The Role of Perceived Value and Risk in Shaping Purchase Intentions in Live-Streaming Commerce: Evidence from Indonesia

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The introduction adeptly establishes the context by addressing the increasing significance of live-streaming commerce and its relevance in the digital marketplace. It defines gaps in the current literature and justifies the necessity for the research.

The examination of demographic variables such as age and gender enhances the research scope.  However, the influence of the global pandemic on online buying is referenced but not explicitly connected to the study's aims. Thus, I suggest to improve this section by providing more details of the impact of pandemic on online shopping and live-streaming commerce. Recent data on these context would ve fruitful. Also, it would be better to identify the research questions and the structure of the following sections of the paper.

The literature review is thorough and organized, addressing essential themes such as live-streaming commerce, the SOR framework, and demographic factors. It offers a robust theoretical framework for the examination.
The hypotheses are thoroughly constructed but would gain from more explicit connections to the literature. Please provide clear justifications and arguements for each hypotheses separetely to enhance coherence.


The methodology sounds good. Howeve, the sample demographics are biased, featuring a predominance of female respondents (80.5%), thereby restricting generalizability. Please give more details on the representativeness of your sample. The study fails to consider potential biases arising from purposive sampling. Please justify the use of purposive sampling strongly. For example, why did you not use the snowball sampling or convenience sampling?

The structural model evaluation is thorough, with detailed statistical analysis. However, the discussion of perceived risk's unexpected positive effect on purchase intention is not adequately explained.

The discussion effectively interprets the findings and compares them with previous research. It highlights practical implications for marketers and businesses. On the other hand, the discussion does not sufficiently address the limitations of the study or the unexpected findings and the role of gender is dismissed without exploring potential reasons for its lack of significance.

The recommendations are somewhat generic and could be more specific to the study's context. Thus, please provide more specific recommendations, such as strategies for enhancing media richness and interactivity in live-streaming commerce.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This is a well-designed empirical study grounded in solid theoretical and methodological foundations. The manuscript addresses an important topic with practical relevance, and the core ideas are promising. The following comments are intended to help the authors strengthen the clarity, rigor, and overall contribution of the work.

1. The study explores a timely and relevant topic, and the overall framework shows potential. However, the manuscript would benefit from a clearer explanation of what makes its contribution distinctive. The proposed model appears quite similar to existing studies using the stimulus-organism-response framework in live-stream commerce. It would help to clarify the theoretical gap and articulate more precisely how this research adds to the existing literature. The literature review feels somewhat broad and could be better structured to build a stronger case for the hypotheses. Some of the hypothesized relationships seem underdeveloped in terms of theory. In the discussion, much of the focus is on repeating the statistical results, and there is limited interpretation of their broader theoretical or practical implications. The terminology throughout the paper is not used consistently, with terms such as streamer, influencer, and seller being used interchangeably. Figures and tables are also missing standard elements such as beta coefficients, R-squared values, and significance labels, which would improve clarity.
2. Some of the constructs in the measurement model seem conceptually mixed, which may weaken construct validity. For example, the items for perceived value appear to combine utilitarian, hedonic, and price-related dimensions, while those for perceived risk include financial, performance, time, and service-related risks. Interactivity and media richness show high correlation, and their discriminant validity appears questionable, with the HTMT value close to the threshold. Several indicators were removed from the analysis, but the manuscript does not explain the reasons for deletion or how content validity was maintained. It may be helpful to explore whether these constructs could be modeled as second-order factors. If this is not feasible, a clearer theoretical justification for the current structure would strengthen the measurement model. Including information on pre- and post-deletion reliability and validity, along with HTMT confidence intervals and cross-loadings, would also improve transparency.
3. The analysis section is generally well-structured, but there are a few areas that need further attention to improve clarity and rigor. Some values, such as beta coefficients, p-values, effect sizes, and significance labels, are inconsistent or unclear across the tables and text. The effect size from perceived value to purchase intention appears unusually large and may warrant closer inspection. In addition, the manuscript does not report indirect or total effects, which would help clarify the mediating role of perceived value and perceived risk. Suppression effects are not explored either, even though perceived risk may act as a suppressor in the model. Although the manuscript mentions PLSpredict, endogeneity testing, and nonlinearity, the explanations are brief and lack detail. It would be helpful to provide more information on how these checks were carried out. Including additional analyses, such as removing or isolating perceived risk, would help assess the robustness of the findings. If there are multiple robustness checks, consider moving those to an appendix to keep the main text more focused.
4. The sample is heavily skewed toward younger participants, with a small number of older respondents, and gender distribution is also imbalanced, with around 80 percent of the participants being female. These imbalances raise concerns about the generalizability of the findings, especially regarding the moderation analysis. Measurement invariance was not tested before interpreting moderation effects, which is important when comparing across groups like age or gender. Additionally, age was treated as a continuous variable in the interaction analysis, even though it was collected in ordinal categories. It would strengthen the analysis to first assess measurement invariance using MICOM, and then conduct a multi-group analysis based on age groups and gender. Reporting post-hoc statistical power for moderation effects would also be useful. Finally, the manuscript should clearly acknowledge the limitations related to sample representativeness. If possible, the authors could consider supplementing the sample or applying statistical weights to reduce bias.
5. There are several important constructs that could influence purchase intention but are not included in the model. These include social presence, trust, user involvement, prior purchase experience, and platform type. Including these factors, if data are available, would help strengthen the explanatory power of the model. In addition, the finding that perceived risk has a positive effect on purchase intention is unusual and not clearly explained. This relationship goes against much of the existing literature and deserves more theoretical discussion. It may be helpful to explore possible underlying mechanisms, such as diagnostic information, optimal stimulation, or consumer involvement. Running additional analyses, such as removing perceived risk from the model or analyzing its dimensions separately, could provide further insight. Including a simple conceptual diagram to illustrate how these mechanisms might work would also make the explanation more accessible and convincing.
6. The manuscript would benefit from greater transparency in how the data were collected and processed. Key details about the sampling frame, the time period during which data were collected, and the criteria used for data cleaning are not clearly reported. For example, it is not clear whether responses were screened for straight-lining, unusually short completion times, or outliers. There is also no mention of whether ethics approval was obtained or whether participants provided informed consent. In terms of common method bias, only full collinearity VIF values are reported, which is not sufficient to rule out bias in self-reported data. It would strengthen the manuscript to include information about procedural measures used to reduce method bias, such as assuring anonymity or randomizing question order. If feasible, consider using a marker variable or conducting a single-factor test to triangulate the risk of bias.
7. The manuscript would benefit from a careful review of language and formatting. There are occasional grammar and spelling issues, such as the use of “Presentase” instead of “Percentage,” along with inconsistent capitalization and citation formatting. Numerical values are sometimes reported with varying decimal precision, which affects clarity and consistency across tables and text. I recommend thoroughly proofreading the manuscript to correct language errors and ensure consistency in terminology, reference formatting, and number presentation, in accordance with the journal’s guidelines. Preparing a list of standardized key terms could also help maintain consistency throughout the text.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

 

The article presents current, important, and interesting issues. However, there are some aspects that are debatable or require changes or additions.

 

The most important such elements include:

1/ The article title identifies perceived value and risk as key categories. The purpose, however, generally discusses factors influencing consumer intentions. Therefore, the scope of the purpose is significantly broader than the article title. This is a significant inconsistency and an error, as the title should indicate the scope of the discussion in the article;

2/ The title lacks any indication of whose the intentions it is about;

3/ The purpose uses the category "consumer," while the article also uses the term "customer," which has a much broader scope. By definition, a customer is a broader concept than a consumer. Therefore, the terminology used throughout the article should be clarified;

4/ The Introduction mentions research questions, but does not provide their content;

5/ The purpose stated in the Introduction does not align with the purpose stated in the Abstract;

6/ The Introduction should conclude with a brief description of the article's structure;

7/ The Introduction makes no reference to the theory within which the discussion is to be conducted. This is mentioned in the Abstract, keywords, and then suddenly appears in a subsection within the Literature Review;

8/ The scope of variables included in the research hypotheses corresponds to the subject scope stated in the purpose formulated in the Abstract, but is significantly larger than the scope of the article's title;

9/ The study characteristics lack important information, including when the research was conducted, etc.;

10/ Too many places in the Discussion present results instead of demonstrating what the Authors found in comparison to the results obtained by other researchers;

11/ The Discussion also contains elements that are the Authors' speculations and not based on research results, e.g., "This result may be attributed to the nature of perceived risk...";

12/ The final section of the article lacks guidance on future research.

 

The article should be supplemented and modified based on the review's content.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear author(s), 
Thank you for the opportunity to review your manuscript. The topic is timely and relevant, particularly given the rapid growth of live-streaming commerce and the ongoing academic interest in the application of the Stimulus-Organism-Response (SOR) framework in digital marketing contexts. Your study is methodologically ambitious and offers insights into how streamer credibility, interactivity, and media richness shape consumer perceptions and behaviour. Below, I provide a detailed evaluation of each part of the manuscript. 

Title and Abstract
The title clearly reflects the study’s scope and theoretical framing. The abstract is well-structured, but it reads somewhat like a condensed results section. While the inclusion of coefficients and p-values demonstrates rigor, it risks overwhelming readers unfamiliar with SEM-PLS techniques. A stronger balance between theoretical framing, practical contribution, and main findings would make the abstract more engaging. Consider devoting more space to the conceptual rationale and managerial implications, and slightly less to raw statistical results. 

Introduction 
The introduction successfully establishes the relevance of live-streaming commerce and situates the study in the Indonesian context. The gap in identification, namely, the lack of an integrated examination of streamer credibility, interactivity, and media richness, is clearly stated. However, the introduction could benefit from a sharper articulation of theoretical contribution. At present, the narrative emphasises practical gaps but does not convincingly demonstrate how this study advances SOR theory beyond prior applications. Moreover, the claim about age and gender as moderators is well-motivated but seems secondary in framing. Consider revising the flow to emphasise the theoretical extension first and then position demographics as an applied nuance. 
Literature Review 
The review covers relevant areas, live-streaming commerce, SOR theory, perceived value and risk, but remains largely descriptive. Many sections summarise previous findings without sufficient critical engagement or discussion of contradictory evidence. For example, the discussion of perceived risk largely assumes it has negative consequences, yet your own findings show an unexpected positive association. Engaging earlier with such tensions would strengthen the originality of your framing. Furthermore, while recent sources are included, some key contributions from 2022 to 2024 are cited without deeper integration into your argument. A more critical synthesis is needed to show precisely where your model innovates relative to existing literature. 

Conceptual Framework and Hypotheses 
The conceptual model is coherent and logically derived from SOR theory. Hypotheses are well stated. However, there are inconsistencies in theoretical expectations versus empirical findings. For example, H11 predicts a negative relationship between perceived risk and purchase intention, yet you find a positive effect. This contradiction suggests the need for either a more nuanced theoretical justification or an ex-post refinement of the conceptual framework. Additionally, hypotheses on gender moderation are not sufficiently grounded in robust prior evidence. Recent studies suggest that gender effects in online commerce are diminishing, which aligns with your null results. This should be acknowledged in the framing. 

Methodology 
The methodological design is generally appropriate and clearly reported. The use of purposive sampling is justified, and power analysis through G*Power strengthens the validity of the sample size. The measurement model is transparent, with indicators and their sources reported. However, several concerns remain: 
Sampling: The dataset is heavily skewed toward younger consumers (46.6% aged 18–22, 35.3% aged 23–29), with only 2% above 40. This imbalance weakens the robustness of moderation analysis on age. While you acknowledge this in limitations, it should be more explicitly discussed in the methodology section itself. 
Measurement Items: Some indicators appear problematic (e.g., PV4 and MR5). While they were excluded in validity testing, this raises questions about construct specification. Clarifying why these items were initially included and whether they reflect cultural/contextual differences in Indonesian live commerce would strengthen confidence in your measures. 
CMB and Robustness Checks: The study goes further than many SEM papers by addressing CMB, endogeneity, and heterogeneity. This is a strength, though the discussion is highly technical. Consider balancing statistical detail with interpretation to ensure accessibility for a wider readership. 

Results 
The results are presented systematically, with tables and figures that are mostly clear. However, there are areas where reporting could be refined: The contradictory finding of perceived risk positively predicting purchase intention is presented but not sufficiently interrogated in the results section itself. It should be flagged as surprising before moving into the discussion. Some tables (e.g., measurement model assessments) could be condensed to improve readability without losing rigour. The reporting of path coefficients, t-values, and f² is comprehensive but at times repetitive. 

Discussion 
The discussion engages with prior literature and attempts to explain findings. The treatment of streamer credibility and media richness is convincing. However, the positive effect of perceived risk is insufficiently theorised. The explanation (that not all negative organismic states suppress behavioural response) remains vague. You should either explore plausible psychological mechanisms (e.g., risk as signalling authenticity, excitement, or social proof) or acknowledge that the result may be context-specific and unexpected. Similarly, the non-significant gender effects should be positioned more directly in light of emerging evidence that gender gaps in e-commerce are narrowing. 

Conclusion, Limitations, and Recommendations
The conclusion summarises findings but tends toward repetition. Consider reducing redundancy and focusing more sharply on what this study contributes theoretically (refinement of SOR in live commerce) and practically (guidelines for enhancing value and balancing risk perceptions). The limitations are candid but could be further developed. For example, the lack of cross-platform analysis is important given heterogeneity across TikTok, Shopee, and Instagram. 

References 
The reference list is extensive and largely up-to-date. However, integration into the main text remains uneven. In some sections, studies are cited only as illustrative examples rather than as part of a critical debate. A more dialogical engagement with recent works (e.g., 2023–2025 systematic reviews on live commerce) would enhance scholarly depth.

Kind regards,

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for your revised paper and responses. O believe that the paper has significant improvements thanks to your revisions.

Author Response

Dear Respected Reviewer,

We are deeply grateful for the constructive feedback you provided during the previous round.  Your feedback significantly enhanced the manuscript's rigor and lucidity.  We are appreciative of your favorable evaluation of the revised version and are prepared to provide any further clarifications that may be necessary during the production process.

Kindly Regards,

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have made meaningful progress in improving the manuscript’s structure, consistency, and overall readability. The paper is now more organized and easier to follow. That said, there are still a few important areas that need further attention before the manuscript is ready for publication.

1. The core model closely resembles existing stimulus–organism–response frameworks, and it remains unclear what specific gap this study addresses. Clarifying how the research builds on or extends previous work would help highlight its contribution more effectively.
2. Constructs like perceived value and perceived risk seem conceptually multidimensional but are modeled as single-layer constructs without explanation. Important validity indicators such as HTMT ratios, cross-loadings, and post-deletion reliability are also not reported, making it difficult to assess measurement quality.
3. The analysis focuses only on direct effects. Mediation, suppression, and total effects are not explored, even though the positive relationship between perceived risk and purchase intention is unexpected. Indirect pathways could offer more insight into how the model operates.
4. The moderation analysis does not test for measurement invariance, despite demographic imbalance in variables like age and gender. Without this step, it is difficult to interpret group comparisons with confidence.
5. Although the writing and formatting have improved, some issues remain. These include inconsistent use of terminology, missing figure labels, and minor grammar problems that affect clarity.

In summary, while the revised manuscript shows clear improvement, several foundational issues still need to be addressed. A stronger articulation of theoretical contribution, improved measurement transparency, and more complete analysis would strengthen the study’s academic value. I recommend another major revision before the manuscript is reconsidered for publication.

Author Response

Dear respected reviewer

Please find the attached document related to our responses of your fruitful feedback.

Kindly regards

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors The Authors have made significant corrections and additions to the article's text. The current version, resulting from their work, is undoubtedly better than the previous version. Therefore, the article in its current form is now available for publication.

Author Response

Dear respected Reviewer,

We sincerely thank for your positive assessment and for the constructive feedback provided in earlier rounds, which materially strengthened the manuscript. We are grateful for the recommendation for publication and remain available to address any minor editorial adjustments during production.

Kindly regards,

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors,

The manuscript has improved substantially since the earlier version. It now makes a clearer theoretical contribution by refining SOR to include dual appraisals of value and risk, with evidence from an understudied market. Methodological transparency has also improved. Some concerns remain, especially the sample skew, the relatively weak theorisation of moderators, and the at times overly technical reporting style, but these do not undermine the overall contribution.

The paper is suitable for publication after minor revisions, primarily aimed at:

Further streamlining technical reporting in the abstract and results.

Strengthening the theoretical integration of recent literature in the review.

Clarifying the exploratory nature of demographic moderators.

Expanding on possible cultural/contextual explanations for the positive risk–intention link.

Kind regards,

Author Response

Dear respected reviewer

Please find the attached document related to our responses of your fruitful feedback.

Kindly regards

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

Overall, the manuscript is in good shape and has improved substantially. I appreciate the effort the authors have made in addressing earlier comments. As a small suggestion for further strengthening the paper, you might consider slightly elaborating on the practical implications in the Discussion section. For example, a few additional sentences explaining how platform managers or streamers could apply these findings would help highlight the study’s relevance for practice. This is not a major concern, but such an addition could make the discussion more impactful.

Author Response

Thank you for the positive assessment and for the helpful suggestions to expand the practical implications. We have incorporated this by elaborating the practical implications in Section 5.4.

We appreciate your constructive feedback, it helped us make the manuscript clearer and more actionable for practice.

Author Response File: Author Response.pdf