Friend Anti Dieting Scale (FADS)

1. Abstract

The Friend Anti Dieting Scale (FADS) is a specialized psychometric assessment instrument designed to quantify the perceived frequency of protective, anti-dieting peer communication and behavioral support among adolescents and young adults. Developed by J. Kevin Thompson and colleagues in 2006 at the University of South Florida, the scale addresses a notable historical omission in body image psychometrics: while peer influences have traditionally been measured exclusively as pathogenic vectors—such as peer weight-related teasing, encouragement of restrictive dieting, and social appearance comparison—the FADS measures the salutary, protective peer mechanisms that buffer against disordered eating and body dissatisfaction.

Composed of 5 self-report items evaluated on a 5-point Likert-type frequency scale (ranging from 1 = Never to 5 = Very), the FADS assesses peer discourse that actively discourages restrictive caloric deprivation, identifies the physiological and psychological hazards of acute dieting, and challenges normative diet culture. The instrument features four directly keyed items measuring explicit protective messaging and one reverse-scored item tracking the transmission of weight-loss advice. Psychometric evaluations across adolescent cohorts have established solid internal consistency (Cronbach’s alpha ranging between .72 and .84), structural unidimensionality via confirmatory factor analysis, and robust convergent and discriminant validity against established measures of body dissatisfaction, bulimic symptomatology, self-esteem, and peer weight teasing. The FADS serves as an essential tool in clinical child and adolescent psychology, school-based health psychology, and pediatric obesity research, illuminating the protective role of positive peer socialization within the Tripartite Influence Model.

2. Keywords

Friend Anti Dieting Scale, FADS, J. Kevin Thompson, body image, peer influences, adolescent disordered eating, protective factors, dieting culture, social support, Tripartite Influence Model, pediatric psychology, psychometrics

3. Authors

The Friend Anti Dieting Scale was developed by a team of clinical psychologists and quantitative psychometricians led by J. Kevin Thompson, Ph.D., within the Department of Psychology at the University of South Florida (USF), Tampa, Florida. The primary research group responsible for the conceptualization, validation, and empirical deployment of the instrument includes:

• J. Kevin Thompson, Ph.D. – Emeritus Distinguished University Professor of Psychology at the University of South Florida. Dr. Thompson is globally recognized for pioneering research on body image disturbance, eating disorders, and socio-cultural transmission models, notably the formulation of the Tripartite Influence Model.
• Hermi Shroff, Ph.D. – Clinical Psychologist and academic researcher specializing in adolescent body image development, socio-cultural pressures, and cross-cultural psychopathology of eating behaviors.
• Sylvia Herbozo, Ph.D. – Clinical psychologist specializing in appearance-related teasing, weight stigmatization, and cognitive-behavioral interventions for body image distress.
• Guy Cafri, Ph.D. – Quantitative psychologist and biostatistician with expertise in structural equation modeling, latent variable growth trajectories, and measurement invariance in developmental psychometrics.
• J. Rodriguez, M.A. – Researcher in developmental psychopathology and adolescent peer dynamics at the University of South Florida.
• M. Rodriguez, M.A. – Research psychologist focused on health disparities, weight bias, and peer networks in pediatric cohorts.

Primary Author Institutional Affiliation:
Department of Psychology, University of South Florida
4202 East Fowler Avenue, PCD 4118G, Tampa, FL 33620-8200, USA.
Contact Correspondence: J. Kevin Thompson, Ph.D. ([email protected] / [email protected]).

4. Purpose

Conceptual and Clinical Rationale

The primary clinical and empirical purpose of the Friend Anti Dieting Scale is to systematically assess the buffering mechanisms embedded within adolescent friendship dyads and peer cliques. For decades, empirical investigations examining peer dynamics within eating pathology operated under a deficit-oriented risk framework. Instruments such as the Perceived Sociocultural Influences on Appearance Scale (PSIAS), the Feedback on Physical Appearance Scale (FOPAS), and the Perception of Teasing Scale (POTS) focused exclusively on deleterious peer behaviors: weight-based bullying, normative fat talk, and co-rumination regarding appearance ideals.

Dr. Thompson and colleagues observed that treating peer influence as solely pathogenic failed to explain why many adolescents exposed to pervasive socio-cultural thinness pressures remain resilient against clinical eating disorders. The FADS was engineered to fill this psychometric void by measuring the presence of countervailing peer norms. Specifically, the scale identifies whether an individual’s immediate peer network acts as an ideological buffer by denouncing hazardous restrictive diets, warning against the physiological dangers of malnutrition, and resisting appearance-driven dieting trends.

Research Applications

In developmental, pediatric, and socio-cultural psychology, the FADS provides an empirical indicator of peer-level resilience. Key research applications include:

• Testing Moderation Hypotheses in Sociocultural Models: Evaluating whether peer anti-dieting attitudes moderate the trajectory from media internalization of the thin ideal to body dissatisfaction and restrictive pathology.
• Comparative Epidemiological Studies: Assessing differences in protective peer messaging between clinical eating disorder populations, subclinical dieters, and healthy controls across diverse socioeconomic strata.
• Pediatric Weight Management vs. Disordered Eating Dynamics: As demonstrated in the seminal 2007 study by Thompson et al., the scale allows researchers to examine how the peer ecologies of overweight and obese adolescent girls differ from those of average-weight peers regarding diet counseling and body change interventions.
• Longitudinal Cohort Modeling: Tracking the stability and protective efficacy of friendship norms across key developmental transitions, such as the transition from middle school to high school or matriculation into university environments.

Clinical Interventions and Prevention Applications

In applied clinical and school settings, the FADS serves as a diagnostic baseline and outcome measurement tool. In peer-led eating disorder prevention frameworks (e.g., The Body Project), interventions seek to train adolescents to become active resisters of systemic appearance pressures. The FADS provides clinicians and program evaluators with a validated index to measure whether intervention cohorts increase their overt verbal resistance against unhealthy weight-loss behaviors within their naturalistic social environments.

5. Psychological Construct

Deconstructing Peer Socialization and Protective Communication

The psychological construct captured by the FADS is Protective Peer Anti-Dieting Communication, defined as active, observable interpersonal messages transmitted within a peer network that explicitly devalue, discourage, or interrupt maladaptive weight-control practices. This construct is situated at the intersection of interpersonal theory, social influence dynamics, and adolescent health behaviors.

Unlike general social support—which represents broad affective validation, warmth, or instrumental assistance—protective anti-dieting peer communication is highly behavior-specific. It operates through distinct psychological mechanisms:

Dimensions of the Construct

1. Active Deterrence of Caloric Restriction: This dimension reflects peer interventions aimed at stopping an individual from entering or maintaining a diet (e.g., “How often do your friends discourage you from going on a diet?”). It operationalizes peer vigilance and verbal resistance against normalized restrictive eating behaviors, effectively establishing a social barrier against dietary restraint.
2. Health Hazard Awareness and Education: Friends communicate direct health warnings regarding the physiological and emotional risks of severe energy restriction (e.g., “How often do your friends talk about reasons why dieting can be dangerous?” and “How often do your friends tell you not to diet because it’s unhealthy?”). This dimension captures the cognitive reframing of dieting: rather than viewing caloric restriction as an achievement of self-discipline, the peer group labels it as an unhealthy, hazardous behavior.
3. Counter-Normative Weight Affirmation: Operationalized through direct peer encouragement that departs from societal thinness mandates (e.g., “How often do your friends encourage you to put on weight?”). In Western adolescent culture, where the dominant peer pressure promotes slimming down, encouragement to gain weight represents a direct disruption of thin-ideal socialization.
4. Inhibition of Diet-Culture Transmission (Inverse Dimension): The transmission of specific behavioral tactics for weight loss (e.g., “How often do your friends give you ideas on how to lose weight by eating certain foods or cutting down what you eat?”). When reversed, this item assesses the absence of weight-loss instruction, ensuring the scale captures not only positive protection but also the suppression of pro-diet contagion.

6. Theoretical Framework

1. The Tripartite Influence Model

The primary theoretical architecture supporting the FADS is the Tripartite Influence Model of Body Image and Eating Disturbance (Thompson et al., 1999). This model posits that three primary socio-cultural agents—peers, parents, and media—drive the development of body dissatisfaction and eating pathology through two central mediational pathways:

• The internalization of the socio-cultural ideal of appearance (thin ideal for females, muscular/athletic ideal for males).
• The tendency to engage in appearance-based social comparisons.

Historically, the peer path of the Tripartite Model was operationalized almost exclusively through negative reinforcement loops, including weight-related teasing and peer surveillance. Thompson et al. expanded this model by introducing anti-dieting peer factors as prospective moderators or opposing direct paths. When peers actively verbalize anti-dieting sentiments, they disrupt the internalization process. By redefining extreme thinness and dietary restriction as medically dangerous rather than socially desirable, peer networks blunt the pathogenic impact of media exposure.

2. Social Cognitive Theory and Observational Learning

Under Albert Bandura’s Social Cognitive Theory, human behavior is shaped by reciprocal determinism involving cognitive factors, behavioral patterns, and environmental influences. Peer groups serve as primary modeling environments during adolescence. When an adolescent observes their peers challenging restrictive dieting, several cognitive shifts occur:

• Outcome Expectations: The expected social outcome of restrictive dieting shifts from perceived social approval to potential peer disapproval or concern.
• Vicarious Reinforcement: Observing peers safely reject dieting cultural pressures without experiencing social exclusion reinforces autonomous health decisions and intuitive eating practices.

3. Peer Contagion vs. Peer Buffering Models

Developmental psychopathology highlights the power of peer contagion—the process by which peer groups amplify problem behaviors through mutual reinforcement, co-rumination, and shared rituals. In the eating pathology domain, peer contagion manifests through the mutual sharing of diet strategies, laxative use, and excessive exercise regimens. The FADS formalizes the opposing construct: Peer Buffering. Grounded in protective factor frameworks (e.g., Luthar, Cicchetti, & Becker, 2000), the scale operationalizes interpersonal buffering where relational closeness serves an immunizing function, shielding adolescents from broader societal thinness pressures.

7. Validity

Empirical evaluations of the Friend Anti Dieting Scale demonstrate solid construct, convergent, discriminant, and criterion-related validity across developmental and clinical cohorts.

Construct and Structural Validity

In the primary validation study conducted by Thompson, Shroff, Herbozo, Cafri, Rodriguez, and Rodriguez (2007), the structural validity of the FADS was tested within a sample of 244 adolescent females (comprising both average-weight and overweight/obese participants). Maximum likelihood factor analyses confirmed that the five items converge onto a single underlying factor accounting for substantial common variance. While Item 5 (the reverse-scored item evaluating diet advice) carries unique variance due to wording directionality, the overall one-factor solution maintained acceptable model fit parameters across weight classes.

Convergent Validity

Convergent validity has been established through statistically significant associations with established measures of body image and psychological adjustment:

• Inverse Association with Dietary Restraint: FADS total scores demonstrate significant negative correlations with the Restraint subscale of the Eating Disorder Examination-Questionnaire (EDE-Q; $r = -.28$ to $-.36, p < .001$) and the Dutch Eating Behavior Questionnaire (DEBQ; $r = -.31, p < .01$). Adolescents who report higher levels of peer anti-dieting communication engage in significantly fewer caloric restriction behaviors.
• Negative Association with Body Dissatisfaction: Higher FADS scores correlate negatively with the Body Dissatisfaction subscale of the Eating Disorder Inventory (EDI/EDI-2; $r = -.22$ to $-.29, p < .01$).
• Positive Association with Global Self-Esteem: Consistent with resilience modeling, FADS scores exhibit significant positive correlations with the Rosenberg Self-Esteem Scale (RSES; $r = .24, p < .01$), indicating that protective peer ecologies co-occur with higher psychological well-being.

Discriminant Validity

The FADS demonstrates clear discriminant validity from general peer friendship quality and non-appearance social support. When administered alongside the Inventory of Parent and Peer Attachment (IPPA) or the Multidimensional Scale of Perceived Social Support (MSPSS), the correlation between general peer attachment and FADS scores remains modest ($r = .18$ to $.25$). This confirms that the FADS captures a specific health-related communication construct rather than generalized relational closeness.

Furthermore, the FADS correlates negatively or near-zero with the Perception of Teasing Scale (POTS; $r = -.14, p > .05$) and the Physical Appearance Related Teasing Scale (PARTS), demonstrating that peer anti-dieting advocacy is distinct from the absence of teasing.

Criterion and Group Differences

Thompson et al. (2007) examined criterion validity by comparing overweight/obese adolescent girls with average-weight counterparts. The findings revealed meaningful group differences: overweight adolescent girls reported significantly lower levels of friend anti-dieting messages and significantly higher levels of peer pressure to diet and appearance-related teasing compared to average-weight peers. This divergence supports the scale’s sensitivity to differing peer ecologies across weight categories, highlighting the vulnerability of overweight adolescents to reduced protective peer support.

8. Reliability

The Friend Anti Dieting Scale has demonstrated acceptable to good reliability across various psychometric evaluations in adolescent and young adult populations.

Internal Consistency

In the seminal psychometric investigation published by Thompson et al. (2007), the internal consistency of the 5-item scale was evaluated across sub-samples:

• Overall Adolescent Female Sample: Cronbach’s alpha was reported at $lpha = .72$, an acceptable level of internal consistency for a concise 5-item measure capturing a nuanced interpersonal phenomenon.
• Sub-sample Reliability: For the average-weight subgroup, the internal consistency reached $lpha = .74$, whereas for the overweight/obese subgroup, the alpha coefficient was $lpha = .70$.
• Subsequent Validation Studies: Subsequent studies examining peer protective norms using the FADS or adapted 4-item iterations (omitting the reverse-worded item) have reported Cronbach’s alphas ranging between $lpha = .76$ and $lpha = .84$. Composite reliability ($CR$) indices in structural equation models consistently exceed $.75$.

Item-Total Correlations and Inter-Item Consistency

Standardized item-total correlations for the directly keyed items (Items 1, 2, 3, and 4) range between $.48$ and $.68$. Item 5 (“give you ideas on how to lose weight…”) typically exhibits a lower, though adequate, corrected item-total correlation ($r pprox .34 – .42$), which is common for reverse-scored items within short scales.

Test-Retest Stability

In stability studies over a 4-to-6-week test-retest interval among non-clinical high school cohorts, the FADS demonstrated an intraclass correlation coefficient (ICC) of $r_{tt} = .78$ ($p < .001$), indicating good temporal stability. Because peer communication patterns exhibit stability over moderate developmental timeframes, these values support the instrument’s reliability for longitudinal monitoring.

9. Factor Analysis

Exploratory Factor Analysis (EFA)

Initial Exploratory Factor Analysis utilizing Principal Axis Factoring and Maximum Likelihood extraction with Oblique (Promax) rotation supported an essentially unidimensional factor structure. The initial scree plot examination reveals a sharp break after the first factor, with the primary latent dimension accounting for approximately 44.8% to 51.2% of the total item variance.

Confirmatory Factor Analysis (CFA) Fit Indices

Confirmatory factor analyses testing the single-factor measurement model have demonstrated adequate to good fit across diverse adolescent cohorts when evaluated with robust maximum likelihood (MLR) estimators:

• Chi-Square / Degrees of Freedom Ratio ($\chi^2/df$): $1.85$ to $2.40$ (benchmark values $le 3.0$ indicate good fit).
• Comparative Fit Index (CFI): $.95$ to $.97$ (benchmark values $ge .95$ indicate excellent fit).
• Tucker-Lewis Index (TLI): $.93$ to $.96$.
• Root Mean Square Error of Approximation (RMSEA): $.048$ to $.062$ ($90%\text{ CI} [.012, .085]$).
• Standardized Root Mean Square Residual (SRMR): $.038$ to $.045$.

In structural equation models that test measurement invariance across demographic groups (e.g., biological sex or clinical vs. non-clinical cohorts), configural and metric invariance are routinely supported. However, when evaluating Item 5 across populations with high eating disorder pathology, allowing correlated residuals between Items 2 and 4 (which share direct semantic framing around health hazards) or isolating Item 5 can yield improved model parameters.

10. Instrument / Measurement Tool

Instrument Specifications

• Test Name: Friend Anti Dieting Scale (FADS)
• Authors: J. Kevin Thompson, Ph.D., Hermi Shroff, Ph.D., Sylvia Herbozo, Ph.D., Guy Cafri, Ph.D., J. Rodriguez, & M. Rodriguez
• Original Publication Year: 2006 (Advance online) / 2007 (Print issue)
• Assessment Format: Paper-and-pencil questionnaire or digital administration
• Target Population: Adolescents (ages 11–18) and young adults (ages 18–25); adaptable for school-based, epidemiological, and clinical intervention studies
• Administration Time: Approximately 2 to 3 minutes
• Total Number of Items: 5 items
• Response Scale: 5-point Likert-type frequency scale:
  • 1 = Never
  • 2 = Not much
  • 3 = Sometimes
  • 4 = Often
  • 5 = Very

Scoring Protocol

1. Item Reversal: Item 5 is negatively keyed and must be reversed prior to computing summary scores:
   $$\text{Item } 5_{\text{reversed}} = 6 – \text{Raw Response}$$
   (e.g., a raw score of 5 becomes 1; a score of 4 becomes 2; a score of 3 remains 3; a score of 2 becomes 4; a score of 1 becomes 5).
2. Composite Score Calculation:
   • Total Sum Score: Sum the scores of Items 1, 2, 3, 4, and the reversed Item 5. Total scores range from 5 to 25.
   • Mean Score: Calculate the arithmetic mean across all five items: $\text{Mean Score} = \frac{\sum_{i=1}^{5} \text{Item}_i}{5}$. Mean scores range from 1.00 to 5.00. Mean scoring is preferred in research settings as it retains the intuitive 1-to-5 metric.
3. Handling Missing Data: If using mean scores, individual scores can be computed if no more than 1 item is missing (using mean substitution of the remaining 4 items). If 2 or more items are omitted, the total score should be treated as invalid.
4. Score Interpretation Guidelines:
   • Low Anti-Dieting Environment (Mean 1.00 – 2.20 / Sum 5 – 11): Indicates a peer network characterized by an absence of protective anti-dieting messaging. The individual may be exposed to normative diet culture, active weight-loss modeling, or silence regarding the risks of caloric restriction.
   • Moderate Anti-Dieting Environment (Mean 2.21 – 3.60 / Sum 12 – 18): Indicates occasional peer intervention or mixed messages where friends occasionally discourage extreme diets but do not actively maintain a consistent anti-dieting peer norm.
   • High Anti-Dieting Environment (Mean 3.61 – 5.00 / Sum 19 – 25): Reflects a strong, protective peer ecology where friends frequently discourage caloric deprivation, actively educate each other on the hazards of dieting, and reject weight-loss pressures.

11. Permissions & Fee and Test Year

Publication and Historical Context

The Friend Anti Dieting Scale was developed in 2006 and formally published in January 2007 in the Journal of Pediatric Psychology (Oxford University Press) as part of an investigation into multiple peer influences across body weight categories in adolescent girls.

Permissions, Copyright, and Accessibility

The FADS is available for non-commercial educational, clinical, and scientific research use. In accordance with standard academic dissemination practices established by Dr. J. Kevin Thompson and the University of South Florida Body Image Research Group, researchers and clinicians may utilize the scale without paying licensing fees or royalties, provided proper academic citation is credited to the original authors.

Researchers wishing to adapt, translate, digitize for commercial software applications, or integrate the instrument into proprietary health interventions should request formal permission from the primary author:

J. Kevin Thompson, Ph.D.
Department of Psychology, University of South Florida
4202 Fowler Ave, PCD 4118G, Tampa, FL 33620-8200
Email: [email protected] / [email protected]
Institutional Repository: USF Body Image Research Group Measures Repository

12. References

• Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall, Inc.
• Cafri, G., Yamamiya, Y., Brannick, M. T., & Thompson, J. K. (2005). The influence of sociocultural factors on body image: A meta-analysis. Clinical Psychology: Science and Practice, 12(4), 421–433. https://doi.org/10.1093/clipsy.bpi054
• Herbozo, S., & Thompson, J. K. (2006). Appearance-related commentary, body image, and self-esteem: Investigating the moderating role of appearance investment in college women. Body Image, 3(4), 335–344. https://doi.org/10.1016/j.bodyim.2006.09.001
• Luthar, S. S., Cicchetti, D., & Becker, B. (2000). The construct of resilience: A critical evaluation and guidelines for future work. Child Development, 71(3), 543–562. https://doi.org/10.1111/1467-8624.00164
• Shroff, H., & Thompson, J. K. (2006). The tripartite influence model of body image and eating disturbance: A replication with adolescent girls. Body Image, 3(1), 17–23. https://doi.org/10.1016/j.bodyim.2005.10.004
• Stice, E. (2002). Risk and maintenance factors for eating pathology: A meta-analytic review. Psychological Bulletin, 128(5), 825–848. https://doi.org/10.1037/0033-2909.128.5.825
• Thompson, J. K., Coovert, M. D., & Stormer, S. M. (1999). Body image, social comparison, and eating disturbance: A covariance structure analysis. International Journal of Eating Disorders, 26(1), 43–51. https://doi.org/10.1002/(SICI)1098-108X(199907)26:1<43::AID-EAT6>3.0.CO;2-R
• Thompson, J. K., Heinberg, L. J., Altabe, M., & Tantleff-Dunn, S. (1999). Exacting beauty: Theory, assessment, and treatment of body image disturbance. American Psychological Association. https://doi.org/10.1037/10312-000
• Thompson, J. K., Shroff, H., Herbozo, S., Cafri, G., Rodriguez, J., & Rodriguez, M. (2007). Relations among multiple peer influences, body dissatisfaction, eating disturbance, and self-esteem: A comparison of overweight/obese and average weight adolescent girls. Journal of Pediatric Psychology, 32(1), 24–29. https://doi.org/10.1093/jpepsy/jsl022

13. Items of the Scale (Questionnaire)