The Psychosocial Health of Black/African Americans Compared with People of Other Races/Ethnic Origins during the COVID-19 Pandemic
, Tore Bonsaksen 2,3, Janni Leung 4, Mary Ruffolo 1, Gary Lamph 5, Karis Hawkins 6 and Amy Østertun Geirdal 7Round 1
Reviewer 1 Report
data analysis: you want to compare the 3 time points too, i.e. not just 3 independent regressions but one with time-point as predictor and interaction with race and the other predictors (with well-being as outcome) - that let's you find out whether there is a change over time
please report the software used for data analysis (e.g. R Studio, Stata, JASP)
given your large N but few who identified as Black/African, I suggest you use propensity matching (matchot package in R) to match the few Black/African by age, gender, education and marital status with a member from the "other" group. In this manner you can directly compare well-being between the groups
Mathematically nearly equivalent but often easier to display is if you run it as an ANCOVA, i.e. your well-being / loneliness etc as outcome, race (binary) as predictor and the other variables as covariates. Then in e.g. JASP you can plot the marginal means, i.e. the plot visualises the effect of race on well-being after controlling for the other variables (co-variates).
please report the overall performance of your regression models, i.e. R square, F value. Your models may explain from 1% to 20% of the variance. If the overall model explains only 1%, this is important information
line 46: empty square bracket
line 86 and 88 Black African Amercian - else you use Black/African or "Black or African" - inconsistent referencing, please choose one writing style
line 83 reference 19 appears before reference 12 and 13 (square brackets)
line 114 ata <- d missing
line 257 Chi Squared test <- Chi-squared test
table 2-5 last line racial identiy <- t missing in identity
line 223 odd order of references in brackets
line 245 not <- no
line 247 (2009) <- should that be a reference in square brackets?
line 293 imcomplete sentence, please finish it
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper has an interesting topic and it may attract the journal's readers. There are some issues that the authors should clarify. Firstly, it was stated that the convenience sampling method was used for data collection. This sampling method provides data that makes it difficult to generalize the results. Secondly, the authors wrote on page 9 lines 278-281 that " This study did not follow participants at each time period, so the responses are based on different participants with different intersections of identities. Although this study did not follow specific participants, survey fatigue from various sources may have impacted both response rates and given responses." To enable clear statistical comparisons, the same participants should have answered the questions at each data collection time point. The authors also mention this issue in the limitation section. Although it is good to acknowledge this, simply mentioning it in the limitations section is insufficient to validate the data and analyses presented in the study.
Another important statistical issue is the lack of information in the manuscript demonstrating that the data follows a normal distribution. Please provide evidence to support that your data is normally distributed.
While it was helpful to see detailed information regarding the scales used in the research, I could not find any information regarding the reliability values of the scales. Please include Cronbach's alpha values for both the original form and the ones you calculated.
The paper has an interesting topic and it may attract the journal's readers. There are some issues that the authors should clarify. Firstly, it was stated that the convenience sampling method was used for data collection. This sampling method provides data that makes it difficult to generalize the results. Secondly, the authors wrote on page 9 lines 278-281 that " This study did not follow participants at each time period, so the responses are based on different participants with different intersections of identities. Although this study did not follow specific participants, survey fatigue from various sources may have impacted both response rates and given responses." To enable clear statistical comparisons, the same participants should have answered the questions at each data collection time point. The authors also mention this issue in the limitation section. Although it is good to acknowledge this, simply mentioning it in the limitations section is insufficient to validate the data and analyses presented in the study.
Another important statistical issue is the lack of information in the manuscript demonstrating that the data follows a normal distribution. Please provide evidence to support that your data is normally distributed.
While it was helpful to see detailed information regarding the scales used in the research, I could not find any information regarding the reliability values of the scales. Please include Cronbach's alpha values for both the original form and the ones you calculated.
The authors could improve their discussion of the findings by situating them within the existing COVID-19 literature on an international basis.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
thank you for revising the manuscript
you find that only 2 to 7% of the variance in quality of life is explained by your regression analysis (R2).
Similar for psychosocial wellbeing (up to 10%,) and loneliness. Marietal status matters a lot.
It should be discussed that many - no measured factors - contribute to your outcome variables.
Given how skewed your sample is (under 5% are African Black American) propensity matching (matchit package in R) is more powerful (see e.g., https://en.wikipedia.org/wiki/Propensity_score_matching) as it controls for confounders - in your case it would be marietal status and age, as it has significant beta's
fine if you are not doing it. After all, there is a multiverse of analysis options, you could mention that in the discussion.
You state four countries, but than only US (line 112 ff)
Please clarify
some space issues before brackets
Author Response
Authors: Thank you for your time and energy to review the manuscript and provide recommendations for revisions. The revised manuscript has been uploaded with track changes saved in the document.
Reviewer 1 (R1): You state four countries, but than only US (line 112 ff) Please clarify which are the four countries? live in the US? Or being US citizen? (hence recruiting from other countries too) what kind of software did you use to analyse your data? R? Stata? SPSS? JASP?
Authors: Thank you highlighting this. This article is only focusing on the data collected in the US, which did ask if they lived in the US. SPSS software was used to analyze the data. This has been adjusted in the revised manuscript.
Reviewer 2 Report
The explanations of methodological issues in the paper are satisfactory to me. It seems that the authors have included Cronbach ALpha values for the scales used in the research. It was also good to see the VIF values for the multicolliniarity check. However, I had recommended that the authors develop the Discussion section by adding the sentence "The authors could improve their discussion of the results by situating them within the existing COVID-19 literature on an international basis". Unfortunately, this does not appear to have been done. Again, I recommend that the authors revisit the discussion section and support it with the international literature. COVID-19 has been a popular topic and there is a huge amount of literature on the subject.
The explanations of methodological issues in the paper are satisfactory to me. It seems that the authors have included Cronbach ALpha values for the scales used in the research. It was also good to see the VIF values for the multicolliniarity check. However, I had recommended that the authors develop the Discussion section by adding the sentence "The authors could improve their discussion of the results by situating them within the existing COVID-19 literature on an international basis". Unfortunately, this does not appear to have been done. Again, I recommend that the authors revisit the discussion section and support it with the international literature. COVID-19 has been a popular topic and there is a huge amount of literature on the subject.
Author Response
Please see the attachment
Authors: Thank you for your time and energy to review the manuscript and provide recommendations for revisions. The revised manuscript has been uploaded with track changes saved in the document.
Reviewer 1 (R1): You state four countries, but than only US (line 112 ff) Please clarify which are the four countries? live in the US? Or being US citizen? (hence recruiting from other countries too) what kind of software did you use to analyse your data? R? Stata? SPSS? JASP?
Authors: Thank you highlighting this. This article is only focusing on the data collected in the US, which did ask if they lived in the US. SPSS software was used to analyze the data. This has been adjusted in the revised manuscript.
