Filling the Gap: Aesthetic Preferences and Demographic Variations in Thigh-Gap Proportions Among the General Population

Abstract

Background

Driven by the influence of social media platforms, the “thigh gap” has emerged as a beauty ideal in modern culture. Despite its prevalence in popular culture, there is no scientific evidence to validate the thigh gap as a realistic or universally attainable beauty standard.

Objectives

The aim of the authors of this study is to evaluate the general public's perception of thigh-gap aesthetics, specifically examining preferences for varying thigh-gap widths and the influence of demographic factors on these preferences.

Methods

Four model images of varying thigh-gap widths and thigh gap-to-hip ratios (TGHRs) of 0.19, 0.09, 0.04, and 0.02 were created and presented to participants in an online survey using public crowdsourcing. Participants rated the attractiveness of these images and ranked them in order of preference. Demographic data, including gender, race, sexual orientation, and income, were collected to assess subgroup trends.

Results

A TGHR of 0.09 was rated as the most attractive by the general population, with preferences for moderate thigh gaps consistently emerging across most demographic groups. Significant variations in preferences were observed based on income, race, and sexual orientation, indicating the role of cultural and socioeconomic factors in shaping aesthetic ideals. Extreme thigh gaps and the absence of a thigh gap were rated as less attractive.

Conclusions

These findings suggest a general preference for moderate thigh gaps with a TGHR of 0.09 over extreme or absent thigh gaps. Our findings highlight the importance of understanding demographic differences in aesthetic preferences to tailor body contouring and reconstructive procedures that meet patient expectations while maintaining natural proportions.

Level of Evidence: 5 (Therapeutic)

The “thigh gap” is defined as the visible space between the inner thighs when standing with feet together. In recent years, the thigh gap has emerged as a beauty ideal in modern culture, largely driven by the pervasive influence of social media platforms. Curated images and videos portraying the thigh gap as an aspirational standard of physical beauty are commonly found across major social media platforms.^1,2

Although its widespread promotion on social media has made the thigh gap a hallmark of contemporary beauty standards, it remains a culturally constructed ideal with significant implications for mental and physical health.^1,3,4 Achieving a thigh gap is unattainable for many people because it is determined by anatomical factors such as pelvic width, shape, and muscle structure rather than BMI.^5,6 The desire to obtain a thigh gap through diet and exercise has been linked to adverse effects such as malnutrition, overtraining, and the predisposition to anorexia nervosa and body dysmorphic disorder.^3,7,8 The trend has also been linked to “thinspiration” and “fitspiration” subcultures, which promote unattainable beauty standards that often demand extreme caloric restriction, intense training regimens, and unhealthy comparisons to others.^9,10

Despite its prevalence in popular culture, there is no scientific evidence to validate the thigh gap as a realistic or universally attainable beauty standard. The purpose of the authors of this study is to fill this gap by assessing the general public's conceptions and choices regarding the aesthetics of the thigh gap. To better understand the public's conceptions, we included a range of demographic variables in our analysis: age, gender, income, sexual orientation, and race. These factors were selected based on their documented influence on beauty norms.^11-13

METHODS

After obtaining UT Southwestern Institutional Review Board approval, we conducted our study to evaluate aesthetic preferences of the female thigh gap as influenced by individual demographics. In order to create natural-looking female thigh gaps with an appreciable degree of variation, a model female patient photograph was purchased from 3D Scan Store (Ten24 Media LTD, Sheffield, United Kingdom) to serve as the base for photographic manipulation. The model was middle-aged Caucasian and had a normal BMI of 20 to 23. This photograph was then contoured using Adobe Photoshop (Version CS6, Adobe, San Jose, CA) to create 3 additional images that varied in the width of their thigh gap (Figure 1). The thigh gap was measured between the medial thighs at the midpoint of the distance between the anterior superior iliac spine (ASIS) and the mid-lateral knee. To standardize these measurements, we utilized the thigh gap-to-hip ratio (TGHR), which was calculated by dividing the thigh gap by the hip width at the level of the ASIS.² Following photographic manipulation of the female model, the senior authors selected photographs that they perceived accurately represented distinct thigh habitus, each with sufficient variation in TGHR to allow raters to discern meaningful differences. The photographs were subsequently screened by study and nonstudy personnel, including multiple plastic surgeons in our department, to ensure an appreciable difference in thigh gap was achieved while maintaining a degree of realism in the images. Four images were eventually selected with ratios of 0.19, 0.09, 0.04, and 0.02 from A to D, respectively (Figure 1).

Figure 1.

(A-D) Model images with varying thigh gap widths and thigh gap-to-hip ratios (TGHR) utilized in the survey. The thigh gap was measured between the medial thighs at the midpoint of the distance between the anterior superior iliac spine (ASIS) and the mid-lateral knee. The TGHR was calculated by dividing this distance by the hip width at the level of the ASIS. The red line, labeled H, indicates the hip width measured at the level of the ASIS. The green lines, labeled TG, indicate the thigh-gap widths.

A survey was created using RedCap to gather public preferences for variations in thigh-gap width. The survey was uploaded to Amazon Mechanical Turk (MTurk), an online crowdsourcing platform operated by Amazon Web Services (Amazon, Seattle, WA). MTurk enables access to large, diverse participant pools and is frequently utilized in aesthetics research for its efficiency and scalability.^14,15 Raters were tasked with a rank-choice vote. All 4 images were displayed on a panel so that raters could see all photographs at once, with each image assigned a letter from A to D. Participants then rated the attractiveness of these images and ranked them in order of preference. The survey also collected demographic information, including age, gender, ethnicity, sexual orientation, and average income, to assess trends across subgroups. The survey was available for respondents on the MTurk platform from November 28, 2022, to December 12, 2022.

Responses were included in the analysis if raters provided the correct submission code generated upon completion. To receive the submission code, raters were required to fill out all demographic information and complete the voting matrix. Raters with the incorrect submission code had their responses omitted; 6 raters did not correctly fill out the survey. Responses were analyzed to determine preferences regarding thigh-gap aesthetics and their correlation with demographic variables.

Statistical Analysis

Using Python 3.11, a series of χ² tests of independence were conducted for each rating category to assess differences in image preference across demographic groups. For any table containing “0” or where expected frequencies fell below the threshold for valid asymptotic inference (<5), we applied a Monte Carlo simulation (10,000 replicates) to obtain an empirical P-value. To quantify the magnitude of association between image ratings and demographic group, we calculated Cramér's V, a standardized effect size for χ² tests. For each table, we also computed bootstrapped 95% CIs for Cramér's V using 10,000 resamples of multinomially simulated data based on the observed proportions. Significance was set at a value of P < .05.

RESULTS

Demographic Analysis

Survey responses were collected from 994 participants, with a mean age of 37.4 years (range, 18-70 years). The sample included 515 cisgender men, 463 cisgender women, and 14 transgender individuals. Racial distribution included 595 Caucasian, 169 Asian, 60 Black, 63 Hispanic/Latino, 76 Native American, and 31 who selected “other.” Regarding sexual orientation, 715 participants identified as heterosexual, 172 as bisexual, 42 as homosexual, 23 as pansexual, and 14 as asexual, whereas 31 selected “other.” The most reported income range was 50K to 100K (n = 336, range, <25K to >200K). For more information regarding the results, tabular data are included in the Supplemental Material section of the manuscript (Supplemental Tables 1-6).

Overall Aesthetic Preferences

We utilized a TGHR to conduct our analysis of thigh-gap aesthetic preferences. We calculated the TGHR by dividing the thigh-gap distance by the hip width at the level of the ASIS. The thigh-gap distance was measured between the medial thighs at the midpoint of the distance between ASIS and mid-lateral knee. Overall, we found statistically significant differences between the most and least attractive images (P < .001). Image B (Figure 1B) with a TGHR of 0.09 was most frequently rated as “Most Attractive” (36.9%), followed by Image A (Figure 1A) with a TGHR of 0.19 (23.6%), Image D (Figure 1D) with a TGHR of 0.02 (20.5%), and Image C (Figure 1C) with a TGHR of 0.04 (18.9%) Conversely, Image D was rated as the “Least Attractive” by most respondents (48.4%), followed by Image A (39.7%), Image C (6.4%), and Image B (5.4%) (Figures 1, 2).

Figure 2.

Violin plot visualizing the distribution of attractiveness ratings for different images (A-D) across 4 ranking levels: most attractive, second most attractive, third most attractive, and least attractive. The width of each violin represents the density of ratings at each level, with wider sections indicating higher frequency.

Age-Based Preferences

When stratified by age group, Image B was most frequently selected as “Most Attractive” by participants across nearly all age groups , including 18 to 29 (39.1%), 30 to 39 (32.6%), 40 to 49 (39.9%), 50 to 59 (42.9%), 60 to 69 (39.5%), and 70+ (50.0%) (P = .728). Image D was consistently rated as the “Least Attractive” across all age groups, with the highest proportions among those older age groups 40 to 49 (54.5%), 50 to 59 (54.3%), 60 to 69 (51.2%), and 70+ (83.3%). Although insignificant, younger groups demonstrated a trend toward a higher proportion finding Image A least attractive compared with older respondents (Figure 3).

Figure 3.

Compiled bar chart indicating relative proportions of participant response to the electronic survey regarding their opinion of the most and least attractive image as organized by age.

Gender-Based Preferences

When stratified by gender identity, Image B was preferred as “Most Attractive” by both males (34%) and females (41%). Transgender participants selected Image A and Image D equally as the most attractive image (28%), but our small sample size does not allow for generalizability within this subgroup. No statistically significant differences were observed across gender groups (all P-values >.05). Image D was consistently rated as the “Least Attractive” across all gender categories (Figure 4).

Figure 4.

Compiled bar chart indicating relative proportions of participant response to the electronic survey regarding their opinion of the most and least attractive image as organized by gender.

Preferences by Race

Analysis of attractiveness preferences by race demonstrated significant variation (P < .001). Among Caucasian and African American respondents, Image B was rated “Most Attractive” (41% and 47%, respectively) and Hispanic respondents preferred Image C (27%). Asian participants displayed balanced preferences for Image B and Image D (28% each). Image D remained the most frequent choice for “Least Attractive” across all racial groups, particularly among African American (50%) and Caucasian respondents (52%) (Figure 5).

Figure 5.

Compiled bar chart indicating relative proportions of participant response to the electronic survey regarding their opinion of the most and least attractive image as organized by race.

Preferences by Sexual Orientation

Statistically significant differences were observed when choosing the most attractive image across sexual orientation groups despite the overwhelming majority of our participants being heterosexual individuals (P = .011). Heterosexual respondents most frequently selected Image B (41%), whereas bisexual and pansexual respondents preferred Image C (39% and 43%, respectively). Asexual respondents showed a higher preference for Image A (33%). Image D was most frequently rated “Least Attractive” amongst heterosexual and asexual individuals. Image A was rated least attractive for the rest of the cohort, particularly among pansexual respondents (87%). Rankings for the remaining images were more evenly distributed across groups (Figure 6).

Figure 6.

Compiled bar chart indicating relative proportions of participant response to the electronic survey regarding their opinion of the most and least attractive image as organized by sexual orientation.

Income-Based Preferences

Significant differences emerged in attractiveness rankings when analyzing the most attractive thigh-gap distance by income group (P = .029). Respondents earning $100K to $200K and $50K to $100K most frequently selected Image B as “Most Attractive” (47% and 36%, respectively), whereas lower-income respondents (<$25K) favored Image C (40%). The “Least Attractive” ranking was dominated by Image D across all income levels, with the highest proportion observed in respondents earning more than $200K (75%, P = .001; Figure 7).

Figure 7.

Compiled bar chart indicating relative proportions of participant response to the electronic survey regarding their opinion of the most and least attractive image as organized by income.

DISCUSSION

This is the first study where the authors provide a nuanced perspective on thigh-gap aesthetics, revealing how different thigh-gap proportions are perceived across diverse populations. Our data indicate that a TGHR of 0.09 (Image B) was consistently rated as the most attractive proportion across most demographic groups. This suggests a clear preference for a moderate thigh gap in the general population that avoids extremes in body contouring, favoring natural proportions.

The influence of demographic factors such as income, sexual orientation, and race on aesthetic preferences underscores the diversity of beauty ideals. Although appearance ideals in the United States have been studied through various lenses, such as age, race, and gender, comparatively little is known about the role of socioeconomic status in shaping body standards—particularly in relation to plastic and reconstructive surgery.^11,12 Socioeconomic status has been shown to shape how individuals internalize sociocultural appearance pressures, with distinct pathways to body image concerns across income levels.¹³ These findings underscore the importance of analyzing thinness ideals such as the thigh gap through a socioeconomic lens and informed our decision to include income in our demographic breakdown. The preference for Image C among lower-income groups and pansexual respondents highlights how cultural and socioeconomic contexts may shape perceptions of attractiveness. This aligns with previous studies that explore regional and cultural variations in beauty trends.^11,12 Additionally, variations in preferences based on race and sexual orientation highlight the importance of understanding the unique aesthetic ideals of different patient populations, which is crucial when planning body contouring procedures and setting realistic expectations with patients.^10,13

Our findings also reveal a disconnect between social media-driven beauty ideals and practical aesthetic preferences. Although social media platforms frequently promote the thigh gap as an aspirational standard, extreme gaps—such as the TGHR of 0.19 (Image A)—were rated as less attractive. This aligns with critiques of social media's unrealistic portrayals of beauty, reinforcing the importance of patient education about the anatomical and genetic factors influencing the thigh gap.^7,14 Moreover, it highlights the physical and psychological risks associated with striving for unattainable ideals.

Clinically, our findings suggest that surgical procedures affecting thigh contours should focus on achieving moderate, natural proportions while avoiding exaggerated proportions. This is particularly relevant not only in body contouring procedures like thigh lifts or liposuction, but also in reconstructive procedures such as profunda artery perforator flaps for breast reconstruction.^2,15 Surgeons equipped with an understanding of aesthetic benchmarks derived from general public preferences, along with a solid appreciation of the underlying anatomy, are well positioned to facilitate patient-facing discussions and informed surgical planning. Our findings suggest that the general preference tends to favor a moderately contoured thigh gap, regardless of demographic subgroup. This preference aligns with our intrinsic appreciation of average anatomy, making the moderate choice a more enduring aesthetic standard even amid cultural fluctuations in body ideals. From a surgical standpoint, overly aggressive liposuction or excessive excision during thigh lifts can create a cavitated appearance of the medial thigh, leading to a less desirable aesthetic outcome—exemplified by Image A.

Limitations of our study include the use of digitally manipulated images. Although necessary to isolate and evaluate specific thigh-gap widths, these images may not fully replicate the nuances of natural anatomical variation. A single image was utilized to serve as a base for manipulation. More models, of varying skin tones specifically, may have provided insight into the difference skin tone makes on thigh-gap preference. However, we feel this could complicate the responses and create a separate dynamic in the results in the form of racial perception. Additionally, not all thigh-gap widths or variations were tested, so it is possible that more ideal thigh-gap widths exist, which were not evaluated in this study. Future research should address this by utilizing larger, more diverse sets of images or by incorporating real patient images to enhance realism and diversity in body types. Our small sample size for transgender respondents limits our ability to generalize our findings to that subgroup of individuals, particularly when compared with such large subgroups of heterosexual and cisgender individuals. Further investigation into the LGBTQIA+ community's aesthetic preference of thigh-gap widths is warranted. Although MTurk offers access to a large participant pool, it is not without limitations. MTurk workers tend to be younger (the majority of our raters are between 18 and 39 years old), more educated, and more technologically literate than the broader population. Additionally, the platform's participant base is predominantly located in the United States and India, which may limit generalizability to other geographic or cultural contexts.

CONCLUSIONS

In conclusion, our data demonstrate that a moderate thigh gap with a TGHR of 0.09 is considered most attractive by the general population. Although preferences vary across demographic groups, these findings underscore the importance of open communication between physicians and patients seeking body contouring or reconstructive surgery to set realistic expectations and emphasize the value of balanced, natural proportions.

Supplemental Material

This article contains supplemental material located online at https://doi.org/10.1093/asjof/ojaf138.

Disclosures

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and publication of this article.