“I Don’t Approve of a Fat Person…”: A Cross-Sectional Survey Exploring the Perceptions of Health, Weight and Obesity
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsOverall, this is a well composed project and manuscript that warrants publication, but not before resolving some issues. Below is a list of comments related to the manuscript that need to be addressed or resolved.
Major Comments:
Results - Table 2 and Table 4: Please change the column “percent reported” to “percent of respondents”, such as in Table 1. It would be more helpful to know the percentage of participants who answered instead of the percentage of responses.
Results - Table 4: Some themes (e.g., “Support” and “Hobbies”) described in Table 4 have so few responses that they cannot be counted as themes. While I am unsure of the exact frequency needed to establish a theme, having less than 1% of total responses does not seem to be a great enough rate to justify it as a theme. Please change or comment on this.
Discussion – Lines 331-337: In this paragraph, research is presented that women are more likely to internalize weight stigma compared to men, but there was no difference in internalized stigma between genders in your results (i.e., lines #242 - #245). This incongruity needs to be discussed in this section within the discussion.
Minor Comments:
Fron Matter - Title: The title uses a list, but no “Oxford comma” or comma that occurs at the end of the list. In the rest of the work, the “Oxford comma” is used, for instance in lines: #37, #75, and #82. Please change the title to: “I don’t approve of a fat person…”: A cross sectional survey exploring the perceptions of health, weight, and obesity.
Results - Table 3: The legend/key at the bottom of Table 3 is incorrect. So called “double daggers” are used for p <= 0.001, but “***” is listed in the legend.
Results – Line #236-240: Report of increased externalized stigma and negative judgment in males contradicts the fact that males scored higher for attraction to those with obesity. Are you suggesting that perceptions of males with obesity received greater externalized stigma and negative judgment, but that male participants reported higher levels of attraction to those with obesity? These lines are unclear and need to be rephrased, as I am not sure if it is discussing the views of male participants or the perception of males with obesity.
Results and Discussion: Please change “sex” to “gender”, as sex was not assessed but rather gender was. Male should be changed to men and female to women throughout results and discussion.
Author Response
We thank the reviewer for taking the time to review our submission. We have addressed the comments raised in an itemised fashion below.
Comment 1: Results - Table 2 and Table 4: Please change the column “percent reported” to “percent of respondents”, such as in Table 1. It would be more helpful to know the percentage of participants who answered instead of the percentage of responses.
Response 1: We agree with the reviewer that reporting percent of respondents is a more helpful metric, and have amended Table 2 and 4 to now report this. We have also amended percentages reported in-text to align with this.
Comment 2: Results - Table 4: Some themes (e.g., “Support” and “Hobbies”) described in Table 4 have so few responses that they cannot be counted as themes. While I am unsure of the exact frequency needed to establish a theme, having less than 1% of total responses does not seem to be a great enough rate to justify it as a theme. Please change or comment on this.
Response 2: These are not themes, but codes identified during thematic analysis. The purpose here is not to try and identify overarching themes (these are captured in the table headers), but instead the different factors that participants reported. To make this clearer, we have amended to table capture to read: "Codes identified during thematic analysis of responses outlining the individual, societal and environmental factors associated with good and poor health (n [%])."
Comment 3: Discussion – Lines 331-337: In this paragraph, research is presented that women are more likely to internalize weight stigma compared to men, but there was no difference in internalized stigma between genders in your results (i.e., lines #242 - #245). This incongruity needs to be discussed in this section within the discussion.
Response 3: We have amended this paragraph to acknowledge the findings, which now reads:
"The present work observed significantly higher weight bias in women, compared with men. Given the increased societal pressures placed on appearance in women [67,68], and the need to adhere with the ‘thin ideal’ [17,18], it is unsurprising such bias exists and is observed in the present sample. In addition, women are more vulnerable to weight stigma [69-71], as are those with higher weight status [24,49,72]. Individuals who more frequently experience weight stigma are more likely to internalise such stigma, endorsing the negative stereotypes and attributing them to themselves [73,74]. This, however, was not shown in the present sample with no significant difference in internalised weight stigma scores. Interestingly, externalised weight stigma was higher in men than women. This suggests that while women appear to hold more negative views, men are more likely to enact harmful social stereotypes associated with weight."
Comment 4: Fron Matter - Title: The title uses a list, but no “Oxford comma” or comma that occurs at the end of the list. In the rest of the work, the “Oxford comma” is used, for instance in lines: #37, #75, and #82. Please change the title to: “I don’t approve of a fat person…”: A cross sectional survey exploring the perceptions of health, weight, and obesity.
Response 4: The title has been updated as advised.
Comment 5: Results - Table 3: The legend/key at the bottom of Table 3 is incorrect. So called “double daggers” are used for p <= 0.001, but “***” is listed in the legend.
Response 5: We thank the reviewers for identifying this error. The key has now been amended to reflect the correct symbols used in the table.
Comment 6: Results – Line #236-240: Report of increased externalized stigma and negative judgment in males contradicts the fact that males scored higher for attraction to those with obesity. Are you suggesting that perceptions of males with obesity received greater externalized stigma and negative judgment, but that male participants reported higher levels of attraction to those with obesity? These lines are unclear and need to be rephrased, as I am not sure if it is discussing the views of male participants or the perception of males with obesity.
Response 6: We apologise, this was a misrepresentation of the data. For this questionnaire (UMB-FAT), higher scores represent more stigmatising views. Rather than males viewing individuals living with obesity as more attractive (as suggested in the sentence), the source of their increased externalised stigma is due to a significantly higher scores for attraction which indicates they find individuals living with obesity less attractive than females. The sentences has been amended to reflect this, and now reads: "Externalised stigma was more prevalent in men (3.8 AU, IQR 0.5 AU) than women (3.6 AU, IQR 0.3 AU) (t = 3.32, p = 0.001). Specifically, men enacted more negative judgement of individuals living with obesity (z = −2.90, p = 0.004) and viewed individuals living with obesity as less attractive (5.2 AU, IQR 2.0 AU) compared with women (4.1 AU, IQR 2.0 AU) (z = −2.90, p = 0.004)."
Comment 7: Results and Discussion: Please change “sex” to “gender”, as sex was not assessed but rather gender was. Male should be changed to men and female to women throughout results and discussion.
Response 7: We could not find any use of the word 'sex' in our manuscript. We have changes male/s to man/men and female/s to woman/women throughout.
Reviewer 2 Report
Comments and Suggestions for Authors- Lines 100-101, please provide a detailed justification for the sample size estimation, rather than simply stating, “in line with other similar studies.” In particular, it should be based on the statistical analysis or the representativeness of the sample. For example, is it convincing to use 133 responses to understand the views of the population?
- Line 105, what is the rationale for asking about perceived weight status only? Although perceived weight status is a significant factor in weight bias studies, why not examine both objective and subjective weight status? It would be better to address this in the Introduction. For instance, is it because of the research gap, or what is known about the differences between objective and subjective weight status? For your information:
https://doi.org/10.1002/oby.22041
https://doi.org/10.1097/NMD.0000000000000869
- This study aimed to explore the views and perceptions of weight and health. Are there any hypotheses or specific research questions to guide your data analysis? Additionally, the research gap was not clearly articulated.
- From the introduction, it appears that the study was going to explore public weight stigma. However, self-stigma was also examined in this study. Please discuss this in the Introduction.
- For the Discussion, it would be better to discuss the implications and significance of the study. Specifically, what additional information has been found compared to the existing literature?
Author Response
We thank the reviewer for taking the time to review our submission. We have addressed the comments raised in an itemised fashion below.
Comment 1: Lines 100-101, please provide a detailed justification for the sample size estimation, rather than simply stating, “in line with other similar studies.” In particular, it should be based on the statistical analysis or the representativeness of the sample. For example, is it convincing to use 133 responses to understand the views of the population?
Response 1: An a priori sample size calculation was not conducted for this study. Although it would have been ideal to establish a minimum sample size, this was not possible at the start of the study. Our sample size was based on observed minimum samples in other published work (as described in-text). It is not appropriate to retrospectively fit a sample size calculation. We have, however, estimated the achieved power for our main analyses using G* Power. While some are low, many of our analyses demonstrate satisfactory achieved power (>0.816, e.g., UMB-FAT total, UMB-FAT Attraction).
Comment 2: Line 105, what is the rationale for asking about perceived weight status only? Although perceived weight status is a significant factor in weight bias studies, why not examine both objective and subjective weight status? It would be better to address this in the Introduction. For instance, is it because of the research gap, or what is known about the differences between objective and subjective weight status? For your information:
https://doi.org/10.1002/oby.22041
https://doi.org/10.1097/NMD.0000000000000869
Response 2: The purpose of this work was not to establish difference in perceptions of health, weight and obesity based on objective/subjective weight status. It is not possible to measure weight status objectively through a cross-sectional survey design. Additionally, the validity of subjective (self-reported) weight has previously been questioned (e.g., https://doi.org/10.1111/j.1467-842X.2000.tb00733.x). Here, we are more interested in perceptions - including the participants' perception of their own weight; i.e., it matters less what their actual weight status is given perceived body image is more indicative of bias/stigmatising views - as the reviewer highlights.
Comment 3: This study aimed to explore the views and perceptions of weight and health. Are there any hypotheses or specific research questions to guide your data analysis? Additionally, the research gap was not clearly articulated.
Response 3: No hypotheses were generated a priori. We did, however, have three specific questions we were interested in addressing, which have now been incorporated into the introduction and read: "Specifically, this project looked to address the following questions: (i) How do individuals perceive ‘health’, and what factors contribute to being ‘healthy’? (ii) Are there specific demographic characteristics associated with weight-based stigmatising views? (iii) Does prior experience of weight bias/stigma affect internalised/externalised stigmatising views?". We have also worked to more clearly articulate the novelty of this work.
Comment 4: From the introduction, it appears that the study was going to explore public weight stigma. However, self-stigma was also examined in this study. Please discuss this in the Introduction.
Response 4: We have included consideration for internalised weight stigma in the introduction, which now reads: "Internalisation of weight stigma, where societal negative stereotypes are applied to oneself contributing to mistreatment or devaluation of self [28], has a mediating role on an individuals’ body image, health-related behaviour (e.g., physical activity), engagement with and experience of healthcare [29]."
Comment 5: For the Discussion, it would be better to discuss the implications and significance of the study. Specifically, what additional information has been found compared to the existing literature?
Response 5: We have mentioned in the discussion where our findings support or refute prior evidence (i.e., how this work builds on previous studies). For example:
"Today health is seen more as a holistic concept, incorporating physical, mental and social wellbeing and not just the absence of disease or infirmity, with some definitions associating health primarily with good fitness [48]. Our findings support this connection, with comments in response to those of healthy weight focussing on physical, mental and social wellbeing."
"Individuals who more frequently experience weight stigma are more likely to internalise such stigma, endorsing the negative stereotypes and attributing them to themselves [73,74]. This, however, was not shown in the present sample with no significant difference in internalised weight stigma scores. Interestingly, externalised weight stigma was higher in men than women. This suggests that while women appear to hold more negative views, men are more likely to enact harmful social stereotypes associated with weight."
"Both diet and physical activity were deemed as important contributors to health – contributing to both good and poor health. Dryer and Ware [75] observed ‘eating more food than you need’, ‘eating the wrong types of food’ and ‘not enough physical activity’ as factors contributing to weight gain. This, again, demonstrates the parallels drawn between weight and health."
We feel the current discussion sufficiently details the way in which our work contributes to the current evidence based, but we are happy to take editorial advise should the reviewer have specific areas they feel would benefit from more explicit description of contribution.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors resolved my comments from the first review. I have no further comments.
Author Response
We thank the reviewer for taking the time to review our submission.
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for the revision. As the authors mentioned that they estimated the achieved power for their main analyses, it would be better to include this information in the manuscript.
Author Response
Comment: Thank you for the revision. As the authors mentioned that they estimated the achieved power for their main analyses, it would be better to include this information in the manuscript.
Response: We have amended the limitations section, where we previously discussed sample size, to include the following statement, "Finally, while the sample size was modest in comparison to other published work. However, a posteriori power calculations suggest that while some analyses have low achieved power, many demonstrate satisfactory achieved power (>0.816, e.g., UMB-FAT total, UMB-FAT Attraction)."
