Parental Expectation, Attitudes, and Home Numeracy Environment in Korea and in the U.S.: Potential Sources of Asian Math Advantages

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review the manuscript “Parental expectation, attitudes, and home numeracy environment in Korea and in the U.S: Potential sources of Asian math advantages” for Education Sciences. In this paper, the authors examine how differences in parental math attitudes, expectations and practices may explain differences in children’s math achievement between Korean and U.S. samples.

Overall, I found the paper interesting to read and the topic broadly applicable to the audience of Education Sciences. I have several questions regarding methodology and analyses as well as suggestions for improvements and clarifications that would strengthen this manuscript.

Introduction

1)    The Abstract includes information about only the Korean sample, but the Method describe participants from the U.S. as well. This should be updated to include both samples or a description of the full sample size with note that families came from either South Korea or the U.S.

2)    The authors review several studies focused on parents’ involvement and children’s math skills in the section, although there are a large number of other studies that could be included (see Daucourt et al., 2021, Psych Bulletin, for meta-analysis of the relations between many aspects of the home math environment and children’s math achievement). I would recommend the authors include reference to additional review articles and meta-analyses to supplement the studies they describe and provide a wider review of the literature (see, for example Douglas et al., 2021, Advances in Child Dev and Behav; Eason et al., 2022, Dev Review).

3)    The conclusion that “research has yet to identify any specific dimensions of the HNE […] that consistently predict child math performance (Mutaf-Yildiz et al., 2020)” seems a bit misleading, given there have been multiple meta-analyses suggesting overall positive, significant effects for various aspects of the home math environment and children’s math skills (Daucourt et al., 2021, Psych Bulletin; Dunst et al., 2017, Early Math Learn and Dev; Silver et al., 2024, J Exp Child Psychol).

4)    There are a few sentences with awkward construction that should be edited (for example, “Through play and conversations, mathematical language (ex: more or less).” on p. 3).

5)    There are a number of citations that could be added to support the claims in the last paragraph on page 3.

Method

1)    How many different schools did families come from in each sample? How many families were from each school? As they were tested at school, I am curious if children were all in the same classroom or had multiple different teachers? How many classrooms participated? This is a concern about clustering in the data (if some children come from the same classroom/schools while others come from other classrooms/schools there may not be independence in the data). Similarly, there may be additional non-independence due to different schools coming from the same/different provinces that also needs to be considered. Critically, the clustering needs to be accounted for in analyses.

2)    The authors note “In the current study, internal consistency reliability is .85, and the total standardized score, which accounts for child’s age, was used. We used a total raw score, not a standardized score, for the U.S. and Korean samples because it was not standardized in Korea.” on page 6. This is confusing: was the standardized score used for calculating Cronbach’s alpha but the raw score used in analyses? Why was the score used in analyses not the one used to calculate internal consistency? Please clarify these sentences as the way it is currently written seems inconsistent (which score was used for what?)

3)    How have the authors ensured that the items queried in the parent questionnaire are appropriate/relevant for families in South Korea, considering they were developed with Westernized families (as the authors note a major rationale for this work is to compare the two samples for possible cross-cultural differences, it is possible the types of items that accurately capture the home environment in one sample may not also be relevant for the other sample)? They note the “measure includes three subscales […] that are relevant in both cultural contexts” (p. 7) but have not described how this was concluded.

4)    Were any participants excluded for missing data, and did they differ from the included sample in country, child age, family SES, child gender, etc? It would be useful to know if the missing data was at random or if the excluded participants differed in meaningful ways from the included participants. Especially if there are potentially barriers leading to why those families did not participate – this is important to know for interpretation of the implications moving forward.

5)    How did the authors control for parental income in analyses? They compared the samples for parent education but not income in their descriptive statistics – was the income variable z-scored within each sample or some other such way to allow for comparing across samples and use as a covariate? This needs to be explained.

6)    It would be helpful to include full lists of all items (as well as which items were included in analyses versus not included in the parcels used for analysis).

Analysis Approach and Results

1)    Most importantly, the authors include reference to Tables in the text, but I do not see tables included in the manuscript or the supplementary material. Evaluating the results is thus nearly impossible and I feel unable to fully complete a review of the results and discussion. I have tried to identify questions and suggestions to the best of my ability below.

2)    I am concerned about potential clustering in the data (nonindependence) resulting from children coming from different schools (such that children from one school are likely to share more similarity with one another than children from different schools) and different schools in different provinces and worry that these random effects need to be considered in analyses.

3)    It is unclear which items were included in the latent construct parcels for each home math environment variable. Why were all items not included?

4)    A real strength of the approach is the close attention to ensuring measurement invariance across samples – I commend the authors for this! Was measurement invariance also tested for the child math assessments?

5)    The various SEM models testing different steps of measurement invariance and their results would be useful for readers to see.

6)    I am concerned about the comparison of scores between the two samples. Given that the authors conclude from their measurement invariance tests that the scalar invariance model did not fit the data well, it seems difficult to draw conclusions from the following results across the two samples and from the difference tests.

7)    What broader model was used for the latent mean comparison between the two countries?

8)    The authors seem to be using different fit indices cut offs to evaluate their models: in the measurement invariance models they describe a model with CFI = .96 and SRMR = .07 as not fitting the data well (likely because the chi-square test is significant) but then in their mediation models describe a model with CFI = .95 and SRMR = .06 (with a large chi-square value and no significance for that test noted) as fitting the data well. Consistency in interpreting model fit would be warranted.

9)    Figure 1 shows two separate mediation models but the authors describe only one set of fit statistics in the text. It is confusing.

10) In the caption for Figure 1 the authors note “The coefficients are unstandardized coefficients while the unstandardized coefficients are presented in parentheses.” I assume one of the labeled “unstandardized” coefficients are actually the standardized coefficient so this caption should be fixed.

11)The results on page 10 are confusing (see the sentence starting on line 480, especially).

Discussion

1)    How do the authors fit their findings of no association between parents’ math expectations and the home numeracy environment, and between the home environment and children’s math in the U.S. sample into the broader literature that tends to find positive associations for each of these?

2)    I find it hard to review the discussion and the conclusions and interpretations given my many questions about the results and analyses (and lack of full results to evaluate).

References

1)    The journal style asks for references to be numbered in order or appearance in text and for in-text citations to be noted with brackets.

Overall Remarks

This study (and its dataset) has the potential to answer interesting and important research questions, however in its present format the analyses conducted do not provide sufficient evidence to be able to answer them.

Author Response

Reviewer 1:

Thank you for the opportunity to review the manuscript “Parental expectation, attitudes, and home numeracy environment in Korea and in the U.S: Potential sources of Asian math advantages” for Education Sciences. In this paper, the authors examine how differences in parental math attitudes, expectations and practices may explain differences in children’s math achievement between Korean and U.S. samples.

Overall, I found the paper interesting to read and the topic broadly applicable to the audience of Education Sciences. I have several questions regarding methodology and analyses as well as suggestions for improvements and clarifications that would strengthen this manuscript.

Introduction

Comment 1: The Abstract includes information about only the Korean sample, but the Method describes participants from the U.S. as well. This should be updated to include both samples or a description of the full sample size with a note that families came from either South Korea or the U.S.

Response 1: We deleted the detailed method from the abstract and added a sample description of the U.S. children and families.

Comment 2: The authors review several studies focused on parents’ involvement and children’s math skills in the section, although there are a large number of other studies that could be included (see Daucourt et al., 2021, Psych Bulletin, for meta-analysis of the relations between many aspects of the home math environment and children’s math achievement). I would recommend the authors include reference to additional review articles and meta-analyses to supplement the studies they describe and provide a wider review of the literature (see, for example, Douglas et al., 2021, Advances in Child Dev and Behav; Eason et al., 2022, Dev Review).

Response 2: We agreed that these are important studies to be added. Thus, we added these suggested studies and slightly reframed our synthetic statement in the paragraph. Thank you for the specific suggestions.

Comment 3: The conclusion that “research has yet to identify any specific dimensions of the HNE […] that consistently predict child math performance (Mutaf-Yildiz et al., 2020)” seems a bit misleading, given there have been multiple meta-analyses suggesting overall positive, significant effects for various aspects of the home math environment and children’s math skills (Daucourt et al., 2021, Psych Bulletin; Dunst et al., 2017, Early Math Learn and Dev; Silver et al., 2024, J Exp Child Psychol).

Response 3: Yes, with those recent meta analysis studies, we found this argument inaccurate and misleading. We deleted the statement and reframed our synthetic statements in the paragraph (e.g., “There are some variations in the associations between the HLE and children’s math achievement. The heterogeneous findings could be due to sample features, methodological variations, or relationship between the two that is not strong as other influences. However, the existing literature, including several meta-analyses, support and weigh the positive associations between various aspects of the HNE and children’s math achievement outcomes).

Comment 4: There are a few sentences with awkward construction that should be edited (for example, “Through play and conversations, mathematical language (ex: more or less).” on p. 3).

Response 4: We edited the sentence that had a grammatical error and copyedited the entire paper thoroughly. 

Comment 5: There are a number of citations that could be added to support the claims in the last paragraph on page 3.

Yes, we added several more as suggested.

Method

Comment 6: How many different schools did families come from in each sample? How many families were from each school? As they were tested at school, I am curious if children were all in the same classroom or had multiple different teachers? How many classrooms participated? This is a concern about clustering in the data (if some children come from the same classroom/schools while others come from other classrooms/schools there may not be independence in the data). Similarly, there may be additional non-independence due to different schools coming from the same/different provinces that also needs to be considered. Critically, the clustering needs to be accounted for in analyses.

Response 6: We acknowledged the limitations of not using multilevel modeling. However, our primary aim was to investigate the mediation effect of parental math practices on the relationship between parental attitudes, and expectations on children’s math skills. Our focus is more on differences between countries rather than clustering effects.

p.14. Other limitations are related to an examination of the associations of parental math attitudes, expectations, and practices with child math outcomes in two different cultural contexts without taking into account the nested structure within each country, (i.e., certain parents may be nested in the same province), potentially leading to biased or inaccurate coefficient estimates

Comment 7: The authors note “In the current study, internal consistency reliability is .85, and the total standardized score, which accounts for child’s age, was used. We used a total raw score, not a standardized score, for the U.S. and Korean samples because it was not standardized in Korea.” on page 6. This is confusing: was the standardized score used for calculating Cronbach’s alpha but the raw score used in analyses? Why was the score used in analyses not the one used to calculate internal consistency? Please clarify these sentences as the way it is currently written seems inconsistent (which score was used for what?)

Response 7: Thank you for your detailed suggestions. We agree that the sentence can be confusing. We clarified the sentence as follows: The Cronbach’s alpha is .85 in the current study. While we used a standardized score for the U.S. sample, for the Korean samples, we used the total raw score for the Korean sample because the standardized score was normalized based on the U.S. populuation.

Comment 8: How have the authors ensured that the items queried in the parent questionnaire are appropriate/relevant for families in South Korea, considering they were developed with Westernized families (as the authors note a major rationale for this work is to compare the two samples for possible cross-cultural differences, it is possible the types of items that accurately capture the home environment in one sample may not also be relevant for the other sample)? They note the “measure includes three subscales […] that are relevant in both cultural contexts” (p. 7) but have not described how this was concluded.

Response 8: We compared items from the Korean version of the Home Environment Measure developed and validated in South Korea (Kim, 2015) with the Home Numeracy Environment Measure from the Western cultural culture (i.e., LeFevre, 2009). We identified items common to both measures, such as a child's ability to count to 20 and perform addition up to 10, which are likely less influenced by cultural factors.

Comment 9: Were any participants excluded for missing data, and did they differ from the included sample in country, child age, family SES, child gender, etc? It would be useful to know if the missing data was at random or if the excluded participants differed in meaningful ways from the included participants. Especially if there are potential barriers leading to why those families did not participate – this is important to know for interpretation of the implications moving forward.

Response 9: We appreciate your insightful comment! Participants with missing data were not excluded from the analysis. Instead, missing data were coded as -99, and Full Information Maximum Likelihood (FIML) was used to handle the missing values. We examined whether those with missing data differed from the rest of the sample in terms of country, child age, family SES, and child gender. No significant differences were found, suggesting the missing data were likely random and did not bias the results.

We handled missing data using Full Information Maximum Likelihood (FIML), which is suitable given the different patterns of missingness observed in our samples. A robust mean and variance-adjusted weighted least square (WLSMV) estimator was employed to account for the non-normality of the data structure in the U.S. sample.

Comment 10: How did the authors control for parental income in analyses? They compared the samples for parent education but not income in their descriptive statistics – was the income variable z-scored within each sample or some other way to allow for comparing across samples and use as a covariate? This needs to be explained.

Response 10: To ensure comparability between the Korean and American samples, parental monthly income was standardized. Income was scaled from 1 to 5 for each sample: in Korea, categories ranged from 1,000,000 won to over 5,000,000 won, and in the U.S., from $20,000 to over $100,000. This standardization facilitated meaningful comparisons across both cultures.

Comment 11: It would be helpful to include complete lists of all items (as well as which items were included in analyses versus not included in the parcels used for analysis).

Response 11: All items were included in the analyses. We realized that the original sentence sounded as if we only selected a few items for parceling, which is not true. We clarified the sentence. Due to the word limit, I have listed a few more example items for home math environments.

8. 8. each of latent construct was divided into three parcels, using all items from each scale.

p.7 These math-related activities include counting objects (e.g., how many spoons?), guessing the names of numbers in books and flyers, learning simple addition (e.g., 2 + 2), collecting objects (e.g., cards, stamps, spinning tops), reading storybooks that involve numbers (e.g., The Colorful Zoo, Twelve Baby Ducks Are Too Many), singing number songs (e.g., Carrot Song), and discussing time using calendars and clocks.

Analysis Approach and Results

Comment 12: Most importantly, the authors include reference to Tables in the text, but I do not see tables included in the manuscript or the supplementary material. Evaluating the results is thus nearly impossible and I feel unable to fully complete a review of the results and discussion. I have tried to identify questions and suggestions to the best of my ability below.

Response 12: We uploaded the tables into the system, but the table may be missed during the administrative process. Thank you for reviewing!

Comment 13: I am concerned about potential clustering in the data (nonindependence) resulting from children coming from different schools (such that children from one school are likely to share more similarity with one another than children from different schools) and different schools in different provinces and worry that these random effects need to be considered in analyses.

Response 13: Thank you for your insightful comment about potential clustering in the data due to children coming from different schools and provinces. We acknowledge this limitation in the discussion section; however, given the complexity of our analyses—particularly the path analysis comparing U.S. and Korean data—our primary aim was to reveal relationships between parental expectation, attitudes, home numeracy environment, and children’s math skills in distinct cultural contexts. While we recognize the importance of addressing clustering, we believe that incorporating random effects for schools and provinces would complicate the analysis without significantly enhancing our understanding of the research questions.

Comment 14: It is unclear which items were included in the latent construct parcels for each home math environment variable. Why were all items not included?

Response 14: All items were included in the analyses. We realized that the original sentence sounded as if we only selected a few items for parceling, which is not true. We clarified the sentence.

8. 8. each latent construct was divided into three parcels, using all items from each scale.

Comment 15: A real strength of the approach is the close attention to ensuring measurement invariance across samples – I commend the authors for this! Was measurement invariance also tested for the child math assessments?

Response 15: Thank you for your commendation regarding our focus on measurement invariance. We did a test for measurement invariance in the child math assessments, ensuring that the scales functioned similarly across Korean and U.S. samples for valid comparisons.

Comment 16: The various SEM models testing different steps of measurement invariance and their results would be useful for readers to see.

Response 16: In Table 3, we outline four steps for establishing measurement invariance, starting from configural invariance and progressing to scalar invariance.

Comment 17: I am concerned about the comparison of scores between the two samples. Given that the authors conclude from their measurement invariance tests that the scalar invariance model did not fit the data well, it seems difficult to draw conclusions from the following results across the two samples and from the difference tests.

Response 17: Thank you for your thoughtful comments on our study. We appreciate your acknowledgment of our focus on measurement invariance, and we confirm that we tested this for the child math assessments as well. Although the scalar invariance model did not fit the data well, making it difficult to draw direct conclusions about differences between the Korean and U.S. samples, we believe the data still offer valuable insights. We will emphasize these limitations in our discussion, clarifying that observed differences may not accurately reflect true differences in the constructs.

Comment 18: What broader model was used for the latent mean comparison between the two countries?

Response 18: Thank you for your question regarding the broader model used for the latent mean comparison between the two countries. We employed a multi-group structural equation modeling (SEM) approach, which allowed us to compare latent means while accounting for measurement invariance.

Comment 19: The authors seem to be using different fit indices cut offs to evaluate their models: in the measurement invariance models they describe a model with CFI = .96 and SRMR = .07 as not fitting the data well (likely because the chi-square test is significant) but then in their mediation models describe a model with CFI = .95 and SRMR = .06 (with a large chi-square value and no significance for that test noted) as fitting the data well. Consistency in interpreting model fit would be warranted.

Response 19: We appreciate your feedback on the interpretation of model fit indices and acknowledge the importance of consistency across different models. In our evaluation, we adhered to Hu & Bentler's (1999) cut-off criteria, where CFI values above .95 and SRMR values below .08 indicate a good fit.

However, we understand the confusion arising from the interpretation differences between the measurement invariance models and the mediation models. In the measurement invariance models, we placed greater emphasis on the Chi-square difference test due to the nested nature of these models. While CFI and SRMR are important, the Chi-square difference test is the primary tool for evaluating whether the additional constraints (such as equal factor loadings or intercepts) significantly worsen the model fit.

In the mediation models, which are not nested, we relied more heavily on overall fit indices like CFI and SRMR, as the Chi-square test becomes less informative in larger samples or more complex models due to its sensitivity to sample size. Thus, even though the Chi-square value was large, the other fit indices (CFI = .95, SRMR = .06) suggested an acceptable model fit in line with established guidelines.

Moving forward, we will ensure clearer communication and consistency in interpreting and reporting model fit indices across different models.

Comment 20: Figure 1 shows two separate mediation models but the authors describe only one set of fit statistics in the text. It is confusing.

Response 20: Thank you for your comment regarding the fit statistics in Figure 1. We conducted separate mediation models within a single analysis using multi-group structural equation modeling (SEM), which is why only one set of fit statistics is presented in the text. This approach allows for a simultaneous comparison of the models across the two groups.

Comment 21: In the caption for Figure 1 the authors note “The coefficients are unstandardized coefficients while the unstandadized coefficients are presented in paraentheses.” I assume one of the labeled “unstandardized” coefficients are actually the standardized coefficient so this caption should be fixed.

Response 21: Thank you for pointing out the inconsistency in the caption for Figure 1. We appreciate your careful review. We will revise the caption to clarify that the coefficients labeled as "unstandardized" are indeed the standardized coefficients, while the coefficients in parentheses are the unstandardized coefficients.

Comment 22: The results on page 10 are confusing (see the sentence starting on line 480, especially).

Response 22: Thank you for your careful reading of the manuscript. We clarified the sentence as follows.

p.11. For the U.S. sample, parents’ math expectation did not predict HNEhome numeracy environment after controlling for child age and parent income (=.16, SE=.19, p>.05), but parents’ math expectation predicted children’s math achievement (=.26, SE=.20, p<.01).

Discussion

Comment 23: How do the authors fit their findings of no association between parents’ math expectations and the home numeracy environment, and between the home environment and children’s math in the U.S. sample into the broader literature that tends to find positive associations for each of these?

Response: This is an important point to consider.  We have described some possibilities for this finding of no association on pg. 13 (last paragraph). Generally, some studies have found positive associations between the HNE and children’s math, some studies have find no association, and some studies have found negative associations between the home environment and children’s achievement. In our sample, we wonder if SES had an impact on this relationship between parent factors and children’s achievement that was stronger for the US sample than the Korean sample.

Comment 24: I find it hard to review the discussion and the conclusions and interpretations given my many questions about the results and analyses (and lack of full results to evaluate).

Response: We appreciate this comment and understand that it can be difficult to review the discussion of the findings when there are lingering questions and the results and analyses. We have made attempts to clean up the methodology section, adding additional details, and making the results more explicit that your doubts have been addressed.

References

Comment 25: The journal style asks for references to be numbered in order or appearance in text and for in-text citations to be noted with brackets.

Overall Remarks

This study (and its dataset) has the potential to answer interesting and important research questions, however in its present format the analyses conducted do not provide sufficient evidence to be able to answer them.

Reviewer 2 Report

Comments and Suggestions for Authors

I think that this is an interesting manuscript, which will attract a lot of interest. It has a good sample size; varied and well-selected measures; and appropriate and well-presented statistical analyses.

There are a few limitations and complexities, which the authors should discuss and may give the manuscript more depth.

(1)   One piece of information that we do not have is the parents’ own level of mathematical ability. It should be mentioned that in future studies, it would be desirable to include this variable, and that it might influence both the extent and effects of home numeracy

(2)   There should be more detailed discussion of the possible role of specific characteristics of Korean and American culture and education, For example, pupils from East Asian countries. including Korea, usually do better than those from other countries in international comparisons of mathematics performance. There is, however, some evidence (Lee, 2009; OECD, 2013) that they also experience high levels of mathematics anxiety compared to pupils in many other countries.

(3)   There should be some mention of the possible role of the attitudes of fathers as well as mothers. Some studies suggest that, due to assortative mating, fathers and mothers may show similarities in attitudes and performance in mathematic (Vanbinst et al, 2020), bur mothers tend to exhibit more anxiety and other negative attitudes toward mathematics.

References:

Lee, J. (2009). Universals and specifics of math self-concept, math self-efficacy, and math anxiety across 41 PISA 2003 participating countries. Learning and Individual Differences, 19, 355–365.

OECD. (2013). PISA 2012 results: Ready to learn. Students’ engagement, drive and self-beliefs (Vol. III). OECD Publishing: Paris.

Vanbinst K, Bellon E, Dowker A. (2020). Mathematics Anxiety: An Intergenerational Approach. Frontiers in Psychoogy,;11:1648. doi: 10.3389/fpsyg.2020.01648. PMID: 32793047; PMCID: PMC7385133.

Author Response

Reviewer 2:

I think that this is an interesting manuscript, which will attract a lot of interest. It has a good sample size; varied and well-selected measures; and appropriate and well-presented statistical analyses. 

There are a few limitations and complexities, which the authors should discuss and may give the manuscript more depth.

Comment 1: One piece of information that we do not have is the parents’ own level of mathematical ability. It should be mentioned that in future studies, it would be desirable to include this variable, and that it might influence both the extent and effects of home numeracy

Response 1: Thank you very much for this comment. We believe that this is captured by the parent's attitudes as we asked questions such as “When I was in school, I was good at mathematics, “When I was in school, I enjoyed mathematics,”.

Comment 2: There should be more detailed discussion of the possible role of specific characteristics of Korean and American culture and education, For example, pupils from East Asian countries. including Korea, usually do better than those from other countries in international comparisons of mathematics performance. There is, however, some evidence (Lee, 2009; OECD, 2013) that they also experience high levels of mathematics anxiety compared to pupils in many other countries.

Response 2: Based on the feedback, we added more explanation on potential cultural differences as the following. “This is consistent with previous studies that despite high scores in math, Korean students tended to have lower math self-concept and math self-efficacy less self-efficacy and more math anxiety (Lee. 2009; OECD, 2019). In East Asian countries, academic achievement, especially math, is highly valued and emphasized, competitions to excel in math are likely to be tense (Ho et al., 2002). Thus, Asian students tend to compare their abilities to peers and feel pressure to do better, which would contribute to negative attitudes and low expectations for their children.” We believe that we provide substantial discussion (the first section in discussion) on cultural differences..

Comment 3: There should be some mention of the possible role of the attitudes of fathers as well as mothers. Some studies suggest that, due to assortative mating, fathers and mothers may show similarities in attitudes and performance in mathematics (Vanbinst et al, 2020), bur mothers tend to exhibit more anxiety and other negative attitudes toward mathematics.

Response3 : This is a great insight.  We have mentioned this as a possible next step for studies (pg. 14).  We didn’t explicitly note whether fathers or mothers had completed the surveys, so it would be challenging to explore that angle in our current dataset. Nevertheless, I do think there are differences between parental attitudes and presents an intriguing angle to explore cross-culturally.

References:

Lee, J. (2009). Universals and specifics of math self-concept, math self-efficacy, and math anxiety across 41 PISA 2003 participating countries. Learning and Individual Differences, 19, 355–365.

OECD. (2013). PISA 2012 results: Ready to learn. Students’ engagement, drive and self-beliefs (Vol. III). OECD Publishing: Paris.

Vanbinst K, Bellon E, Dowker A. (2020). Mathematics Anxiety: An Intergenerational Approach. Frontiers in Psychoogy,;11:1648. doi: 10.3389/fpsyg.2020.01648. PMID: 32793047; PMCID: PMC7385133.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the opportunity to review the revised version of the manuscript “Parental expectation, attitudes, and home numeracy environment in Korea and in the U.S: Potential sources of Asian math advantages” for Education Sciences.

The authors have addressed sufficiently addressed my previous comments and I thank them for their efforts.

Comments on the Quality of English Language

N/A