All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
The reviewers are satisfied that you have addressed the previous feedback and your manuscript is now suitable for publication.
[# PeerJ Staff Note - this decision was reviewed and approved by Jafri Abdullah, a PeerJ Section Editor covering this Section #]
The authors have addressed all my comments.
-
-
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
The author has arranged the paper very well and written in a professional and fluent English style. The abstract gives a general idea of the study's purpose, method, important findings, and implications in a very brief form. Some of these sections are, however, not descriptive enough. The introduction does the work and puts the social anxiety between college students into perspective, and links this to physical activity, family support, and self-efficacy. The reasons why the chain mediation effect may be hypothesized (physical activity, family support, self-efficacy, and social anxiety) are not strong enough. Though it is stated that these individual mediators are supported, the logic behind the fact that they are interacting sequentially with each other (i.e., how family support causes self-efficacy and not vice versa) is not discussed fully (Lines 48-63). Moreover, the information that is quite significant is not reflected in the Methods section:
The fact that there is no information regarding validity (reliability is the only one that is present), as can be seen, there is no information regarding validity according to the RS-3 Scale, Line 96.
The description of the PSS-Fa lacks transparency when it comes to scoring binary Likert (Lines 109-119). The statement that parents enjoy a good family support (Table 3) is not clear since the variable is binary (0/1), but mean scores were not indicated in that manner (M=0.60, SD=0.25; Line 141).
Ethical approval: I see that the consenting process is mentioned (Line 84), but the method of data collection (written consent, digital consent, etc.) and the anonymization of the data are not clear.
The cross-sectional type of the research is another important weakness as it does not enable making any causation conclusions concerning mediation, which is also recognized by the scientists but is not sufficiently clarified by the authors (Lines 318 320). Another factor that boosts family support is physical activity, which is another claim that does not reflect the opinion of causation held in the design (Line 190). The generalization is limited by sampling (that is, the stratified random sampling by the Sichuan Province does not allow extending the results to people in general, e.g., outside the Chinese or the university). Demographic data (Table 1) is not comprehensive enough: such variables as the socioeconomic status and family structure, which could be used to confound the relationships between family support or self-efficacy, are not mentioned. The scales used are well proven, but there is no uniformity in soliciting them.
General Self-Efficacy Scale (GSES): It was assumed to be unidimensional (Line 123), but there really are some multidimensional homologations. The latter approach is unjustified.
No Common Method Bias: there is no Harmon report (Line 129). Regarding this kind of design, single-wave, self-reporting, and common methodology variance seem to lead to overestimates in the correlation (e.g., PA ↔ Family Support: r = 0.53; PA ↔ Self-Efficacy: r = 0.52).
Their descriptive and statistical analysis is mostly rigorous (SEM, bootstrapping) and lacks homogeneity, making the reliability of certain analyses weak:
Confidence Effect Size: It gives the true effects of total direct as 66.91 (Line 150), whereas the Results section gives 66.86 (Line 233). The first effect is significant when using SEM (beta = -.187, p <.001; Line 214) and not significant in boots coding (beta = -.181, [95% CI = -.266, -.097]; Lines 230231). They must answer such contradictions
Pathway of Mediation: The chain mediation (Path 3) showcases only 13.35 percent of the total impact, unlike in the discussion of mediation pathways, where there is overemphasis on its role without indicating the fact regarding its low percentage when compared to that of the other paths (Path 1: 30.16 percent; Path 2: 23.40 percent).
The existence of covariates (gender, age, education) was reflected in the models that were used in PROCESS procedures (Line 203) but not in the SEM. That is, they do not report their impact on the model's fit or coefficients.
In the meantime, SEM tests are within an acceptable range (var/df = 3.346 and CFI = 0.978; Line 222), RMSEA ( 0.044), and SRMR ( 0.039) are good, but the justification for introducing correlated mediators is not present. Including a comparison model (parallel vs. serial mediation) would enhance validity.
The topic is within the time, and the use of the chain mediation is new in identifying psychosocial pathways. The study goes one step further (SEM, bootstrapping), and the raw information can be retrieved (supplemental files).
There are quite a few changes to pronounce on Verbalize longitudinal/experimental follow-ups.
Explain Statistical Irregularities: explain the impacts of the statistical irregularities (e.g., the significance of the direct effect) and give full results of Harman tests.
Theoretical Justification: Why is family support preceded by self-efficacy in the chain (e.g., refer to a social cognitive theory of Bandura)?
Section 2. Description of scoring of PSS-Fa: Describe consecutive scales as either coded as binary or composed and give equivalent scores (Table 3 and the text).
Generalizability: State the limitations of the sampling methodology that applies to Sichuan Province and suggest replication of the research with various territories.
Ethics & Consent: Disclose the consent procedure (digital/physical) and data anonymization procedure.
Discussion Balance: It is important to note that chain mediation improves the effects to a considerable extent, but it accounts for a small percentage of effects as compared to individual pathways.
- The English is clear throughout the manuscript.
- Most of the references are key and adequately introduce the theoretical framework of the introduction. For example: Bandura (1997) – a classic on self-efficacy. However, the most recent references are from 2018–2019 (e.g., Holt et al., McDowell et al.), but references from the last 3–4 years (2020–2025) are missing. It is recommended to update the theoretical framework and discussion by incorporating more recent scientific literature (last 5 years), especially relevant systematic reviews or longitudinal studies that appeared between 2020 and 2025.
- The figures and tables are legible, clear, and adequate. The raw data have been attached for review of the article. However, I cannot find throughout the article whether these will be deposited in a repository (e.g., OSF) to increase the transparency of the research. If this is the case, it should be added or explained.
- The research is original, uses a real-life sample of university students in China (391), and applies a chain mediation model to explore how physical activity influences social anxiety through family support and self-efficacy.
- The research question is inferred but not explicitly presented. It is justified that there is a knowledge gap, but the introduction could define it more explicitly, explaining why it is relevant to address it now (for example, we are in a post-pandemic context). It is recommended that the introduction more clearly frame the study's contribution compared to previous work and that its question or hypothesis be more prominently stated from the outset.
- Ethics approval and informed consent are provided, and it is indicated that the Declaration of Helsinki was followed.
- The design is appropriate for the type of study conducted.
- The assessment instruments are validated in the Chinese population, and Cronbach's α is appropriate for each of them.
- The statistical analysis is adequate: SEM with AMOS, mediation analysis with PROCESS, and bootstrapping with 5,000 iterations are used.
However, weaknesses are noted:
- The text does not include how the sample size was determined (statistical power analysis or justification), which would be desirable to reinforce rigor. If this has not been done, it is recommended that it be detailed.
- The text does not include whether the instrument was administered under controlled conditions (e.g., classroom, timetable) or entirely online without supervision. This affects the quality of the self-reported data. If this has not been done, it is recommended that the context of the questionnaire administration be included, or, failing that, that it be included in the study's limitations.
- The methods are described in sufficient detail for replication (instruments used, reliability, score calculation, PA categorization, and family support are detailed; statistics are adequately explained, including software and versions; and control variables are included).
- The figure provides a visual representation of the SEM model.
- The underlying data are robust, statistically valid, and controlled.
- The study uses appropriate methods: Pearson correlations, Structural Equation Modeling (SEM), mediation analysis using the PROCESS macro, and bootstrapping with 5,000 iterations.
- Standardized coefficients, confidence intervals, and significance levels are reported, meeting appropriate statistical standards.
- SEM model fit indices (CFI, TLI, RMSEA, SRMR) are reported, all within acceptable ranges.
- Demographic variables (sex, age, educational level) are controlled for in the regression models.
However, several recommendations are made:
- It is recommended that the data included in this review be hosted in a public repository to improve transparency and replication.
- The study is cross-sectional and correlational, so it cannot establish causality. It is recommended that all expressions that can be interpreted as causal be reviewed throughout the text. For example, "Physical activity reduces social anxiety" should be replaced by "is negatively associated with social anxiety." It is recommended that these be reviewed throughout the manuscript (abstract, discussion, conclusions, etc.).
Conclusions
- The conclusions adequately reflect the results obtained, highlighting the mediating effects of family support and self-efficacy, and cautiously interpreting the importance of the knock-on effect.
- Limitations are also appropriately mentioned.
Importance of the Contribution
- The article explores a mediation model already partially explored in previous studies. However, it provides added value by testing knock-on mediation.
However, it is recommended:
- Strengthen the theoretical justification of the study as a new contribution in a more specific, in-depth, and literature-based manner.
Congratulations to the authors for their work.
The work is adequate, interesting, and well presented.
However, significant changes would increase its value.
- The use of old references for key concepts and issues is understandable. However, it is recommended that the introduction be strengthened with new citations that update the state of the art. It would also be interesting to add a section to the introduction to "Current Study" specifically expressing the research gap that this article seeks to address, as well as the research questions.
Likewise, more recent citations in the discussion would be welcome; the most recent ones used are from within the last six years.
- The raw data should be made public in a repository to increase the transparency of the research, or it should be made clear that these can be requested from the authors if publication in a repository is not possible.
- Throughout the text, the results should be expressed without indications of causality, referring to associations between variables.
- The limitations should include the errors that can be induced by the use of self-administered questionnaires to strengthen this section. Furthermore, "future directions" are indicated and not explicitly shown. In addition to adding the "future directions," it is recommended to include the practical implications of the findings.
Overall, the manuscript is well presented, and the method is appropriate. However, the introduction and discussion sections should be strengthened.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.