SERVQUAL to Determine Relationship Quality and Behavioral Intentions: An SEM Approach in Retail Banking Service

Round 1

Reviewer 1 Report

This manuscript fills the gaps in the past literature and has good logic. Here are some of my reviews of this manuscript.

1. The first research question of this manuscript is "(1) How do SQ dimensions under the SERVQUAL model influence REQ and BVI in retail banking?" But this manuscript only discusses "ERVQUAL model influence REQ and BVI". It does not solve the first question.

2. Using too many "Therefore" in this manuscript.

3. The manuscript's definition of behavioral intention is vague. The reader can't understand whether it is repurchasing intention or loyalty. In the concluding chapter, loyalty is discussed. The contextual description of this manuscript confuses the reader.

4. Hypothesis 1 has nothing to do with the research question. This makes the research topic out of focus.

5. The SQ scale has reliability and validity analysis. Why is there no reliability and validity analysis for the other two scales?

6. How is the influence large compared determined? Scales are not standard measurement tools. There is no origin and no standard unit. Therefore, it is not recommended to use big or small to compare.

7. Table 3. ”t value: (0.05-0.17)” Why is the hypothesis supported? Is it correct? In addition, which method is used to test whether the mediation is supported?

8. "The banking sector of Bangladesh is suffering from a deposits crisis, bad loans, and various scams over the last one decades" This manuscript cannot get the result of this research. It is recommended to focus on the conclusions got from the research hypothesis.

9. Expanding the SERVQUAL dimension is not the key point of this research. The conclusion of this manuscript also points out that "SQ dimensions of SERVQUAL seemed to be more satisfactory to the customers". The statements in this section are contradictory.

11. Line 338. "The direct connection between SQ and 388 BVI was not observed in this paper". This statement is inconsistent with the hypothesis test result.

Author Response

Thank you for your review comments and suggestions. As your guidelines, our response to the details can be found in the Word document.

Author Response File: Author Response.docx

Reviewer 2 Report

Reviewer statement for the manuscript

SERVQUAL to Determine Relationship Quality and Behavioral Intentions: A SEM Approach in the Retail Banking Service

Reviewer Decision

1. The introduction must be rewritten. Why the solution methods (SERVQUAL and Structural Equation Models) were chosen should be mentioned in the introduction. In addition, although Structural Equation Models are used as a solution method in the study, no information is given about them in the introduction or any other section.
2. In the literature review section, the conceptual information that should be given in the introduction is given. Literature studies should be classified in terms of solution methods, SERVQUAL dimensions, application area, etc. Maybe, giving a table would be explanatory.
3. Hypothesis Development: The information given in the first two paragraphs of this section appears to be relevant to methodology and literature rather than hypotheses.
4. Figure 1 and Figure 2 are not cited in the text. Furthermore, descriptive and detailed information about these figures is not given.
5. Detailed information about the questionnaire used to collect data should be given. Maybe, the URL for an example of the survey could be given.

Author Response

Thank you for your review comments and suggestions. As your guidelines, our response to the details can be found in the Word document.

Author Response File: Author Response.docx

Reviewer 3 Report

Thank you so much for the opportunity to read this paper. The aim of the study was to visualize the impacts of the SERVQUAL or service quality model on the relationship quality and behavioral intentions of consumers in the retail banking services of Bangladesh. The work is within the scope of the journal; however, the following points are the concerns:

Abstract:

·         The abstract is not in a sustainability format.

·         The objective seems unclear. “The aim of the study was to visualize the impacts of the SERVQUAL or service quality model on the relationship quality and behavioral intentions of consumers in the retail banking services of Bangladesh”.

The impact of the SERVQUAL model on the relationship ….is confusing. “It may be the impact of SERVQUAL constructs on”.

·         The abstract need to extend the methodology in one or two sentences (research design, data collection methods and sample size etc.).

Keywords:

“SERVQUAL or Service Quality” is weird. Use only one. SERVQUAL is well known.

Introduction:

·         The introduction section is not well documented. The problem statement is insignificant to be able to attract the readers initially.

·         Please improve the background of the study. Particularly, explaining the retail banking industry in BD with recent statistics.

·         The literature gaps are very shallow or almost missing. The study lacks an explanation of the existing knowledge, and to what extent things are known and unknown.

·         Please include the significance of your study in the introduction section.

Literature review:

·         Give an overview of different service quality models

·         Give an empirical literature review (meaning what works done within the last 5 years or so) with a detailed Table.

·         Hypothesis development seems alright but is full of old literature. You will find so many papers on service quality.

Methodology:

·         Please enlighten us with the population of your survey and the samples of your study.

·         In your measurement of constructs, you did not discuss the items and that is why it is difficult to judge. Please add the questions in the appendix or abridged form of your items in Table 2 in the item section so that it helps readers to understand quickly.

·         Please give a detailed data collection procedure with the few unanswered questions, that need to address:

o   Why did you choose the purposive sampling method? Justify.

o   Which data collection methodology have you adopted?

o   When did you collect your data?

o   Was it a voluntary or rewarded survey?

o   Was it a questionnaire survey or face to face interview?

o   Address the ethical issues of the survey. Was it with consent or without the consent of respondents? Did you take approval from the ethical committee?

·         My serious concern is the use of the SERVQUAL Model where gap analysis is done. But the gap analysis was done. Probably you used ‘performance” only excluding expectation which is usually used for finding service gaps. Your work most probably matches with the SERVPERF model of Cronin and Taylor (1992). Although they used the same scale but only considers the performance. Please clarify this matter. There are serious flaws in the paper.

Result:

·         You used both the Fornel Larker Method but the HTMT calculation is superior to the prior one. Please incorporate the HTMT analysis.

·         Probably this study used a cross-sectional method, where there is a possibility of common method bias (CMB) issues. Please incorporate the CMB issue and how you handled those issues.

·         You used abbreviated words which need to be elaborated. Please add the elaboration in all tables where necessary.

·         The AGFI is below 0.90, please justify with reference.

·         You did not test the R² (models explanatory power) which is required. Even that is missing in the AMOS output. You need to re-run the analysis for that.

Discussion:

o   In the discussion section, please add a paragraph explaining what was your framework and how your model fits with the empirical study or like this.

o   You must include the hypothesis number to locate the discussion more easily by readers.

o   Please bring more comparisons of your results with the past results and critically discuss how your work filled the gaps.  

Implication:

o   Implications are too general and show a lack of theoretical contribution leading to the lack of novelty of the study.

o   You did not highlight the retail banking perspective.

o   Practical implications are also general. Please specify the suggestion on how to improve which service with example.

Limitations:

·         Please extend the limitations section a bit with further suggestions in terms of which construct could be included or which analysis could be incorporated in future studies particularly for the mediation or moderation constructs.

Reference:

The reference section is not updated. It used old references throughout the manuscript. Please update intensively.

Based on the above observation, I find the paper might contribute to the retail banking sector if properly addressed the issues mentioned earlier. However, there are many major concerns with the paper. Thus, I would like to recommend a major revision of the manuscript.

I hope this comment will not discourage you and will improve in a corrected submission. Wish you all the best!!

Author Response

Thank you for your review comments and suggestions. As your guidelines, our response to the details can be found in the Word document.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The author made major revisions in the revised version. But there are still two small opinions please pay attention to.

 1. Line 67.  "many studies have adapted the model of SERVQUAL". But it only cites one article. "many" is not suitable.

2. Figure 2. There are many factor loadings that are equal. Were these factor loadings set as free or fixed parameters when constructing the model?

Author Response

Comment: 1

Line 67.  "many studies have adapted the model of SERVQUAL". But it only cites one article. "many" is not suitable.

Author clarification:

We have corrected this; please see the red marked lines (67 and 68).

Comment: 2

There are many factor loadings that are equal. Were these factor loadings set as free or fixed parameters when constructing the model?

Author clarification:

No, none fixed parameters were used. Few factor loadings have such case.

Reviewer 2 Report

Thank you very much for your authors' interest. The manuscript can be published in Sustainability with this version.  

Author Response

We are very thankful to the you for your earlier anonymous review comments and suggestions. 

Reviewer 3 Report

Thank you so much for the corrections. The paper has been improved a lot, still some concern left to address. Please address the following issues before the paper gets published: 

1. SERVQUAL model works with the gap analysis based on Expectation-Performance scale. If you want to work with only performance, this must be clarified. Justify your scale in the manuscript. For your reference, please follow the paper: 

https://doi.org/10.3390/ijfs11010033

2. Correct the grammatical mistake. "did not required approval from the ethical committee". Explain why you did not require the approval from the ethical committee? What precaution you had taken? Or so.

3. Please incorporate the HTMT test as this is superior to the Fornell Larker. 

4. Please replace the figure 2 with the R square value. Also, fine tune the model in terms of its aesthetic part. 

Wish you all the best

Author Response

Comment:1

SERVQUAL model works with the gap analysis based on Expectation-Performance scale. If you want to work with only performance, this must be clarified. Justify your scale in the manuscript. For your reference, please follow the paper: https://doi.org/10.3390/ijfs11010033

Author clarification

Yes we have addressed the rationality of our theoretical ground in the 3^rd paragraph of section 1. Our clarified statements are as follows:

“Several studies have been accomplished on SQ in the banking industry using the SERVQUAL model and one of its derivatives (Kheng et al. 2010; Kumar et al. 2010). On the contrary, some studies have been criticized on both theoretical and operational grounds for the SERVQUAL model (Van Dyke et al. 1997; Cronin and Taylor 1992). In addition, Ranasinghe's (2021) studies have adapted the SERVQUAL model, despite problems associated with its dimensionality and its strength in various contexts, to measure SQ. However, there are still some limitations; for example, this model is not designed to include all possible influences on consumer decision-making in the banking sector. Consequently, our study explores empirical support for the use of a SERVQUAL modeling approach to SQ as perceived by retail bank customers in Bangladesh. Generally, the SERVQUAL model (with five dimensions: tangibles, reliability, empathy, assurance, and responsiveness) is the most extensive model for evaluating SQ around the world in numerous domains (Dey et al. 2019).

Comment: 2

Correct the grammatical mistake. "did not required approval from the ethical committee". Explain why you did not require the approval from the ethical committee? What precaution you had taken? Or so.

Author clarification

According to the suggestion, we have rewritten this sentence; please check red marked lines 246.

In our country, as far as I know there is no legal ethical committee who provide such consent. We used to use kind of filtering questions before answering the survey questions. For instance, in our study at the very beginning of the questionnaire, we clearly mentioned that “It was a voluntary survey and the participants' responses would be kept strictly confidential and would only be used for research purposes.” Also, participants are completely free to opt out of submitting a response if they wish to withdraw their submission.

Comment: 3 and 4

Please replace the figure 2 with the R square value. Also, fine tune the model in terms of its aesthetic part. 

Please incorporate the HTMT test as this is superior to the Fornell Larker. 

Author clarification

We are highly agreed with these comments, which have been drawn in earlier review as well. However, due to a technical problem we cannot analyze further. Unfortunately, our computer gets crashed, in which data and analysis was kept. We can understand how awkward situation we fall in now. Even though, we know how important to show R² value in structural model and HTMT test as well, but we are helpless now. If you permit us, we may remove structural model figure (Figure 2) from our manuscript, as path analysis results are also reported in Table 4.

Therefore, please pardon us in this regard, and we hope you can understand our case.

Round 3

Reviewer 3 Report

Thank you so much for the review. But I think authors are unable or unwilling to address my issues.

Author Response

Your perspective on the specific topic of discriminant validity was greatly appreciated. Yet, we repeat our previously stated position. We hope you will understand our situation.