Parent-Child Communication and Children’s Democratic Citizenship: The Roles of Growth Mindset and Peer Relationship

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This study focuses on Children’s Democratic Citizenship in South Korea and its possible predictors and mediators such as parent-child communication, growth mindest and peer relationship basing on children’s self-reports.

 

I found the study well written, the methods are clear and also the findings in a field not yet explored. I have only some minor concerns.

 

In the introduction I suggest to add some theoretical model who is the basis for the experimental design of the study.

 

In the data analysis, I am wondering why you didn’t check possible gender or age differences. I think that they could be useful to understand better the significative associations or predictions. The cultural factors could be very informative in this sense.

 

Probably it could be interesting to have information also on the location of the schools and on socio-demographic information of parents and to understand the possible associations.

 

I think that you can put this information in the discussion section as possible further studies.

Author Response

Comment 1: In the introduction I suggest to add some theoretical model who is the basis for the experimental design of the study.

Response 1: Thank you for pointing this out. I/We agree with this comment. Therefore, I/we have added an explicit theoretical framework to the introduction, grounding the study in three complementary theories that provide the basis for the proposed serial mediation model. The updated text can be found on page 5, paragraph 10, as follows:

"To provide a theoretical foundation for the present study, this research is grounded in three complementary frameworks: socialization theory, positive psychology, and ecological systems theory. First, socialization theory posits that parent–child interactions play a critical role in shaping children's social values and behaviors, including civic attitudes and responsibilities. Second, positive psychology emphasizes the importance of individual strengths, such as growth mindset, as key internal resources that influence adaptive functioning and social engagement. Third, ecological systems theory (Bronfenbrenner, 1979) highlights the interaction between individual and environmental factors, suggesting that children's development is shaped by both family (microsystem) and peer contexts. Based on these theoretical perspectives, this study proposes a conceptual model in which parent–child communication influences children's democratic citizenship both directly and indirectly through growth mindset and peer relationships."

Comment 2: In the data analysis, I am wondering why you didn't check possible gender or age differences. I think that they could be useful to understand better the significative associations or predictions. The cultural factors could be very informative in this sense.

Response 2: Thank you for pointing this out. I/We agree with this comment. Therefore, I/we have acknowledged this limitation and added gender, grade-level, and cultural factors as specific directions for future research in the Limitations and Future Research section. The updated text can be found on page 16, paragraph 5 (Limitation 5), as follows:

"Fifth, this study did not include detailed socio-demographic variables related to parents, such as educational level or socioeconomic status, nor did it examine differences across student demographic characteristics such as gender and grade level. These variables may play an important role in shaping children's social and civic development, including democratic citizenship. Previous research has suggested that demographic characteristics can be associated with variations in civic-related outcomes. Therefore, future studies should incorporate parental socio-demographic factors and examine differences across student demographic groups to provide a more comprehensive understanding of the contextual and individual influences on children's democratic citizenship."

Comment 3: Probably it could be interesting to have information also on the location of the schools and on socio-demographic information of parents and to understand the possible associations.

Response 3: Thank you for pointing this out. I/We agree with this comment. Therefore, I/we have acknowledged that the current study was limited to schools in City D, an urban area in South Korea, and that parental socio-demographic information was not collected. We have noted that these factors may be important for understanding contextual associations and have added them as directions for future research. The updated text can be found on page 16, paragraphs 2 and 5 (Limitations 2 and 5), as follows:

"Second, the study utilized nonprobability sampling and focused exclusively on upper-grade elementary school students from a single city (City D). As such, the findings should be interpreted as context-specific rather than broadly representative of South Korean children. The sample's geographic and demographic characteristics may limit the applicability of results to students in rural areas, different socioeconomic contexts, or other cultural settings. Future research should expand the sample to include students from all grade levels and diverse geographic regions to enhance the external validity of the results and to better capture variations across developmental stages and sociocultural contexts."

"Fifth, this study did not include detailed socio-demographic variables related to parents, such as educational level or socioeconomic status… Therefore, future studies should incorporate parental socio-demographic factors and examine differences across student demographic groups to provide a more comprehensive understanding of the contextual and individual influences on children's democratic citizenship."

Comment 4: I think that you can put this information in the discussion section as possible further studies.

Response 4: Thank you for pointing this out. I/We agree with this comment. Therefore, I/we have incorporated all of the above information — including gender and grade-level differences, school location, and parental socio-demographic factors — as specific directions for future research within the Limitations and Future Research section (Section 6), as the reviewer suggested. These additions ensure that the practical and contextual implications of these variables are clearly communicated to readers as avenues for subsequent investigation. The relevant text can be found on page 16, paragraphs 2 and 5 (Limitations 2 and 5), as cited in the responses to Comments 2 and 3 above.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

General Assessment

This manuscript attempts to define democratic citizenship through four dimensions—respect for human rights, responsibility, cooperation, and awareness of democratic procedures—while seeking to distinguish these from general morality or prosociality. While the research problem is relatively well-conceived, the manuscript suffers from a significant lack of understanding regarding statistical methodology and requires extensive revisions to its theoretical framework. In its current form, the paper remains a "well-packaged" correlational study; it lacks the theoretical and methodological rigor necessary to propose a serial developmental model. I hope the following comments assist the authors in revising this into a publishable manuscript.

 

1. The authors construct democratic citizenship using four dimensions, yet it remains unclear how these are statistically or conceptually distinct from moral development or general prosocial behavior. The manuscript needs to demonstrate how it measures the specific political and social attributes of citizenship—such as the will to participate in social issues and communal responsibility—moving beyond the simple virtues of a "well-behaved child.“

 

2. While the link between parent-child communication and democratic citizenship is plausible, the logic for selecting specific independent and mediating variables is weak. There is insufficient theoretical explanation as to why "growth mindset" is positioned as the primary mediator and "peer relationships" as the secondary one. Treating growth mindset as "psychological capital" and labeling peer relationships simply as "social capital" feels like a conceptual leap that lacks theoretical refinement. Furthermore, the logic connecting the proposed path (Parent-Child Communication → Growth Mindset → Peer Relationships → Democratic Citizenship) to the resolution of macro-level social issues is overextended.Crucially, the authors present a strong narrative suggesting that parental influence manifests as social behavior through internal beliefs, even though the direct path from parent-child communication to peer relationships was not significant. This approach appears to be an post-hoc rationalization of results rather than a test of a strictly pre-designed theoretical model.

 

3. The study utilizes self-reported, cross-sectional data from a single informant at a single time point. Despite this, the model assumes a developmental sequence. This path implies a temporal order that cross-sectional data simply cannot verify. While the manuscript acknowledges the limitations of a cross-sectional design, the Discussion and Conclusion sections treat the serial pathway as if it were a proven functional process. This is a classic case of over-interpretation and must be corrected. The authors should consider comparing competing or alternative models; at the very least, a robust theoretical argument for the current model's superiority is required. Proposing a "serial mediation" so prominently without such grounding is untenable.

 

4. A glaring issue is that the manuscript describes the analysis as Structural Equation Modeling (SEM) while actually using PROCESS Macro Model 6. The text cites Kline (2016) regarding a sample size of over 200 for SEM and states, "Prior to conducting structural equation modeling analysis." However, the actual analysis is a regression-based bootstrap mediation. These are not the same; PROCESS Macro uses observed variables, not latent variables. Therefore, applying SEM sample size logic is inappropriate, and the term "structural equation modeling analysis" is factually incorrect. This discrepancy is a serious error.Furthermore, the statement that the "overall model demonstrated good fit (F = 30.897, p < .001, R2 = .308)" is fundamentally flawed. F and R2 are indicators of a regression model's explanatory power, not "model fit" indices (like CFI, TLI, or RMSEA). Confusing these basic concepts significantly undermines the statistical credibility of the entire manuscript.

 

5. Total Scores vs. Dimensions: Although democratic citizenship is defined by four sub-dimensions, it appears to have been treated as a single total score in the analysis. There is no explanation as to why the sub-factor structure is valid for this sample or why using a total score is theoretically appropriate. Treating it as a unidimensional construct overlooks how parent-child communication or peer relationships might relate differently to "human rights" versus "democratic procedures.”

Mindset Scale: The authors reverse-coded fixed mindset items to create a total growth mindset score. While common, this requires a theoretical justification for placing fixed and growth mindsets on a single bipolar continuum.

Reliability Reporting: There is a conflict in the reporting of Cronbach’s alpha for the growth mindset scale (stated as .911 in the text but .977 in the table). Such inconsistencies suggest a lack of care in data processing and reporting.

 

6. The Discussion section fails to cautiously address the actual pattern of results (where the direct effect and the independent peer mediation path were non-significant). Instead, it generalizes the importance of peer relationships as if they were a broad, independent mediator, which the results do not support.More critically, in the interpretation of Table 4, the text describes the serial mediation path (X > M1 > M2 > Y) as having a confidence interval of ".144 to .326." However, the table shows this range actually belongs to the Total Indirect Effect. The text and table do not align, which calls into question the accuracy of the entire results section.

 

7. The manuscript uses various terms—citizenship, democratic citizenship, civic engagement, civic consciousness, democratic procedures, and prosociality—interchangeably without precise demarcation. Despite discussing South Korean elementary students, actual contextual variables such as the Korean national curriculum, classroom climate, or experiences with school democracy are entirely absent. Consequently, while the title suggests a study on "Democratic Citizenship Education," the actual analysis feels more like a generic study on positive developmental resources (family/peer assets). The focus of the paper remains blurred.

 

8. The data was collected in November 2022, but the projected publication is 2026. The authors should discuss how social changes during this interval (e.g., the shift in classroom environments post-pandemic) might influence the results.

 

9. Rather than offering generic suggestions like "studying the role of teacher support," the authors should propose specific core elements or curriculum structures based on their actual findings. This would provide the necessary "added value" to distinguish this paper from existing literature.

Author Response

Response to Reviewer 2

Comment 1: The authors construct democratic citizenship using four dimensions, yet it remains unclear how these are statistically or conceptually distinct from moral development or general prosocial behavior. The manuscript needs to demonstrate how it measures the specific political and social attributes of citizenship—such as the will to participate in social issues and communal responsibility—moving beyond the simple virtues of a "well-behaved child."

Response 1: We sincerely thank the reviewer for this insightful and valuable comment. We agree that clarifying the conceptual distinction between democratic citizenship and general prosocial behavior is essential. Therefore, we have strengthened the theoretical explanation in the Introduction section by explicitly distinguishing democratic citizenship as a socio-political construct. We clarified that, unlike general prosocial behavior, which reflects interpersonal kindness or helping tendencies, democratic citizenship involves active engagement with collective issues, participation in decision-making processes, and a sense of responsibility toward the broader community. Furthermore, we have revised the Measurement section to clarify how the instrument captures these socio-political attributes. Specifically, we emphasized that the items assessing democratic procedures, responsibility, and cooperation reflect children's willingness to participate in group contexts, resolve conflicts through dialogue, and engage in collective decision-making. The updated text can be found on lines 72–79 (Introduction) and lines 321–325 (Measurement section), as follows:

"This conceptualization distinguishes democratic citizenship from broader moral development or general prosocial behavior in three key ways. First, while moral development focuses on individual ethical reasoning (Kohlberg, 1984), democratic citizenship specifically emphasizes civic engagement and collective action within a political community. Second, whereas general prosociality encompasses any helpful behavior toward others (Eisenberg & Mussen, 1989), democratic citizenship requires deliberate participation in community decision-making processes and adherence to democratic norms and procedures. Third, democratic citizenship is not merely attitudinal or value-based but inherently behavioral—it requires active demonstration of civic competencies such as dialogue, compromise, and collaborative problem-solving in real social contexts (Westheimer & Kahne, 2004)."

Comment 2: While the link between parent-child communication and democratic citizenship is plausible, the logic for selecting specific independent and mediating variables is weak. There is insufficient theoretical explanation as to why "growth mindset" is positioned as the primary mediator and "peer relationships" as the secondary one. Treating growth mindset as "psychological capital" and labeling peer relationships simply as "social capital" feels like a conceptual leap that lacks theoretical refinement. Furthermore, the logic connecting the proposed path to the resolution of macro-level social issues is overextended. Crucially, the authors present a strong narrative suggesting that parental influence manifests as social behavior through internal beliefs, even though the direct path from parent-child communication to peer relationships was not significant. This approach appears to be a post-hoc rationalization of results rather than a test of a strictly pre-designed theoretical model.

Response 2: We sincerely thank the reviewer for this insightful and constructive comment. We agree that the theoretical rationale for the ordering of the mediators required further clarification. Therefore, we have strengthened the theoretical foundation of the serial mediation model by adding an explicit explanation at the end of Section 2.3 (lines 285–297). Specifically, we clarified that the proposed model is grounded in a theoretically driven developmental sequence. Parent–child communication is conceptualized as a primary socialization process that shapes children's internal belief systems, including growth mindset. Growth mindset, as a cognitive–motivational construct, influences how children interpret challenges and engage in social contexts, which is subsequently reflected in the quality of peer relationships. Furthermore, we emphasized that the sequential pathway (parent–child communication → growth mindset → peer relationships → democratic citizenship) represents a theoretically informed progression from family-based socialization to internal cognitive frameworks and then to social interaction contexts. We also explicitly stated that this model should be interpreted as a conceptual framework rather than a deterministic or causal developmental process. The updated text can be found on lines 285–297, as follows:

"The proposed serial mediation model is grounded in a theoretically driven developmental sequence… Thus, the sequential pathway represents a theoretically informed progression from family-based socialization to internal cognitive frameworks, and finally to social interaction contexts. This model does not assume a deterministic developmental process but rather represents a conceptual framework for understanding how these factors may be interrelated."

Comment 3: The study utilizes self-reported, cross-sectional data from a single informant at a single time point. Despite this, the model assumes a developmental sequence. The Discussion and Conclusion sections treat the serial pathway as if it were a proven functional process. This is a classic case of over-interpretation and must be corrected. The authors should consider comparing competing or alternative models; at the very least, a robust theoretical argument for the current model's superiority is required.

Response 3: Thank you for pointing this out. I/We agree with this comment. Therefore, I/we have revised the Discussion section to avoid causal or developmental interpretations and clarified that the serial mediation model represents a theoretically informed conceptual framework rather than a confirmed process. We also added statements emphasizing the cross-sectional limitation and suggested alternative or competing models for future research. The updated text can be found on lines 513–555, as follows:

"However, these relationships should be understood as theoretically informed associations rather than empirically verified developmental processes… Additionally, future studies should consider testing alternative or competing models (e.g., different ordering of mediators or bidirectional relationships) to better understand the robustness of the proposed framework."

Comment 4: A glaring issue is that the manuscript describes the analysis as Structural Equation Modeling (SEM) while actually using PROCESS Macro Model 6… the term "structural equation modeling analysis" is factually incorrect. Furthermore, the statement that the "overall model demonstrated good fit (F = 30.897, p < .001, R² = .308)" is fundamentally flawed. F and R² are indicators of a regression model's explanatory power, not "model fit" indices.

Response 4: We sincerely thank the reviewer for identifying this important issue. We fully acknowledge that the previous description of the analysis was inaccurate. Therefore, we have revised the Methods section to clearly state that the study employed a regression-based mediation analysis using PROCESS Macro Model 6. We removed all references to Structural Equation Modeling (SEM) and clarified that the analysis was conducted using observed variables with bootstrapping procedures to estimate indirect effects. We have also refined the description of the analytical procedure by replacing the previous phrasing with a more accurate explanation based on Hayes (2013), and clarified that this approach does not involve latent variable modeling. The updated text can be found on lines 382 and 388, as follows:

"The sample size (N = 212) was considered adequate for regression-based mediation analysis, particularly when using bootstrapping procedures… PROCESS Macro Model 6 estimates indirect effects based on observed variables using bootstrapping procedures… The regression model explained 30.8% of the variance in democratic citizenship (R² = .308), indicating a moderate level of explanatory power."

Comment 5: Although democratic citizenship is defined by four sub-dimensions, it appears to have been treated as a single total score in the analysis. There is no explanation as to why using a total score is theoretically appropriate. / The authors reverse-coded fixed mindset items to create a total growth mindset score. This requires a theoretical justification for placing fixed and growth mindsets on a single bipolar continuum. / There is a conflict in the reporting of Cronbach's alpha for the growth mindset scale (stated as .911 in the text but .977 in the table).

Response 5: We sincerely thank the reviewer for this important and insightful comment. We agree that further clarification regarding the use of total scores was necessary. In this study, democratic citizenship was conceptualized as a higher-order construct reflecting an overall disposition toward civic values and behaviors. While it consists of multiple sub-dimensions, we employed a composite total score to capture a general tendency of democratic citizenship, which aligns with prior studies that have treated it as a global construct in quantitative analyses. From a methodological perspective, the use of PROCESS Macro Model 6 requires observed variables, and composite scores were used to represent each construct. However, this decision was not solely driven by analytical convenience; rather, it was based on the theoretical assumption that democratic citizenship can be meaningfully represented as an integrated construct. We have also added this point as a limitation and suggested that future research examine differential effects across sub-components. Regarding the growth mindset scale, following Dweck's framework, fixed and growth mindsets are conceptualized as opposite ends of a bipolar construct, and reverse-coding is a commonly accepted approach to represent an overall mindset orientation. Finally, the Cronbach's alpha value has been corrected and standardized to α = .977 throughout the manuscript. The updated text can be found on line 347 (reliability) and lines 321–325 (measurement justification), as follows:

"Given that Kim's (2005) framework conceptualizes democratic citizenship as an integrated construct in which the four dimensions are theoretically interrelated and collectively define civic competency, a composite total score was used in the analysis… The four items measuring fixed mindset were reverse-coded for the analysis, following Dweck's (2006) original conceptualization that treats fixed and growth mindset as opposite ends of a single continuum; this is consistent with prior studies employing the same scale (Ryu, 2018)… The scale demonstrated excellent reliability (Cronbach's α = .977 for the total scale, with α = .918 for growth mindset items and α = .897 for fixed mindset items)."

Comment 6: The Discussion section fails to cautiously address the actual pattern of results. More critically, the text describes the serial mediation path (X > M1 > M2 > Y) as having a confidence interval of ".144 to .326." However, the table shows this range actually belongs to the Total Indirect Effect.

Response 6: We sincerely thank the reviewer for this careful and insightful observation. Therefore, we have made two revisions. First, we acknowledge that there was an error in the interpretation of the confidence intervals in the Results section. Specifically, the confidence interval (.144 to .326) was incorrectly attributed to the serial mediation pathway, whereas it corresponds to the total indirect effect. This has been corrected in the revised manuscript to ensure consistency between the text and Table 4. Second, we have substantially revised the Discussion section to more accurately reflect the actual pattern of results. We clarified that the independent mediating effect of peer relationships was not statistically significant and therefore should not be interpreted as a standalone mediator. We have moderated our interpretation to avoid overgeneralization and ensured that each pathway is discussed in accordance with its corresponding confidence interval. The updated text can be found on lines 454–458 (Results) and lines 524–527; 536–538 (Discussion), as follows:

"Finally, the total indirect effect of parent–child communication on democratic citizenship was statistically significant, as the 95% confidence interval did not include zero (β = .233, 95% confidence interval = .144 to .326). The specific serial mediation pathway through growth mindset and peer relationships (X → M1 → M2 → Y) was also statistically significant (β = .050, 95% confidence interval = .019 to .088)."

Comment 7: The manuscript uses various terms interchangeably without precise demarcation. Despite discussing South Korean elementary students, actual contextual variables such as the Korean national curriculum, classroom climate, or experiences with school democracy are entirely absent.

Response 7: Thank you for pointing this out. I/We agree with this comment. Therefore, I/we have incorporated this point into the Limitations section by adding a dedicated recommendation emphasizing the need to integrate contextual and educational factors, including the Korean national curriculum, classroom climate, and school-based democratic experiences, as directions for future research. The updated text can be found on lines 608–626, as follows:

"Sixth, while this study focused on the relationships among parent–child communication, growth mindset, peer relationships, and democratic citizenship, it did not directly incorporate contextual variables such as the Korean national curriculum, classroom climate, or students' experiences of school-based democratic practices… In particular, incorporating elements of the Korean national curriculum related to civic and moral education, as well as examining classroom-level democratic experiences (e.g., student participation in decision-making, classroom discussion culture), would provide a more comprehensive understanding of democratic citizenship development."

Comment 8: The data was collected in November 2022, but the projected publication is 2026. The authors should discuss how social changes during this interval might influence the results.

Response 8: We sincerely thank the reviewer for this valuable comment. Therefore, we have addressed this issue by adding a statement in the Discussion section acknowledging the time gap between data collection and publication, and noting potential changes in post-pandemic educational contexts. We also included this point in the Limitations section, suggesting that future research should consider longitudinal approaches to capture evolving social and educational environments. The updated text can be found on lines 627–633, as follows:

"Seventh, considering that the data for this study were collected in 2022, future research should account for potential changes in social and educational environments over time, particularly in the post-pandemic context. The COVID-19 pandemic has significantly influenced classroom dynamics, peer interactions, and communication patterns between parents and children. Thus, longitudinal or repeated cross-sectional studies are needed to examine how these evolving conditions may affect the development of democratic citizenship."

Comment 9: Rather than offering generic suggestions like "studying the role of teacher support," the authors should propose specific core elements or curriculum structures based on their actual findings.

Response 9: We sincerely thank the reviewer for this valuable and constructive suggestion. We agree that more concrete and practice-oriented implications would strengthen the contribution of the study. Therefore, we have revised the Limitations section by adding a set of specific educational elements derived from the study findings. These include (1) structured parent–child communication activities such as guided dialogue prompts, reflective journaling, and home-based discussion tasks; (2) classroom instructional strategies to explicitly foster growth mindset through feedback focused on effort and improvement and challenge-based learning; and (3) intentional peer interaction components such as cooperative learning, group problem-solving, and student-led discussions. By integrating these elements, we proposed a more concrete and structured framework for democratic citizenship education that connects family, individual, and peer-level influences. The updated text can be found on lines 634–652, as follows:

"(1) Programs should include structured parent–child communication activities, such as guided dialogue prompts, reflective journaling, and home-based discussion tasks… (2) Instructional strategies should explicitly foster growth mindset through classroom practices, including feedback focused on effort and improvement, opportunities for challenge-based learning, and activities that encourage persistence and reflection. (3) Peer interaction components should be intentionally designed within classroom settings, such as cooperative learning, group problem-solving, and student-led discussions, to provide contexts in which students can practice democratic values and social responsibility."

Author Response File: Author Response.pdf