The Role of Breast Morphology in Women’s Rival Derogation Tactics
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsI have read the manuscript “The role of breast morphology on women’s rival derogation tactics” by Garza and Pazhoohi. I believe the manuscript is publishable, but not in its present form. The manuscript is full of typographic errors that need to be addressed. Otherwise, the methodology and the statistical analysis support the authors conclusions that breast size and not ptosis affect women’s non-physical aggression. The one further suggestion I would make is to potentially soften this conclusion as both of their aggression measures are based on single questions. While these questions are valid, taken from apparently valid and reliable scales, they are still only single question indicators. Possibly an acknowledgement of this and a suggestion for future work increasing the number of items from other sources.
Comments on the Quality of English LanguageThe typographic errors need to be addressed.
Author Response
I have read the manuscript “The role of breast morphology on women’s rival derogation tactics” by Garza and Pazhoohi. I believe the manuscript is publishable, but not in its present form. The manuscript is full of typographic errors that need to be addressed. Otherwise, the methodology and the statistical analysis support the authors conclusions that breast size and not ptosis affect women’s non-physical aggression. The one further suggestion I would make is to potentially soften this conclusion as both of their aggression measures are based on single questions. While these questions are valid, taken from apparently valid and reliable scales, they are still only single question indicators. Possibly an acknowledgement of this and a suggestion for future work increasing the number of items from other sources.
Response: We have assessed all typographical errors throughout. We have also included the following in the limitation section to address the reviewer’s concern, and it now reads, “The use of a single item to measure verbal and indirect aggression is a limitation. Using multiple measures for these constructs in future studies could add to the validity that aggressive behaviors are being measured accurately. The findings represent a single study of women from a university sample; therefore, future studies could expand by using a more diverse sample or implementing a cross-cultural comparison. One approach that could be beneficial to enhance the significance of the findings is to follow an approach similar to Widman et al. [36], where measures of physiological activity can be used to gauge women’s level of competition and vigilance. Perhaps, future work could incor-porate an eye-tracking method that assesses how women are vigilant to women’s breast morphology and if visual attention also correlates with measure of verbal and indirect aggression.”
The typographic errors need to be addressed.
Response: We have combed through the manuscript and fixed these errors. For instance, on line 61 of page 2, we mistakenly included a statement beginning with “Women men…”, and it was supposed to read, “Men place a premium…”. Also, we noticed that some of the numerical references did not include a closing bracket “[# “ , and they have been corrected to appear as
“[# ]”.
Reviewer 2 Report
Comments and Suggestions for Authors
The current investigation relied on a single online study to examine whether Hispanic college women were more likely to self-report greater frequency of derogating female targets with larger and less saggy breasts. The authors found that college women reported a greater frequency of derogating targets with larger breasts. They did not find an effect of women’s dispositional intrasexual competitiveness on derogation.
Although the logic underlying the study is sound, I have some concerns with the methods.
The study has numerous limitations, which, at minimum, should be articulated in the discussion. For example, the use of a single (small) sample, an online survey, self-report measures, and single-item measures are each problematic.
Due to the small sample, I am not convinced that this study was sufficiently powered. Indeed, some of the findings were trending in the predicted direction. The authors should provide post-hoc sensitivity analyses to assess whether they were sufficiently powered to detect these effects.
Did the authors use any attention checks to ensure data quality?
In the abstract, the authors describe verbal aggression as an indirect aggressive tactic. However, as worded, the measure describes yelling at the female target or saying something hurtful to her. These are direct forms of aggression because they are directed at the target and should not be described as indirect. I have further concerns with this operationalization because verbal aggression is a slightly more male-typical form of aggression (see e.g., Archer, 2004). Thus, I worry that these self-reports might not actually represent women’s likely behavioral responses.
It seems to me that the response scales of the aggression items (never to very often) do not really make sense based on the questions’ wording (“how likely are you”). Moreover, frequency of aggression is not really the construct of interest. One disclosure might be sufficient to spread a rumor that harms a woman’s reputation. Thus, willingness is perhaps more meaningful than frequency.
The authors’ takeaways from their non-significant findings should be tempered in light of the study’s limitations. For example, “However the findings from the current study diverge from this hypothesis, suggesting that women do not necessarily resort to aggression, either verbally or indirect, against others with non-ptotic breasts…This indicates that breast size, rather than ptosis, may be a more critical factor in signaling a woman's reproductive potent.” These conclusions seem very premature given that this was a single, likely underpowered, self-report online study.
Please clarify in section 2.2.3 whether the two aggression items were combined to form a composite or analyzed separately.
The abstract should describe the sample and that the study used an online survey to provide context for readers.
Please clarify in the text whether verbal aggressive responses to the C-cup significantly differed from the other sizes.
The authors describe their sample as Hispanic. Were all female participants Hispanic? Please provide frequencies.
Author Response
The current investigation relied on a single online study to examine whether Hispanic college women were more likely to self-report greater frequency of derogating female targets with larger and less saggy breasts. The authors found that college women reported a greater frequency of derogating targets with larger breasts. They did not find an effect of women’s dispositional intrasexual competitiveness on derogation.
Although the logic underlying the study is sound, I have some concerns with the methods.
Response: We thank the reviewer for providing these helpful comments, and we know their suggestions and comments have significantly improved the manuscript. Below, we address each point.
The study has numerous limitations, which, at minimum, should be articulated in the discussion. For example, the use of a single (small) sample, an online survey, self-report measures, and single-item measures are each problematic.
Response: We have included the g*power analysis for a-priori sample sizes for power. The power analysis indicated that 71 participants would be needed to detect a small to medium effect. We have included this in the participants section, and it now reads: “A G*Power analysis for a 4(breast: A-, B, C-, & D-cup) x 3(ptosis: non, low, & high) repeated measures design indicated that 71 participants were needed to detect a small to medium effect (f = .10, 80% power).” Nonetheless, we have included in the limitations section that the use of a single study is a limitation.
Due to the small sample, I am not convinced that this study was sufficiently powered. Indeed, some of the findings were trending in the predicted direction. The authors should provide post-hoc sensitivity analyses to assess whether they were sufficiently powered to detect these effects.
Response: We disagree that the sample size was insufficient. This study was a 2-way repeated measures design and did not include any group comparisons. We have now included the power analysis for further clarification in the participant’s section. Further, by running a post-hoc sensitivity analysis, it determined that we had enough power given the sample size (Power = 96%). We are confident that the study was sufficiently powered, and the findings are robust. We included the a-priori sample size estimation in the participants section, but if the reviewer feels more comfortable with us including the sensitivity post-hoc analysis, we do not object to including that as well.
Did the authors use any attention checks to ensure data quality?
Response: The study did not use any attention checks, as we did not have a very long survey nor did we launch the survey using samples that were out of our reach (e.g., mturk samples). We used a university sample, and we also shortened the breast image set from 36 to 12 to reduce the possibility of participants not taking the survey seriously. To add, we included an honesty question to be able to determine if participants took the survey seriously, and they did. The question asked, “On a scale from "1= not honest at all" to "9 = very honest", how honest were you in providing ratings to the images shown”, and the participants reported a high degree of honesty (M = 8.53, SD = 1.27) in their ratings.
In the abstract, the authors describe verbal aggression as an indirect aggressive tactic. However, as worded, the measure describes yelling at the female target or saying something hurtful to her. These are direct forms of aggression because they are directed at the target and should not be described as indirect. I have further concerns with this operationalization because verbal aggression is a slightly more male-typical form of aggression (see e.g., Archer, 2004). Thus, I worry that these self-reports might not actually represent women’s likely behavioral responses.
Response: We apologize for this oversight, and the reviewer is correct. We have rephrased the abstract to report on rival derogation tactics to mean both verbal and indirect forms, rather than including verbal as an indirect form of aggression. This has also been corrected throughout the manuscript. It now reads, “
Abstract: Physical features that are desired by the opposite sex may drive competition between members of the same sex to gain access to potential mates. Women’s breasts are considered sexually attractive to men, and it has been shown that women may engage in competitive tactics to compete or derogate women with ideal physical traits (i.e., physical attractive features). In the current online study, we investigated Hispanic women’s (N = 114) perceptions of breast stimuli that had been manipulated to display 4 levels of breast size (A-, B-, C-, and D-cup) and 3 levels of ptosis (i.e., levels of sagginess: non, low, and high) and their likelihood of engaging in rival derogation tactics, such as verbal and indirect aggression. The findings demonstrated that women were more likely to verbally and indirectly aggress towards women with larger breast sizes. Women’s dispositional level of intrasexual competition did not play a role in derogation tactics. The results are in line with previous research suggesting that women’s derogation tactics are likely going to be targeted towards women with attractive features that are desired by men.”
It seems to me that the response scales of the aggression items (never to very often) do not really make sense based on the questions’ wording (“how likely are you”). Moreover, frequency of aggression is not really the construct of interest. One disclosure might be sufficient to spread a rumor that harms a woman’s reputation. Thus, willingness is perhaps more meaningful than frequency.
Response: We apologized for this oversight, as we were looking at the original Mini-DIAS response options, but the way that our questions were asked were a modified version of the Mini-DIAS, where the response options were in line with the question using “likelihood”. We have updated the section 2.2.3 to include the following,
“The dependent variables were 2 questions from the Mini-DIAS [32]. These questions measure one’s propensity to engage in verbal and indirect aggression. Although the complete Min-DIAS includes a question on physical aggression (i.e., likelihood of kicking, shoving), most research has pointed to men engaging in more physical confrontation [4]; therefore, we used the verbal and indirect items. We modified the Mini-DIAS to represent participants’ “likelihood” of engaging in the behaviors rather than the original response that asks “how often”. For the verbal aggression question, participants were asked, “How likely are you to be verbally aggressive against the woman? (For example: yell at her, call her bad names, or say something hurtful to her). For the indirect aggression question, participants were asked, “How likely are you to be indirectly aggressive against the woman? (For example: gossip maliciously about her, spread harmful rumors about her, or try to socially exclude her?). The responses for verbal and indirect aggression were on a 7-point scale, where options varied from “ 1 = not very likely” to “7 = very likely”. For this study, we used each item to represent a verbal and indirect aggression score.”
The authors’ takeaways from their non-significant findings should be tempered in light of the study’s limitations. For example, “However the findings from the current study diverge from this hypothesis, suggesting that women do not necessarily resort to aggression, either verbally or indirect, against others with non-ptotic breasts…This indicates that breast size, rather than ptosis, may be a more critical factor in signaling a woman's reproductive potent.” These conclusions seem very premature given that this was a single, likely underpowered, self-report online study.
Response: We have removed the following statement to not sound premature or confusing in our findings, “This indicates that breast size, rather than ptosis, may be a more critical factor in signaling a woman's reproductive potential”. However, our sample is sufficiently powered to be able to detect effects based on the repeated measures nature of the design.
Please clarify in section 2.2.3 whether the two aggression items were combined to form a composite or analyzed separately.
Response: In this section, we have included the following, “For this study, verbal and indirect aggression were assessed separately rather than forming a composite variable.”
The abstract should describe the sample and that the study used an online survey to provide context for readers.
Response: We have clarified this statement in the abstract, and it now reads, “In the current online study, we investigated Hispanic women’s (N = 114) perceptions of breast stimuli that had been manipulated to display 4 levels of breast size (A-, B-, C-, and D-cup) and 3 levels of ptosis (i.e., levels of sagginess: non, low, and high) and their likelihood of engaging in rival derogation tactics, such as verbal aggression and gossiping.”
Please clarify in the text whether verbal aggressive responses to the C-cup significantly differed from the other sizes.
Response: We have included in verbal aggression writeup the following, “Further, women were more likely to verbally aggress towards women with C-cup breast sizes compared to B-cup breast sizes (p = .03), but the difference was not significant compared to A-cup breast sizes (p = .40).”, and for indirect aggression, “Women were more likely to indirectly aggress towards women with C-cup breast sizes compared to B-cup breast sizes (p = .04), but the difference was not significant compared to A-cup breast sizes (p = .73).”
The authors describe their sample as Hispanic. Were all female participants Hispanic? Please provide frequencies.
Response: We have included the following in our participants section, “One-hundred and fourteen predominantly Hispanic women (M = 23.82, SD = 5.51) from Texas A&M International University participated in this online study in exchange for course credit. Women in the study were self-identified heterosexuals. In our sample, 45% of women reported being single and 71% reported being in a relationship. The sample demographics consisted of Hispanic (N = 111) and White (N = 3) participants.”
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsI appreciate that the authors have been responsive to my feedback. The manuscript is now stronger.
Author Response
I appreciate that the authors have been responsive to my feedback. The manuscript is now stronger.
Response: We thank the reviewer for this positive comment. We are glad we answered everything accordingly.
