Examining the Most Insidious Stressor: Systemic Protective Factors and Mental Health Outcomes for Latina/e/x Sexually Expansive Women

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

1.  Line 28.  Since the authors begin with an expansive definition of LGBTQ etc., why narrow it down to only gay and lesbian?  How were the other aspects of sexually expansive persons screened out of the study?  How many were screened out?  Did the invitation for the study only ask for lesbians and gay men to participate?
2.  Line 37.  Either leave out the "a" or add a noun at the end of the sentence.
3.  Line 54 seems confusing.
4.  Line 81.  Since religion is brought up, I would suggest that one tenet of most religions is that you must follow moral ideas or principles rather than emphemeral feelings.  You might have a strong crush on someone in terms of feelings but it could turn out to be a disaster of a relationship.  Most religions might tell all those of varying sexual orientations, that you have to be careful about letting feelings drive the train of your decision-making and behaviors.  My understanding is that youth suicides are often occurring after a relationship break-up (regardless of sexual orientation).  Maybe if all persons were more careful with starting relationships (on more than just feelings), perhaps suicide rates could be lowered for all sexual orientations.
5.  Line 124.  If one were to use social exchange theory rather than just sexual minority theory, certain more interesting questions would be raised.  Social exchange theory says that people make decisions based on perceived rewards and costs/risks seeking overall profit.  There is so much talk about stigma and the negatives of being LGBTQ, what about the positives?  There are articles on the positives.  I can name some.  
Lesbians:  No need to worry about getting pregnant, very unlikely.  If you can't get pregnant, may never have to make choices about abortion or adoption.  Less worry about date rape.  Don't have to try to figure out how men tick or put up with their ignorance about what makes women tick.  Your partner understands female physiology and hot spots far better than a man normally would.  Probably easier to have high quality sex without a super horny man demanding relief now.  More compatibility on how your brains work (given some sex differences in neurology).  Less worry about a man getting violent with you and able to inflict severe damage on you.  You get your monthly cycle; how often does any man?  If you were abused by a man during your childhood, you can spare yourself having to relive any of that.  In other words, with all these positives, why  would microaggressions or social disapproval come close to overcoming all of these advantages or benefits?  Your costs are less, your risks less, your benefits more.  Except that social exchange theory proposes that if you feel overbenefitted, you feel guilty.  Which raises the possibility that internalized homophobia, per social exchange theory, might be caused by felt overbenefit rather than some externally imposed moral guilt.  I am not saying that's IT but just that no one ever seems to have tested it as a possibility, should at least be something mentioned for future research, even if it's a dumb idea, why not spend a bit of time to disprove it?
Gay men:  No need to learn how to approach women without being rejected or walking away in fear.  You can approach gay men having a good sense that they probably want sex, too, whereas with women that's a matter of timing and probability and commitment.  Men understand how much you want sex, it's not something to be despised.  You can have casual or more committed relationships, being open and upfront without offending.  You are not going to get some female pregnant against her wishes and force her to deal with possible abortion or adoption.  If you have sexual capability of having several orgasms an evening, you can find willing partners in men but how often would a woman put up with sex several times an evening (keeping in mind that their testosterone levels are far lower than that)?  Many women want to trade sex for commitment but that is an option with gay men, not a requirement.  The point being is that there can be many rewards, lower costs, lower risks; how could some family rejection overcome all of these profits?  Again, it could be a bad theory, but at least someone ought to examine/test it.
6.  Line 145.  I agree that internalized heterosexism ought to be measured with more precision.  Not just if it exists, but how and why?  There might be several alternative pathways to it, do those pathways matter?  Again, something to consider for future research.  The pathways might differ by gender, too.
7.  Line 173.  Incomplete sentence.
8.  Line.  210.  If long-term relationships can provide protection, is there a chance that short-term relations do not as much?  In other words, if any person (regardless of sexual orientation) has only a series of short-term relations might that lead to frustration, discouragement, self-criticism?  
9.  Line 332.  Did the 120 cases come from the first wave or from all three waves?  
10.  If the age ranges were 18-25 and 34-41, would that not leave a gap from ages 26 to 33?  Is age in your sample distributed randomly or not?
11.  If nearly half of the Generations Study was bisexual and since bisexuality is more common among women than men, did selecting out bisexuality bias the sample of women more than it did men?
12.  Line 340.  You did not mention the percent of the sample involving gay men?  Please do. 
13.  Line 344.  The word "and" is confusing.  Did you mean that 32% earned less than $25K and in addition 46% were employed full-time or did you mean that 46% of the 32% were employed full-time?
14.  As another idea from social exchange theory, if a person wishes they were not LGBTQ, what are the barriers, real or perceived, and alternatives, seen and unseen, impacting their internal thinking?  I am not sure we know very much about this, another area for future research.
15.  Line 380.  I don't recall seeing anything about Cronbach's alpha for the scales used or evidence for their discriminant or construct validity.  Please provide.
16.  Why not include correlation tables for the relations of the demographic variables with each other and with the variables in Table 1?
17.  If age seems important, is there a possible issue with the unusual age distribution (if it exists) and how age is working in your sample?
18.  Tables 2 and 3 are confusing.  Usually a oneway ANOVA table has the means/SDs for the dependent variable across the levels of the independent variable, even if the results are not significant.  In Table 2, the dependent variable is, I presume, internalized heterosexism but the only variable shown is sexual orientation but not in two levels (gay, lesbian).  Yet the F test shows 7 levels with F(6, 115), which doesn't add up since that would suggest 6 + 115 + 1 (or two) subjects, at least 122 when your sample only has 120 subjects???
19.  In Table 3, the degrees of freedom in the F test look more reasonable.  But the table remains very incomplete.
20.  Table 4.  If you are predicting internalized heterosexism from EIC then the beta should be the same (when using only two variables) as the correlation but the correlation in Table 1 is -.19 but the beta in Table 4 is -.019.  Please advise.
21.  In Table 5, same issue, in Table 1 the relation is -.29 but in Table 5 it's -.28.  Please advise.
22.  In Table 6, the numbers match with Table 1.
23.  In Table 7, the R squared decreases when you add more predictors?  The adjusted R squared might do that but not the raw R squared.
24.  In Tables 8 and 9, the R squared are increasing or staying the same as more variables are added, which is what would be expected.
25.  Line 443 and Line 469 differ on the SIC and IH, which may account for issue in #21 above.
26.  Line 473 and Table 1 agree but disagree with results in Table 4.
27.  Line 483 has -.17 but Table 1 has -.18, please account for the discrepancy.
28.  In Table 1 you had eight possible correlations with IH and only two were significant, and not significant with the two outcomes.  If IH was as pervasive as theorized, would you not expect stronger patterns?
29.  Running oneway anovas, correlations, and two variable regression models are variations on the same underlying statistical tests, note how the significance levels are nearly all the same.  The parsimony of reporting only correlations would make more sense than reporting redundant results.
30.  Line 569.  The meaning of this result depends on the size of the correlation, which isn't reported.  That's why we need a table reporting all the correlations between the demographic variables with each other and with and the variables in Table 1.  Along with significance levels, of course.
31.  Line 692.  I must have missed how you compared IH with behaviors.  Please advise.
32.  Line 697.  You suggest erasure of non-binary categories but did you not select out bisexuals and others yourself?  You are complaining as it were about the inadequacy of your own sample selection methods; if such erasure were so vital, why did you ignore it to begin with?

Author Response

Please see the attachment 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This study explored the mental health outcomes for sexually expansive Latina/e/x women, who are routinely underrepresented in minority stress research. Findings indicate that identity affirmation and social support provide protection and positive support for one’s state of mental health. The following comments are provided to further improve the strength and quality of this study:

Introduction

• Provide a brief definition/explanation for the meaning of Latina/e/x as an identity, as not all readers may be familiar, particularly in how this relates to LGB/sexual expansive identities
• The introduction section is rather long and could benefit from being more succinct. In its current state it reads more like a dissertation versus a research manuscript. Please review and edit where able to shorten the length of the introduction/literature review so that it is more direct and provides the theoretical foundation for this study
• Consider moving certain sections at the beginning of the introduction section (start at line 38/39) to the end of the introduction section, as there are sections here that are more appropriate to discuss the purpose of the study after providing the literature background

Methods

• To clarify, if data was collected from LGB individuals, does this mean that transgender and/or queer individuals were not included in initial data collection? Please clarify
• Line 333-334, “As well as createdmeasures and reliability statistics over .80 threshold for social sciences.” This appears to be out of place. Please clarify or delete as appropriate.
• Methods and Results need to be clearly differentiated from one another. Separate data here accordingly and have a distinct Methods section that only discusses the methods used for data collection and analysis, and a separate Results section that discusses findings and outcomes of data analysis.

Discussion and Conclusion

• Much of the information found in this section almost reads more like an expansion to the literature review. Please take more time to specifically discuss and reflect upon the results of your data analysis versus discussion at the more theoretical level
• The conclusion section provides no final conclusions but instead reads more like a limitations section. Have a separate Limitations subsection to discuss study limitations, and do not include discussions about study limitations in the Conclusion subsection.
• Please write a succinct conclusion section that summarizes the key findings and applications of this study. What are the ways in which study findings can be practically applied to those working with sexually expansive Latina/e/x people?

Misc

• Please review the entire manuscript again for sentence structure, punctuation, and flow of language, as there are several instances throughout the document with some errors in these areas
• Throughout the document, be consistent in your use of LGB, LGBT, LGBTQ, LGBTQIA+, LGBTQ+, etc, as there are instances where different forms of the acronym are used, and you want to remain consistent

Author Response

Please see attached 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This is a capably conducted study that can benefit from attention to the following. 

1. The abstract should define "sexually expansive." References to mental health in the abstract should indicate improved or diminished mental health (valence).  
2. Since this is a cross-sectional study, a good case can be made for eliminating causal language. The correlation matrix (Table 1) is appreciated, but a cross-sectional survey does not permit causal order to be determined. So, x predicting y could become x is associated with y. The word effect/effects should be reconsidered. The language of "mental health outcomes" could become "mental health factors" or, more simply, "mental health." I am not a purist where this consideration is concerned, and leave it to the authors and editors to make this determination. But it is more scientifically accurate to speak of associations or correlations among variables in data of this sort. At the very least, some nod to causal order assumptions and the limitations associated with these assumptions needs to be featured.  
3. More effort could be made to establish statistical power despite this stated limitation in the Conclusion. Please provide assurances that the sample size has sufficient statistical power for the full range of analyses that are conducted. Were power analyses conducted? Also, the reference to "the sample of (n = 120)" in line 393 does not need parentheses, in my view. It could be: "the sample (n) of 120..."  
4. The treatment of the chronosystem should be accompanied by a statement that these data are not longitudinal. The study does not address change over time. 
5. Since qualitative data were also collected, it seems reasonable to mention how such data could be used in a follow-up study to extend what's learned here. This point could be integrated into the Conclusion in more than the single sentence that is currently there. 

Author Response

Please see attachment 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript is substantially improved, thanks to the authors!

At line 357, should not the beta be -0.286?

I have no further suggestions.

Author Response

Comment 1: 

This manuscript is substantially improved, thanks to the authors!

At line 357, should not the beta be -0.286?

I have no further suggestions.

Response 1: 

I sincerely appreciate this feedback and thorough review of the results! The inverse association has been updated to adhere to the correct sign. 

Reviewer 2 Report

Comments and Suggestions for Authors

I thank the authors for all of the edits and changes that they have made to this manuscript. I have no further suggested edits or changes to recommend at this time. 

Author Response

Comment 1: 

I thank the authors for all of the edits and changes that they have made to this manuscript. I have no further suggested edits or changes to recommend at this time. 

Response 1: 

I sincerely appreciate this feedback and thank the reviewer for their time! 

Reviewer 3 Report

Comments and Suggestions for Authors

I commend the author on a sound revision. I have no further revisions to recommend and appreciate the attention given to the prior comments. Well done!  

Author Response

Comment 1: 

I commend the author on a sound revision. I have no further revisions to recommend and appreciate the attention given to the prior comments. Well done!  

Response 1:

I sincerely appreciate this feedback and thank the reviewer for their time!