5 Self-Report Measures
5.1 Demographics
Demographics captured in the current study include: Race, Sexual Orientation, Gender, Ethnicity, Student Status, Working Status, Current Living Situation, Assessment Date, and Age
Scoring
Demographics are primarily drawn from the data collection form (measure prefix ‘dcf’). Cleaning steps include:
Selection of variables that are part of the data collection form (prefix ‘dcf’) and winnow variables to be included in the master dataset.
Calculation of age in years and remove birthdate (HIPPA Identifier) from the data set.
Changing assessment date to only capture month and year of assessment for further de-identification.
Binning categorical variables to response options most appropriate for race and ethnicity
Key Variables
age (Participant Age in Years)
race (Participant Race, recoded to: White[0], Black[1], Asian[2], Mixed Race/Other[3])
ethnicity (Hispanic[1] or non-Hispanic[0])
race_eth_1 (Race and ethnicity recoded to: White[0], Non-Hispanic Black [1], Non-Hispanic Asian [2], Hispanic [3], Mixed/Other [4])
race_eth_2 (Race and ethnicity recoded to: White [0], Asian [1], Other [2])
sex_orientation (Participant sexual orientation, recoded to: Heterosexual [0], Bi/Pansexual [1], Gay/Lesbian [2])
sex_minority (Participant sexual orientation recoded to Hetersoxesual[0] or Sexual Minority[1])
gender (Participant gender identity recoded to Male[0], Female [1], NonBinary/Other[2])
gender_TNB (Participant gender identity recoded to: Cisgender [0] or Trans/Nonbinary [1])
gender_female_nb (Participant gender identity coded to Male = 0 and Female/Nonbinary = 1)
sex_gender_minority (Participant sexual and gender identity coded to CisHet [0] or Sexual/Gender Minority [1])
assessment_month (assessment recoded to assessment month and year)
5.2 ED100k Plus
The ED100k Plus is a survey that compiles questions from validated measures of dieting and disordered eating including the ED100k (Thornton et al., 2018), Eating Disorder Diagnostic Scale (EDDS; (Stice et al., 2004)), Eating Disorder Examination Questionnaire (EDE-Q; (Luce & Crowther, 1999)), and the Dieting and Weight History Questionnaire (DWHQ; (Witt et al., 2013)) to provide a more comprehensive assessment of past and current symptoms of disordered eating. The survey begins by addressing weight suppression, lowest adult weight, and dieting history. The survey then addresses eating disorder behaviors and cognitions in both a past and present sense. The final questions of the ED100k Plus address menstrual information for female participants.
Scoring
Data cleaning steps include:
1-2. selects variables; recodes the compensatory exercise variable to remove missing codes, and recodes intentional weight control variables to remove missing codes and set ‘0’ equal to a ‘No/Never’ answer
3. overrides skip logic for exercise and other weight control variables such that if individuals have an initial ‘no’ answer where skip logic has overridden future questions, a ‘0’ (reflecting a ‘no/never’ answer) is entered in the subsequent question regarding details of the behavior.
4. Second recode for these questions such that missing codes are change to ‘NA’ and the ‘0’ answers are labelled appropriately
5-7. specifically calculate logic necessary to specify exercise-related outcome variables, described in key variables below.
8. cleans up the dataset by removing erroneous variables that are no longer needed.
After these steps ar complete, additional weight suppression variables are calculated and added to the data in the R-script
Key variables
ED100k_exercise_icb (Was exercise ever used as a behavior to intentionally control weight or shape?)ED100k_ex_compulsive (Ever felt compelled to exercise for wt and shape control OR uneasy / distressed if unable to exercise)ED100k_ex_interfere (Did exercise ever interfere in one’s life (changing eating habits, declining opportunities to be with friends, exercising despite illness or injury)ED100k_ex_excessive (>1 month of psychologically driven exercise every day or nearly every day)ED100k_ex_addictive (Having psychologically compulsive/driven exercise (feeling compelled to exercise or distressed if unable) that lasted for at least one month AND at least one life interfering symptom)ED100k_ex_compulsive_1mo (Compulsive/Driven Exercise that lasted for at least one month)ED100k_ex_maladaptive_1mo (reports driven exercise for at least one month OR any compensatory exercise)ex_current (Do you currently exercise to control weight and shape AND feel compelled to exercise or distress if unable to exercise?)Eating Disorder Behavior Variables
Case Status Variables
Weight suppression variables
ED100k_wt_suppress_high_current - current weight suppressionED100k_wt_suppress_high_lowest - difference between highest weight and lowest weight at adult heightED100k_wt_suppress_high_AN - difference between highest ever weight at adult height and weight during an AN episodeED100k_wt_suppress_current_AN - difference between current weight and weight duirng an AN episodeED100k_bmi_suppress_high_current : current BMI suppressionED100k_bmi_suppress_high_lowest: high-low BMI suppressionED100k_bmi_suppress_high_AN : high-AN BMI suppressionED100k_bmi_suppress_current_AN : current-AN BMI suppression (difference between current BMI and AN BMI)5.3 Eating Pathology Symptom Inventory
We used the 45-item Eating Pathology Symptom Inventory (EPSI; (forbushDevelopmentValidationEating2013?)) to assess eating disorder symptoms at pre-intervention, post-intervention, and 4-week follow-up. The EPSI has been validated in clinical, college, and community samples, with 7-8 subscales in various populations, with the 8-subscale version being the most commonly-employed version of the measure (coniglioFactorialIntegrityValidation2018?). The 8 subscales include: Body Dissatisfaction, Binge Eating, Cognitive Restraint, Purging, Restricting, Excessive Exercise, Negative Attitudes toward Obesity, and Muscle Building.
Scoring
Each item is scored on a 5-point Likert-style scale (0 = Never; 4= Often) to describe how well each item describes the participant experiences. Scores are derived by summing responses across the questions included in each subscale. The eight subscales of the measure are: Body Dissatisfaction, Binge Eating, Cognitive Restraint, Purging, Restricting, Excessive Exercise, Negative Attitudes toward Obesity, and Muscle Building.
Key Variables
All variables below reflect the final scoring of weighted subscale sum scores with < 25% missing data.
epsi_body_dissatisfaction_25
epsi_binge_25 (binge eating)
epsi_restraint_25
epsi_purging_25
epsi_restrict_25
epsi_exercise_25 (excessive exercise)
epsi_neg_attitude_25 (negative attidtude towards obesity)
epsi_muscle_building_25
5.4 Fat Phobia Scale - Short Form
The shortened Fat Phobia Scale (FPS) utilizes 14 questions to assess fatphobia (from the 50-item original scale). It is intended to measure fat phobia levels in the tested population through various personality traits and whether there is bias when people apply them to an overweight person (Bacon et al., 2001). The shortened FPS questionnaire demonstrates high reliability when compared to the original 50 question scale.
Scoring
The FPS uses a 5-point sliding scale with 1 being equal to a different negative descriptive variable for each respective question and 5 being equal to a different positive variable for each respective question. Some items are reverse scored prior to data aggregation such that th 1 (negative) to 5 (positive) valuation is maintained across the scale. A mean score is created from the 14 items.
Key Variables
fps_mean (Mean score of all 14 items)
5.5 Child Food Security Survey Module (CFSSM)
The U.S. Adult Food Security - Survey Module (US FSSM) (carlsonMeasuringFoodInsecurity1999?) was developed by the USDA Food and Nutrition Services in 1995 for assessing food security within households, specifically for use in the Current Population Survey. The US FSSM was adapted for administration with children, creating the Child FSSM (CFSSM), which is a 9-item survey that demonstrates sufficient reliability in measuring food insecurity (Connell et al., 2004).
Scoring
Respondents had to answer the 9 items with either “A lot”, “Sometimes”, or “Never”. Responses “A lot” and “Sometimes” were dichotomous as affirmative while “Never” remained negative. The sum of affirmative responses to the 9 questions is then the raw score on the scale.
Food security status is assigned as follows: | | - Raw score 0-1: Food secure | - Raw score 2-5: Food insecure without hunger | - Raw score 6+: Food insecure with hunger | |
Key Variables
5.6 Generalized Anxiety Disorder 7-item (GAD-7)
Generalized Anxiety Disorder 7-item (GAD-7) (spitzerBriefMeasureAssessing2006?) is a seven-item instrument that is used to measure or assess the severity of generalised anxiety disorder (GAD). Each item asks the individual to rate the severity of his or her symptoms over the past two weeks. Response options include “not at all”, “several days”, “more than half the days” and “nearly every day”. The GAD-7 is a widely-used and well-validated measure for anxiety screening (sapraUsingGeneralizedAnxiety2020?). the GAD-7 is scored on a unidimensional scale summing all 7 items, with cutoffs that can also be employed for interpretation of clinical severity: | | Minimal anxiety = 0-4 | Mild Anxiety = 5-9 | Moderate Anxiety = 10-14 | Severe Anxiety = 15-21. | |
Scoring
The sum of the respondent’s answers is used to generate an idea of the respondent’s anxiety level. A score of 10 points or higher indicates probably anxiety disorder, with 89% sensitivity and 82% specificity for GAD. We have also created a dichotomized measure of likely anxiety (cutoff score of 10 or greater), along with clinical severity index (minimal, mild, moderate, severe).
Key variables
gad_sum_25 (Sum of GAD-7, 25% or less items missing and weighted by N items completed)
gad_cutoffs (GAD-7 with anxiety severity cutoffs employed)
gad_anx_disorder (GAD-7 probable anxiety disorder Yes/No)
5.7 Goldfarb Fear of Fat Scale
The Goldfarb Fear of Fat Scale (GFFS) is a 10-item scale which was introduced in 1985 as a diagnostic tool to measure one’s fear of fat as it can be an indicator for the development of an eating disorder (goldfarbGoldfarbFearFat1985?). When used as a screening tool for both clinical and non-clinical groups, GFFS has strong test-retest reliability and psychometric properties (przybyta-basistaExaminationFactorStructure2022?). In addition, the test also has shown high validity for both populations with and without eating disorders (Goldfarb, 2010) While original factor structure is defined as a single-factor, (przybyta-basistaExaminationFactorStructure2022?) also supported a two-factor structure in a non-clinical sample: (1) fear of weight gain and (2) fear of losing control over eating/weight.
Scoring
GFFS uses a 4-point scale with 1 being equal to very untrue and 4 being equal to very true. The GFFS is scored by summing the items to create a total score. The measure also contains two subscales (fear of weight gain (FOWG) and loss of control (LOC) which can also be summed.
Key Variables
For variables below, _25 appendix reflects that individuals with <25% missing items on a scale were included with weighted sums based on number of items completed
gffs_sum_25 (sum of all items)
gffs_fowg_25 (sum of fear of weight gain subscale)
gffs_loc_25 (sum of loss of control items)
5.8 Mini Menstrual Questionnaire (MMQ)
The Mini Menstrual Questionnaire, or MMQ, is a tool that the EMBARK Lab created to assess menstruating individuals’ menstrual history and current menstrual status. This measure assesses the regularity of the individual’s menstrual cycles, the form of birth control (if any) the individual uses, if they are pregnant or breastfeeding, and past gynecological procedures they have undergone (if any).
5.10 Beck Depression Inventory (BDI)
The Beck Depression Inventory (or BDI) is a 21-item, self-report survey that measures characteristic attitudes and symptoms of depression (Beck et al., 1961). Participants are asked to rank each question, which reflects a symptom or attitude related to depression, on a scale from 0 to 3, with zero reflecting the absence of the symptom or attitude and 3 reflecting the most severe experience of the symptom or attitude. For example, one question asks about feelings of being a failure and the options range from “I do not feel like a failure.” to “I feel like I am a complete failure as a person.”
Scoring
Scoring is completed by summing the totals of each of the 21 items. A higher score represents higher symptomatology of depression (Beck et al., 1988). Although Beck and colleagues urge clinicians to use their clinical judgement in assigning cut-off scores, typically a score of 10 or lower represents none or minimal depression, 10-18 indicates moderate depression, 19-29 indicates moderate to severe depression, and scores 30 or above indicates severe depression (Beck et al., 1988).
Key Variables
5.11 Sociocultural Attitudes Towards Appearance Questionnaire (SATAQ)
The Sociocultural Attitudes Towards Appearance Questionnaire-4 (SATAQ) is designed to assess societal and interpersonal aspects of appearance ideals. The SATAQ has five validated factors: Internalization of Thin/Low Body Fat, Internalization of Muscular/Athletic, Pressures from Family, Pressures from Media, and Pressures from Peers. This scale structure was confirmed in 3 independent and geographically diverse samples of women from the United States (East Coast N = 440, West Coast N = 304, and North/Midwest N = 349). SATAQ-4 scale scores demonstrated excellent reliability and good convergent validity with measures of body image, eating disturbance, and self-esteem. The measure was also validated in college males from the United States (N = 271); however, there was some evidence of an underlying structure unique to men. For the purposes of the current data, the 5-factor structure is employed in scoring.
Scoring
For the SATAQ, a total score is generated based on the mean of the respondent’s answers. Additionally, subscale scores (internalization of thinness/low body fat, internalization of muscular/athletic. pressures - family, pressures - peers, and pressures - media) are also calculated.
Key Variables
Variables with the _25 appendix indicate that these variables include those with 25% or less missing data
sataq_mean_25 (mean of SATAQ)
sataq_thin_ideal_25 (mean of internalization of the thin ideal)
sataq_athletic_ideal_25 (mean of internalization of athletic ideal)
sataq_parental_pressure_25 (mean of parental pressure)
sataq_peer_pressure_25 (mean of peer pressure)
sataq_media_pressure_25 (mean of media pressure)
5.12 Traditional Masculinity-Femininity Scale (TMFS)
The Traditional Masculinity-Femininity Scale (TMFS) which was introduced in 2016 to assess self-ascribed masculinity and femininity (Kachel et al., 2016). The TMFS has demonstrated high reliability for both sexes. Additionally, TMFS shows validity due to its ability to reduce issues of social desirability when given as a self-assessment. TMFS uses a 7-point likert scale with 1 being equal to totally masculine and 7 being equal to totally feminine. In terms of scoring, the mean is taken of all given answers with an average less than 4 implying masculinity and an average greater than 4 implying femininity.
Scoring
The mean is taken of all given answers with an average less than 4 implying masculinity and an average greater than 4 implying femininity.
Key Variables
tmfs_mean_25 (average of scores in response to questions asking about different situations where one could consider their femininity and masculinity. 25% or less missing data)
5.13 Fat Stereotypes Questionnaire (FSQ)
Davison and Birch (Davison & Birch, 2004) developed the Fat Stereotype Questionnaire (FSQ) to assess fat stereotypes among 9 year old girls. This scale asks participants to think about attributes such as happiness, intelligence, attractiveness, laziness, having a lot of friends, and goodness/badness in relation to fatness and thinness.
Scoring Participants respond to prompts that are descriptors of both “thin” and “fat” people as the subject. They respond using a four-point Likert scale, which ranges from “really disagree” to “really agree,” which produces 18 scores (9 of which are with “fat” subjects and 9 of which are with “thin” subjects). Scoring is completed by calculating discrepancy scores (fat people attribute - thin people attribute) and calculating the mean of the 9 discrepancy scores. Scores above 0 indicate that the participant more strongly endorses that attribute for thin people, whereas scores below 0 indicate that the participant more strongly endorses that attribute for fat people. Importantly, before calculating the mean score, the attributes need to be reverse coded in order to interpret all values in the same direction. The internal consistency for this measure in Davison and Birch’s (2014) sample was ɑ = 0.71.
Key Variables
5.14 Perception of Teasing Scale (POTS)
The Perception of Teasing Scale (also known as POTS) was developed by Thompson et al.(Thompson et al., 1995) to assess children and adolescents’ history of being teased in relationship to a high-weight status and social competency. López-Guimerà et al. (López-Guimerà et al., 2012) validated an 11-item version of POTS with Spanish adolescents. The measure contains two subscales: weight-related teasing and competency-related teasing. Both subscales have a satisfactory level of internal consistency. A response-to-teasing is also assessed by asking the participant if they experienced any emotional distress due to the teasing they experienced.
Scoring
The teasing scale ranges from “never” to “always,” with “always” occupying the number five. The upset scale is also on a 5-point Likert scale from “not upset” to “very upset”. The final teasing score is generated by generating a sum of the score for all of the teasing items. A separate score for emotional distress is calculated by creating a sum for all of the repsonse-to-teasing items.
Key Variables
5.15 Feedback Survey
The feedback survey is a lab-created 15-question survey that asks questions about the participants’ experiences participating in BAM High. This survey asks participants to reflect on their experiences participating and evaluate the program in terms of facilitator strength, helpfulness, level of distress related to the intervention, and more. There are some items that ask for a “yes” or “no” answer, while others provide space for open comment.