand and platform trust. The idea is interesting. However, the methodology and discussion are weak.

In theory:

In hypothesis development, as a cross-sectional survey study, please do not use a casual relationship term, like promote and lessens.

In the discussion section, there is not much to say. Please explain more about the insignificant result of H5a. Also, pls discuss more about the significant results.

Pls add a conclusion section in the article.

The academic contributions of the study is weak. Please explain more in the context of sharing apparel. How does your result affect brand management , sharing economy and fashion business literature?

In methodology, please explain in detail about the data collection. It is a snowball sampling, how did you control the common method bias or external validity?

A table of sample profile should be added. You need to explain your sample profile to support your study.

Which regression analysis model did you use?

Pls explain the reasons of control variables selected.

Author Response

Thank you so much for your constructive comments! We believe that we have been able to address your comments and suggestions, and that our paper has been substantially improved thanks to you. Below, we indicate how we responded to each of your comments.

1. In hypothesis development, as a cross-sectional survey study, please do not use a causal relationship term, like promote and lessens. Thank you so much for this advice. We have changed the terms to be correlations, such as positively/negatively associated with.
2. In the discussion section, there is not much to say. Please explain more about the insignificant result of H5a. Also, pls discuss more about the significant results. Thank you so much this advice. We have added more explanations about the insignificant result of H5a and H5b, and discussions about the significant results.
3. Pls add a conclusion section in the article. Thank you so much for this advice. We have added a brief conclusion section.
4. The academic contributions of the study is weak. Please explain more in the context of sharing apparel. How does your result affect brand management, sharing economy and fashion business literature? Thank you so much for this advice. We have explained more about how this study contributes to the theory and practice, especially in the field of sharing apparel.
5. In methodology, please explain in detail about the data collection. It is a snowball sampling, how did you control the common method bias or external validity? Thank you so much for this advice. We have added more details about how the survey was conducted and a new section of “3.3 Common Method Bias” to explain how we control common method bias and why common method bias was not a severe problem in our study.
6. A table of sample profile should be added. You need to explain your sample profile to support your study. Thank you so much for this advice. In fact, we have already introduced the samples in the text. Maybe it is not necessary to added a table of sample profile? If it is necessary or other information about the samples should be provided, we will do that as suggested.
7. Which regression analysis model did you use? Thank you for this comment. We had constructed a moderated mediation model and tested the hypotheses with Boot-strapping method using the PROCESS V4.0 developed by Hayes in the SPSS. We first chose the “Model 4” in the PROCESS to test the mediating effects and then chose the “Model 14” to test the moderated mediating effects.
8. Pls explain the reasons of control variables selected. Thank you so much for this advice. We have added some explanation about why the control variables were selected.

Reviewer 3 Report

After reading the manuscript I must admit that my general opinion is fairly positive. The topic is interesting, indeed not widely covered by existing research while the research procedure sound and adequately developed. Theoretical background, research questions, hypotheses and data analysis are in my opinion valid and correct and the research in fact has merits and it is contributing.

My particular comments are:

1. In the abstract – there is an expression that ‘satisfaction lessens brand switching intention’ and right next – question if it is still true in the particular context – market. If in general it is true (and undoubtedly it is), are there any reasons to believe that it is not true in some particular case – here of ‘sharing apparel’?
2. Line 13 – I believe it should be ‘integrating the moderating effect of’.
3. It would be very beneficial to put one or two sentences of conclusions or some result interpretation at the end of the abstract as now it ends with analysis results’ description and thus seem to be incomplete.
4. Since ‘sharing apparel’ is a core term, and obviously it is not self-explanatory, it must be defined at the beginning of the manuscript; is it possible to say: ‘consumption of sharing apparel’ (line 43)?
5. In the line 34 it is mentioned that the business model of sharing apparel is questioned – it should be baked with some references, esp. given that the idea may be perceived differently in different countries.
6. I believe the statement that there is no consensual definition for satisfaction (line 77) is a bit farfetched, esp. when you are referring to the publication which is 22 years old, please refer to R. Olivier who provided unequivocal and widely accepted (second ed of his book, from 2014 has nearly 20k citations); the amount of factors governing satisfaction has nothing in common with the possibility to create a definition; nevertheless, I admit that there is a constant debate over the idea, factors and mechanisms behind satisfaction;
7. In the line 263 it is mentioned that all items were measured with 5-point Likert scale, while in ‘Variety seeking’ we have mentioned a 7-point scale; was it also a Likert-like scale or other type?
8. In ‘consumer satisfaction’ you aggregated two areas of satisfaction – product-related and platform-related; given the calculated Cronbach’s alpha participant does not seem to differentiate between these two constructs, so do not treat it as a remark, but don’t you think it would be a good idea to consider them separate entity – esp. that in the model you have two separate areas of trust; maybe it would be a good idea to discuss it in conclusions;
9. Path analysis seem to be sound, although I believe that assumptions of the method should be discussed and there should be an answer to what degree they are met (esp. that you used ordinal-level measurement) and what to what possible consequences it may lead; since you used bootstrapping, maybe it would be a good idea to present confidence intervals for each effect.
10. (mainly because of the remark I put in point 8) I cannot fully agree with the statement in the line 390 – that this is different from the classical approach that satisfaction simply lessens brand switching intention because in fact you have two different areas of satisfaction (product vs platform) and then two areas of trust (and in fact - two products in one model – apparel and platform) – they may work in different direction; that ‘traditional view’ you mentioned assume homogenous satisfaction (e.g. from a product or service) and one-dimensional product (i.e. a particular, one thing, not an association of product and platform;
11. The part with the discussion could be more developed, esp. in the area of the managerial implications, it would also be a good idea to put your results in the context of previous studies.

Author Response

Thank you so much for your constructive comments! We believe that we have been able to address your comments and suggestions, and that our paper has been substantially improved thanks to you. Below, we indicate how we responded to each of your comments.

1. In the abstract – there is an expression that ‘satisfaction lessens brand switching intention’ and right next – question if it is still true in the particular context – market. If in general it is true (and undoubtedly it is), are there any reasons to believe that it is not true in some particular case – here of ‘sharing apparel’? Thank you so much for this comment. Sorry about that we didn’t express clearly. Actually, we mean that satisfaction lessens brand switching intention in the common purchasing context, while sharing apparel is more like renting than purchasing. We have revised the expression to avoid misunderstanding.
2. Line 13 – I believe it should be ‘integrating the moderating effect of’. Thank you so much for this comment. You are right about it. We have revised the expression.
3. It would be very beneficial to put one or two sentences of conclusions or some result interpretation at the end of the abstract as now it ends with analysis results’ description and thus seem to be incomplete. Thank you so much for this advice. We have added some interpretation about the results at the end of the abstract.
4. Since ‘sharing apparel’ is a core term, and obviously it is not self-explanatory, it must be defined at the beginning of the manuscript; is it possible to say: ‘consumption of sharing apparel’ (line 43)? Thank you so much for this advice. We have added an explanation of what is sharing apparel at the beginning of the manuscript.
5. In the line 34 it is mentioned that the business model of sharing apparel is questioned – it should be baked with some references, esp. given that the idea may be perceived differently in different countries. Thank you so much for this advice. We have revised the expression and added two references to support our argument.
6. I believe the statement that there is no consensual definition for satisfaction (line 77) is a bit farfetched, esp. when you are referring to the publication which is 22 years old, please refer to R. Olivier who provided unequivocal and widely accepted (second ed of his book, from 2014 has nearly 20k citations); the amount of factors governing satisfaction has nothing in common with the possibility to create a definition; nevertheless, I admit that there is a constant debate over the idea, factors and mechanisms behind satisfaction. Thank you so much for this comment. We are not clear about which book you are referring to. We had searched for “Olivier 2014” but found no perfect match. Actually, we don’t need a consensual definition of satisfaction because we need to put forward a new definition for consumer satisfaction in the context of sharing apparel. We had referred to Giese & Cote (2000) just because we need to know what components should be included in our definition of consumer satisfaction of sharing apparel. Therefore, we have revised the expression from “no consensual definition” to “constant debate”, and use the three essential components in our definition. Thank you again for the advice.
7. In the line 263 it is mentioned that all items were measured with 5-point Likert scale, while in ‘Variety seeking’ we have mentioned a 7-point scale; was it also a Likert-like scale or other type? Thank you so much for this comment. We have checked the data and the manuscript. Actually, we mentioned that we used the “seven-item scale” from Menidjel et al.’s. It means that the scale includes seven items. It is also a 5-point Likert scale.
8. In ‘consumer satisfaction’ you aggregated two areas of satisfaction – product-related and platform-related; given the calculated Cronbach’s alpha participant does not seem to differentiate between these two constructs, so do not treat it as a remark, but don’t you think it would be a good idea to consider them separate entity – esp. that in the model you have two separate areas of trust; maybe it would be a good idea to discuss it in conclusions. Thank you so much for this comment. We did not realize this problem before. It will be interesting and insightful to consider them as two subdimensions in further studies. We have revised the expression in the manuscript, and discussed it as a major limitation/future direction of this study.
9. Path analysis seem to be sound, although I believe that assumptions of the method should be discussed and there should be an answer to what degree they are met (esp. that you used ordinal-level measurement) and what to what possible consequences it may lead; since you used bootstrapping, maybe it would be a good idea to present confidence intervals for each effect. Thank you so much for this comment. We have added confidence intervals for each effect.
10. (mainly because of the remark I put in point 8) I cannot fully agree with the statement in the line 390 – that this is different from the classical approach that satisfaction simply lessens brand switching intention because in fact you have two different areas of satisfaction (product vs platform) and then two areas of trust (and in fact - two products in one model – apparel and platform) – they may work in different direction; that ‘traditional view’ you mentioned assume homogenous satisfaction (e.g. from a product or service) and one-dimensional product (i.e. a particular, one thing, not an association of product and platform. Thank you so much for this comment. We have revised the expression to be more modest about our contribution and called for further studies.
11. The part with the discussion could be more developed, esp. in the area of the managerial implications, it would also be a good idea to put your results in the context of previous studies. Thank you so much for this comment. We have added more discussion about the theoretical contribution and practical implications.

Round 2

Reviewer 2 Report

The authors have tried to address my concerns. To improve this article, please:

1. Improve the section of 3.3. Pls try to use CFA with a marker variable (Williams et al., 2010). What items were reversed for testing? What do you think that response within 2 mins should be deleted from the sample size? What is the effective response rate? For your last argument, please provide a reference.

Williams, L. J., Hartman, N., & Cavazotte, F. (2010). Method variance and marker variables: A review and comprehensive CFA marker technique. Organizational research methods, 13(3), 477-514.   2. Include a limitation about the sampling method.   3. For hypothesis testing, it seems that you are using OLS regression. Pls state it in the text.   4. Pls explain more about the H5. It seems that you still believe that it is a significant relationship.  What do you mean "that brand trust are above the average level"? What is the average level of brand trust?   5. So, do you mean that the consumer start to trust the platform more than the brand, so switching intention is about the brand, but not the platform. I am a bit confused.   6. Finally, you have some interesting findings in your paper, but you need to interpret and discuss them enough to make it a good paper. 

Author Response

Thank you so much for your constructive comments! We believe that we have been able to address your comments and suggestions, and that our paper has been substantially improved thanks to you. Below, we indicate how we responded to each of your comments.

 

1. Improve the section of 3.3. Pls try to use CFA with a marker variable (Williams et al., 2010). What items were reversed for testing? What do you think that response within 2 mins should be deleted from the sample size? What is the effective response rate? For your last argument, please provide a reference.

 

Thank you so much for the advices.

 

Firstly, since we didn’t include a marker variable in the survey, we cannot conduct CFA with a marker variable with the present data. We have added this problem in the discussion of limitation.

 

Secondly, the item BSI3 was reversed coded, as presented in Table 2.

 

Thirdly, the survey contains 23 items, several demographic questions, and quite a few words of explanation, it takes more than 2 minutes to read and answer the questions as we do it by ourselves. It is also observed that most participants spent about 3-4 minutes to finished the questionnaire. It is reasonable to believe that the participants are not taking the survey seriously if they finish the questionnaire in less than 2 minutes. It is true that many of their answers are of the same score, including the reversed item.

 

Fourth, the effective response rate is 90.1%, which was reported in the section of 3.1.

 

Fifth, we have added three references in the section of 3.3, to support why the three steps helps to reduce common method bias and why a model with positive effects, negative effects and moderating effects will be less likely to be interfered by common method bias. This is one of the reasons why we didn’t include a marker variable in the first place.

 

2. Include a limitation about the sampling method.

 

Thank you for this advice. We have added a limitation about the sampling method.

 

3. For hypothesis testing, it seems that you are using OLS regression. Pls state it in the text.

 

Thank you for this advice. We have added a statement about OLS regression.

 

4. Pls explain more about the H5. It seems that you still believe that it is a significant relationship. What do you mean "that brand trust are above the average level"? What is the average level of brand trust?

 

Thank you for this advice. We realize that the explanation was not convincing enough. We meant to argue that H5 will only be true with all kinds of brands taken into consideration, but in the context of sharing apparel, most of the brands are premium brands. Consumers may be more loyalty to premium brands than to mass brands, because there are many substitutes to mass brands. As the data contained only consumers of premiums brands, the effect of brand trust on brand switching intention will always be strongly negative, no matter how we separate them into groups.

 

5. So, do you mean that the consumer start to trust the platform more than the brand, so switching intention is about the brand, but not the platform. I am a bit confused.

 

Thank you for this comment. In our study, switching intention is about the brand not the platform. We argue that consumers trust both the brand and the platform.

 

6. Finally, you have some interesting findings in your paper, but you need to interpret and discuss them enough to make it a good paper.

 

Thank you for this advice. We have added more interpretations about the findings.