Development of the Body Checking

Questionnaire: A Self-Report Measure of Body

Checking Behaviors

Deborah L. Reas,1 Brooke L Whisenhunt,1 Rick Netemeyer,2

and Donald A. Williamson1,3*
1
Department of Psychology, Louisiana State University, Baton Rouge, Louisiana
2
Department of Marketing, Louisiana State University, Baton Rouge, Louisiana
3
Pennington Biomedical Research Center, Baton Rouge, Louisiana
Accepted 19 April 2001

Abstract: Objective: The purpose of this study was to develop a brief self-report inventory to
assess body checking behaviors. Method: Using exploratory and con®rmatory factor ana-
lyses, the 23-item Body Checking Questionnaire (BCQ) was developed. The BCQ measures
the global construct of body checking behaviors with three correlated subfactors that assess
checking related to overall appearance, checking of speci®c body parts, and idiosyncratic
checking rituals. Results: The BCQ was found to have good test-retest reliability (.94) and the
subfactors had good internal consistency (.88, .92, and .83). The measure correlated highly
with other measures of negative body image and eating disorders, demonstrating its con-
current validity. Additionally, the BCQ was found to differentiate normal controls and eating
disorder patients, as well as nonclinical participants scoring high and low on a measure of
concern with body size and dieting. Discussion: Due to the potential role of ritualistic body
checking in the maintenance of body dissatisfaction by directing excessive attention to body
shape/weight, the BCQ may prove to be a useful clinical tool in the assessment and treatment
of eating disorder patients. Ó 2002 by Wiley Periodicals, Inc. Int J Eat Disord 31: 324 333,
2002; DOI 10.1002/eat.10012

Key words: Body Checking Questionnaire; body checking behaviors; negative body image

INTRODUCTION

Negative body image and overconcern with body size/shape are cardinal features of
eating disorders (American Psychiatric Association [APA], 1994). The past 15 years have
witnessed a burgeoning interest in the study of body image disturbances, resulting in a
growing consensus that body image is a multidimensional phenomenon, involving per-
ceptual, attitudinal, and behavioral features (Thompson, 1996; Williamson, 1990). The
majority of this research has focused on the perceptual and attitudinal aspects of body


Correspondence to: Donald A. Williamson, Ph.D., Pennington Biomedical Research Center, 6400 Perkins
Road, Baton Rouge, LA 70808. E-mail: [email protected]
Ó 2002 by Wiley Periodicals, Inc.
Development of Body Checking Questionnaire 325

image, and only a few studies have investigated the behavioral manifestations of a neg-
ative body image. For example, of the 42 body image assessment procedures reviewed by
Thompson (1996), only three instruments specifically measured behaviors related to
negative body image: the Body Image Avoidance Questionnaire (BIAQ; Rosen, Srebnik,
Saltzberg, & Wendt, 1991), the Physical Appearance Behavior Avoidance Test (PABAT;
Thompson, Heinberg, & Marshall, 1994), and the Physical Appearance Comparison Scale
(PACS; Thompson, Heidelberg, & Tantleff, 1991).
Patients diagnosed with an eating disorder often organize their lifestyle to accom-
modate a negative body image. They may avoid social situations that emphasize
physical appearance, compulsively compare themselves with others, or ritualistically
check their body weight/size. Checking behaviors may resemble compulsions, in that
they prevent or ``undo'' distress stemming from preoccupation with body size and
shape (Rosen, 1997). Because these behaviors are often time-consuming and distressing,
it may be important to include them in the assessment and treatment of eating dis-
orders.
Paradoxically, frequent checking of body size may serve to reinforce body dissatis-
faction in eating disorder patients by directing excessive attention to estimates/judg-
ments of size/shape (Williamson, Muller, Reas, & Thaw, 1999). A recent cognitive-
behavioral theory of anorexia nervosa proposed by Fairburn, Shafron, and Cooper (1999)
emphasized the important role of body checking in the maintenance of anorexia nervosa.
The authors suggested that as the disorder progresses, control over eating, body shape,
and body weight become the primary indices of self-worth and self-control. Body
checking is used to monitor changes in weight and size, but has the effect of magnifying
perceived imperfections in appearance, thus encouraging further checking and increased
preoccupation with weight/size. As a result, hypervigilant body checking maintains
beliefs regarding fatness and body size. Further, the authors hypothesized that normal
variations in body weight may result in large mood swings for patients who weigh
themselves many times per day.
Due to their role in maintaining preoccupation with body size, the elimination of body
checking behaviors could be an important focus of treatment for many eating disorder
patients. Because body checking may also play a role in the etiology of eating disorders,
the measurement of body checking habits may also be useful in programs designed to
prevent the development of clinical eating disorders. To date, no method for objectively
measuring body checking behaviors has been developed. This psychometric study aimed
to develop a brief and easily administered self-report measure designed to assess body
checking rituals that are often observed in patients with eating disorders. The study had
two phases: initial development of the questionnaire and confirmation of factor structure
and tests of reliability and validity.

STUDY 1: TEST DEVELOPMENT

Participants
In Study 1,259 research participants, including 244 female college students and 15 fe-
males diagnosed with an eating disorder (APA, 1994), participated in a factor analytic
study designed to define the domain of body checking behaviors. The eating disorder
sample (n = 15) included 5 women diagnosed with anorexia nervosa (AN), 5 diagnosed
with bulimia nervosa (BN), and 5 diagnosed with an eating disorder not otherwise speci®ed
(ED-NOS). These diagnoses were established using the Interview for the Diagnosis of
326 Reas et al.

Eating Disorders- 4th version (IDED-IV; Kutlesic, Williamson, Gleaves, Barbin, & Murphy-
Eberenz, 1998), which is a reliable and valid measure for the diagnosis of eating disorders
using criteria de®ned in the 4th ed. of the Diagnostic and Statistical Manual of Mental
Disorders (DSM-IV; APA, 1994).
Overall, the average age of the participants was 21.8 years, with a range of 15 51 years.
The sample comprised Caucasians (n = 163; 62.9%), African Americans (23.6%; n = 61),
and other races (13.5%). Most participants reported that they had never been married (n
= 221; 85.3%). The average body mass index (BMI; kg/m2) for the college students was
23.0 (range, 15.4 44.7) and the average BMI for the eating disorder sample was 19.7
(range, 12.4 27.0). Consent forms explaining the purpose and procedures of the study
were read and signed by all participants. The eating disorder patients were administered
the Body Checking Questionnaire (BCQ) as part of an assessment upon entering treat-
ment at either an outpatient clinic or day hospital program. The college undergraduates
were recruited from psychology classes and received extra course credit in exchange for
their participation in the study.

Assessment Measures
The BCQ
The initial item pool of the BCQ consisted of 38 items selected to assess the frequency of
body checking behaviors. Items were generated by two of the authors (DLR and DAW)
based on a review of the literature pertaining to the various habits of eating disorder
patients. Input regarding the initial item pool was obtained from graduate students in
clinical psychology and from several eating disorder patients being treated in a university
outpatient program. Each item was scored on a 5-point Likert-type scale, ranging from 1
= never to 5 = very often.
To establish the content validity of the BCQ items, a panel of experts within the field of
eating disorders was selected to evaluate the relevance and representativeness of the
questions and instructions of the BCQ. Based on recommendations by Haynes, Richard,
and Kubany (1995), the experts reviewed the response format, structure, appropriateness,
clarity, and representativeness of each of the 38 BCQ items. The experts included nine
clinical psychologists (PhD) and one psychiatrist (MD).1 They were instructed to rate how
strongly they agreed with statements regarding the purpose, clarity, relevance, and ap-
propriateness of the 38 BCQ items using rating scales ranging from a score of 1 (do not
agree) to 7 (strongly agree). Results indicated that the experts rated the content of the BCQ
very highly, with average ratings ranging from 6.2 to 6.7 on the 7-point scale for each of
the 38 items. These results lend support to the content validity of the initial pool of items
that would form the BCQ.1

Demographic Questionnaire
A questionnaire that gathered information regarding age, race, and marital status was
administered to all participants. For the 15 eating disorder patients, diagnostic infor-
mation regarding the eating disorder was obtained by interview using the IDED-IV.

1
On average, the panel of experts had 8 years of experience in the treatment and/or research of eating
disorders. Nine of the 10 experts had written a dissertation related to the ®eld of eating disorders.
Development of Body Checking Questionnaire 327

College females were asked whether they were currently on a diet for the purpose of
weight loss. Height and weight were measured using weight scales and a stadiometer
and BMI was calculated using the formula kg/m2. BMI has been found to be a valid
measure of body fat relative to height (Garrow, 1983).

Results
Exploratory Factor Analysis
To examine the factor structure of the BCQ, responses were subjected to an ex-
ploratory principal components analysis (PCA) using oblique rotation. We hypothe-
sized that the items would form positively correlated factors consistent with a higher-
order factor structure. Therefore, an oblique rotation was used to assess the level of
correlation among the components (Hair, Anderson, Tatham, & Black, 1999). The PCA
extracted three components with eigenvalues >1.0 that accounted for over 50% of the
total variance. An inspection of the rotated solution suggested that the components
could be labeled as an overall appearance factor, a specific body parts factor, and an
idiosyncratic checking factor. From this solution, 10 items were deleted due to low
loadings on a factor (<.50) or because they loaded equivalently on more than one factor
(component), leaving 28 items.
Another PCA was conducted on the remaining 28 items. This analysis also ex-
tracted three components with eigenvalues >1 and the 28 items were retained after
the second PCA. The overall appearance factor contained 13 questions that measured
checking behaviors related to general appearance and accounted for 20.6% of the
total variance. The specific body parts contained nine items that measured body
checking behaviors related to specific body parts and accounted for 15.5% of the
variance. The idiosyncratic checking factor contained six items related to checking
behaviors that can be considered unusual or idiosyncratic and accounted for 15.2%
of the total variance. As hypothesized, the correlations among these components
were high, ranging from r = .56 to r = .85, suggesting that a higher-order factor
structure was tenable. This factor structure was tested using con®rmatory factor
analysis in the second study.

STUDY 2: CONFIRMATORY FACTOR ANALYSIS AND TESTS

OF RELIABILITY AND VALIDITY

Participants
The 28-item version of the BCQ, a demographic questionnaire, and several other
measures used to establish concurrent and discriminant validity (described below)
were administered to 149 female college students and 16 women diagnosed with an
eating disorder. The test-retest analysis was also conducted using this sample. The
clinical sample was composed of 4 women with AN, 9 with BN, and 3 with ED-NOS
according to DSM-IV criteria (APA, 1994). The mean age of the participants was 20.8
years, with a range of 16 56 years. Average BMI for the entire sample was 22.1, with a
range of 16.3 37.7. The sample comprised Caucasians (n = 129; 78.2%), African
Americans (n = 20; 12.1%), and Hispanic and other races (n = 16; 9.7%). Most par-
ticipants reported that they had never been married (n = 156; 94.5%).
328 Reas et al.

Results
Con®rmatory Factor Analysis
To confirm that the BCQ is a single measure of a higher-order factor (body checking)
with three subfactors that are highly correlated, a maximum likelihood factor analysis
was conducted.
The results showed that although the individual item loadings for the three subfactors
were strong and significant (.60 .88, p < .01) and the loadings of these subfactors to the
higher-order factor (gamma paths) were strong and signi®cant (.78 .97, p < .01), the
model showed marginal ®t. These marginal ®t values were due to within-factor corre-
lated measurement errors associated with ®ve items. Given these items also had re-
dundant wordings with several other items, they were deleted and a 23-item higher-
order structure was estimated.
The resulting model fit the data well: Comparative Fit Index (CFI) = .90, Incremental Fit
Index (IFI) = .90, and Root Mean Square Error of Aproximation (RMSEA) = .076. Fur-
thermore, the item loadings for the subfactors were strong and significant (.58 .85, p £
.01) and the loadings of these subfactors to the higher-order factor were strong (.79 .93, p
< .01). In sum, the reduced 23-item higher-order factor model was strongly supported,
supporting the hypothesis that the BCQ is a measure of a higher-order factor (body
checking) with three subfactors that are highly correlated (idiosyncratic checking, speci®c
body parts, and overall appearance).

Cross-Validation
To cross-validate this new higher-order structure of the BCQ, we used confirmatory
factor analysis to evaluate the 23-item BCQ on the original sample of 259 participants
from Study 1. The fit indices confirmed the adequacy of the 23-item BCQ: CFI = .90, IFI =
.90, and RMSEA = .074. Table 1 illustrates that the item loadings were strong and sig-
nificant (.49 .85, p < .01). The loadings of these three ®rst-order factors (gamma paths) to
the higher-order factor of body checking were strong and signi®cant (.76 .94, p < .01).
The intercorrelations of the three subfactors were .69, .73, and .81. In sum, the reduced 23-
item higher-order factor model cross-validated well in the original sample.

Internal Consistency
Internal consistency of the factor structure was examined using Chronbach alpha co-
efficients. Reliability estimates were r (165) = .88, .92, and .83 for overall appearance,
speci®c body parts, and idiosyncratic checking, respectively. These results are indicative
of adequate internal consistency.

Test-Retest Reliability
To assess the test-retest reliability of the measure, a second administration of the BCQ
was conducted with 54 participants selected randomly from the sample of 149 partici-
pants after an average of 2 weeks (range of 7 21 days). Pearson product-moment
correlation coefficients were calculated to determine the stability of the BCQ over time.
Test-retest reliability for the total BCQ score was r (54) =.94 (p < .01). Results from the test-
retest reliability analysis indicated very good temporal stability for each of the three BCQ
Development of Body Checking Questionnaire 329

Table 1. Factor loadings for the 23 BCQ items across the con®rmatory factor analysis (CFA)
sample and the cross-validation sample

Item Factor Loading

CFA sample Cross-Validation

Factor 1, overall appearance

Q3 .69 .72
Q5 .62 .70
Q8 .76 .78
Q11 .63 .63
Q12 .64 .68
Q13 .67 .74
Q15 .58 .62
Q17 .69 .67
Q21 .69 .49
Q22 .63 .61
Factor 2, speci®c body parts
Q1 .79 .65
Q2 .69 .72
Q6 .70 .70
Q9 .83 .85
Q10 .75 .72
Q14 .85 .79
Q16 .77 .76
Q19 .73 .81
Factor 3, idiosyncratic checking
Q4 .85 .76
Q7 .61 .71
Q18 .70 .73
Q20 .80 .62
Q23 .68 .64

factors. Reliability coef®cients for the three subfactors were r = .94, .91, and .90 (p < .01)
for overall appearance, speci®c body parts, and idiosyncratic checking, respectively.

Concurrent and Discriminant Validity

The total score of the 23-item version of the BCQ was correlated with the following
measures to establish concurrent and nomological validity: Body Shape Questionnaire
(BSQ; Cooper, Taylor, Cooper, & Fairburn, 1987); Eating Attitudes Test-26 (EAT; Garner
& Garfinkel, 1979); BIAQ (Rosen et al., 1991); and the Shipley Intelligence Test (Zachery,
1986).
The BSQ is a 34-item self-report questionnaire that assesses the desire to lose weight,
body dissatisfaction, and feelings of low self-esteem associated with body weight and
shape. Good reliability and validity of the BSQ have been demonstrated and a factor
analytic study has shown the BSQ to be a valid measure of body dissatisfaction (Wil-
liamson, Barker, Bertman, & Gleaves, 1995). The BSQ was selected as a measure of body
dissatisfaction to test the concurrent validity of the BCQ. It was hypothesized that scores
on the BSQ would be significantly correlated with scores on the BCQ.
The EAT is a 26-item self-report questionnaire that evaluates the attitudes and be-
haviors associated with AN. It was also shown to be of use in the assessment of BN
(Williamson, Anderson, Jackman, & Jackson, 1995). The EAT-26 has been found to have
good reliability and validity (Williamson et al., 1995) and was chosen to be a measure of
330 Reas et al.

eating disturbances. It was hypothesized that the EAT-26 would correlate positively with
the BCQ as a measure of the validity of the BCQ.
The BIAQ is a 19-item self-report questionnaire that measures behavioral avoidance of
situations that trigger anxiety about physical appearance. It was shown to have satis-
factory internal consistency, retest reliability, and validity. This measure was chosen as a
measure of behavioral avoidance related to negative body image and was used to test the
concurrent validity of the BCQ. As body avoidance and checking behaviors both stem
from a negative body image, a positive correlation with the BCQ was hypothesized.
The Shipley is a brief measure of general intelligence found to correlate significantly
with the full scale (Wechsler Adult Intelligence Scale Intelligence Quotient) WAIS-R IQ
scores (Shear, Harrison, & Sherman, 1986). The Shipley was used to test discriminant
validity, as it was hypothesized that a measure of intelligence would not correlate sig-
nificantly with a measure of body checking behaviors.

Results of Validity Studies

Results supported the concurrent validity of the BCQ. The total score on the BCQ was
strongly correlated with the BSQ and the EAT-26, r (159) = .86, p < .01 and r (160) = .70, p <
.01, respectively. These results indicated that body checking behaviors were associated
with more negative attitudes toward weight and shape as well as symptoms of eating
disorders. A positive correlation was found between the BIAQ and the BCQ, r (162) = .66,
p < .01. This correlation indicates that behaviors related to body image avoidance and
body checking rituals were correlated. The BCQ did not correlate with the Shipley Test of
Verbal Intelligence scale, supporting discriminant validity for the measure, r (159) =
).104, p > 19.

Group Differences on the BCQ

As an additional assessment of validity, total BCQ score from the Study 2 sample was
compared between eating disorder patients and college females The eating disorder
patients obtained significantly higher scores on the BCQ (M = 82.1, SD = 18) than the
college-aged females, M = 56.0, SD = 16, t (164) = 6.48, p < .001.
Nonclinical participants were divided into two groups based on scores on the BSQ as
an additional test of validity. We selected a cutoff score of 110, which is 1 SD above the
mean for normal controls, as the cutoff for which participants were classified as either
less concerned or very concerned about body shape and size. Based on this cutoff score,
99 participants were classified as less concerned and 45 participants were classified as
very concerned. The mean BCQ score of the very concerned group was 74.1 ‹ 14.9. This
score was significantly higher than the mean score for the less concerned group, M = 49.0
‹ 11.3, t (144) = 10.8, p < .001.
As a final test of validity, the nonclinical participants who reported currently being on
a weight loss diet were compared with nondieters. Of the 149 participants, 23 college
females indicated they were currently on a diet. A comparison of BCQ scores between
dieters and nondieters in the control group showed that, on average, the dieters reported
significantly more body checking behaviors than the nondieters (M = 71.3 ‹ 17 vs. 54.2 ‹
16.1, respectively). These results indicate that persons who are intentionally trying to
restrict their caloric intake tend to exhibit more body checking behaviors than those who
are not currently dieting, t (147) = 4.6, p < .001. Table 2 illustrates means and standard
Development of Body Checking Questionnaire 331

Table 2. Means and standard deviations of BCQ scores across different groups

Group Mean (SD) T-test (p-value)

Eating disorder patients 82.1 (18) 6.5

College female 56.0 (16)
Very concerned (high BSQ) 74.1 (15) 10.8
Less concerned (lower BSQ) 49.0 (11)
Dieters 71.1 (17) 4.6
Nondieters 54.2 (16)

Note: BCQ= Body Checking Questionnaire; BSQ = Body Shape Questionnaire


p< .01.

deviations of BCQ scores for eating disorder patients versus college-aged females, dieters
versus nondieters, and very concerned (high BSQ scores) versus less concerned (lower
BSQ scores) groups.

DISCUSSION

The BCQ was found to be a reliable and valid measure of body checking behaviors. It is
a 23-item self-report questionnaire that can be administered easily and requires only a
few minutes to complete. Initial results of the factor structure indicate that the BCQ is a
single measure of a higher-order construct comprising checking behaviors related to
overall appearance, specific body parts, and idiosyncratic checking.
Higher scores on the BCQ were associated with more intense body dissatisfaction, fear
of fatness, body image avoidant behaviors, and general eating disturbances, as indicated
by significant correlations with the BSQ, BIAQ, and the EAT. The BCQ was shown to
have good internal consistency and test-retest reliability. Preliminary evidence was
demonstrated for the ability of the BCQ to discriminate eating disorder patients from
controls, as well as dieters from nondieters.
A negative body image is a multidimensional phenomenon, as evidenced by behav-
ioral features as well as perceptual and attitudinal components. Because only a few
measures have been developed that assess the behavioral manifestations of a negative
body image, the BCQ represents an important addition to the existing literature on the
assessment of body image. To our knowledge, the BCQ is the only psychometrically
validated measure of body checking behaviors that has been described in the research
literature.
Ritualistic checking of body shape and size may serve to reinforce body dissatisfaction
in eating disorder patients by directing excessive attention to the body. Thus, the iden-
tification of body checking behaviors is important to address when treating eating dis-
order patients. A reduction or elimination of checking behaviors is desirable and can be
achieved by response prevention or self-management techniques (Rosen, 1997) or by the
general reduction of shape and weight concerns. In addition to maintenance of eating
disorder symptoms, body checking behaviors may play a role in the etiology of eating
disorders. Therefore, the measurement of body checking habits may prove useful in the
prevention of clinical disorders.
The significant relationship between checking behaviors and avoidance behaviors is
interesting and worth further investigation. It has been argued that checking rituals of
oneÕs weight and shape may become highly aversive and over time, eventually become
totally avoided (Fairburn et al., 1999). For example, a woman who checks her weight/
332 Reas et al.

shape ritualistically may begin to wear baggy clothing and avoid mirrors or scales al-
together. Thus, avoidance behaviors can lead to the maintenance of eating disorder be-
haviors in a similar way as repeated checking, because avoidance may allow beliefs that
the individual is failing at her weight-control efforts to persist. Because these behaviors
are often distressing and time- consuming, the measurement of checking and avoidance
behaviors could be included as part of an overall assessment of the various aspects of
body image. In conclusion, the BCQ may be useful as a measure of body checking
behaviors in the research and treatment of eating disorders and in research on body
image in nonclinical samples.

We are grateful to Roz Shafron for her comments on an earlier draft of this paper. We also extend
our gratitude to Drew Anderson, Kevin Thompson, Cheryl Funsch, Paula Varnado, David Gleaves,
Phil Watkins, Monique Smeets, Erich Duchman, and Rene Bruno for their help in rating the content
of the BCQ items.

REFERENCES
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.).
Washington, DC: Author.
Cooper, P.J., Taylor, M.J., Cooper, Z., & Fairburn, C.G. (1987). The development and validation of the Body
Shape Questionnaire. International Journal of Eating Disorders, 6, 485 494.
Fairburn, C.G., Shafron, R., & Cooper, Z. (1999). A cognitive behavioural theory of anorexia nervosa. Behaviour
Research and Therapy, 37, 1 13.
Garner, D.M., & Gar®nkel, P.E. (1979). The Eating Attitudes Test: An index of the symptoms of anorexia nervosa.
Psychological Medicine, 9, 273 279.
Garrow, J.S. (1983). Indices of adiposity. Review of Clinical Nutrition, 3, 697 708.
Hair, J.F., Anderson, R.E., Tatham, R.L., & Black, W.C. (1999). Multivariate data analysis (5th ed.). Englewood
Cliff, NJ: Prentice Hall.
Haynes, S.N., Richard, D.C., & Kubany, E.S. (1995). Content validity in psychological assessment: A functional
approach to concepts and methods. Psychological Assessment, 7, 238 247.
Kutlesic, V., Williamson, D.A., Gleaves, D.H., Barbin, J.M., & Murphy-Eberenz, K.P. (1998). The Interview for the
Diagnosis of Eating Disorders IV: Application to DSM-IV diagnostic criteria. Psychological Assessment, 10,
41 48.
Rosen, J.C. (1997). Cognitive-behavioral body image therapy. In D.M. Garner & P.E. Gar®nkel (Eds.), Handbook
of treatment for eating disorders (pp. 188 201). New York: Guilford Press.
Rosen, J., Srebnik, D., Saltzberg, E., & Wendt, S. (1991). Development of a Body Image Avoidance Questionnaire.
Psychological Assessment, 3, 32 37.
Shear, J.M., Harrison, W.R., & Sherman, C.J. (1986). Estimating WAIS IQ of neuropsychiatric patients at three
educational levels. Psychological Reports, 58, 947 950.
Thompson, J.K. (1996). Assessing body image disturbance: Measures, methodology, and implementation. In J.K.
Thompson (Ed.), Body image, eating disorders, and obesity (pp. 49 83). Washington, DC: American Psy-
chological Association.
Thompson, J.K., Heinberg, L., & Marshall, K. (1994). The Physical Appearance Behavior Avoidance Test (PA-
BAT): Preliminary ®ndings. The Behavior Therapist, 17, 9 10.
Thompson, J.K., Heinberg, L., & Tantleff, S. (1991). The Physical Appearance Comparison Scale (PACS). The
Behavior Therapist, 1, 174.
Williamson, D.A. (1990). Assessment of eating disorders: Obesity, anorexia, and bulimia nervosa. New York:
Pergamon Press.
Willianson, D.A., Anderson, D.A., Jackman, L.P., & Jackson, S.R. (1995). Assessment of eating disordered
thoughts, feelings, and behaviors. In: Methods for the assessment of eating behaviors and weight related
problems. (Allison, A.J., ed). Sage Publications: Newbury Park, CA. p 347 386.
Williamson, D.A., Barker, S., Bertman, L.J., & Gleaves, D. (1995). Body image, body dysphoria, and dietary
restraint: Factor structure in nonclinical subjects. Behavior Research and Therapy, 33, 85 93.
Williamson, D.A., Muller, S., Reas, D.L., & Thaw, J.M. (1999). Cognitive bias in eating disorders: Implications for
theory and treatment. Behavior Modi®cation, 23, 556 578.
Zachery, R.A. (1986). Shipley Institute of Living Scale, revised manual. Los Angeles, CA: Western Psyclogical
Services.
 APPENDIX

Body Checking Questionnaire

Circle the number which best describes how often you engage in these behaviors at the
present time.
1 = never
2 = rarely
3 = sometimes
4 = often
5 = very often

Table Appendix (Continued)

1. I check to see if my 1 2 3 4 5
thighs spread when I'm sitting down.
2. I pinch my stomach to measure fatness. 1 2 3 4 5
3. I have special clothes which 1 2 3 4 5
I try on to make sure they still ®t.
4. I check the diameter of my wrist 1 2 3 4 5
to make sure it's the same size as before.
5.1 check my re¯ection in glass 1 2 3 4 5
doors or car windows to see how I look.
6. I pinch my upper arms to 1 2 3 4 5
measure fatness.
7. I touch underneath my chin to 1 2 3 4 5
make sure I don't have a ``double chir.''
8. I look at others to see how my 1 2 3 4 5
body size compares to their body size.
9. I rub (or touch) my thighs while 1 2 3 4 5
sitting to check for fatness.
10. I check the diameter of my 1 2 3 4 5
legs to make they're the same size as before.
11. I ask others about their weight or 1 2 3 4 5
clothing size so I can compare
my own weight/size.
12. I check to see how my bottom 1 2 3 4 5
looks in the mirror.
13.1 practice sitting and standing 1 2 3 4 5
in various positions to see how
I would look in each position.
14. I check to see if my thighs rub together. 1 2 3 4 5
15. I try to elicit comments from 1 2 3 4 5
others about how fat I am.
16. I check to see if my fat jiggles. 1 2 3 4 5
17. I suck in my gut to see what it is like 1 2 3 4 5
when my stomach is completely ¯at.
18. I check to mare sure my rings ®t the 1 2 3 4 5
same way as before,
19. I look to see if I have cellulite on my 1 2 3 4 5
thighs when I am sitting.
20. I lie down on the ¯oor to see if I can feel my 1 2 3 4 5
bones touch the ¯oor.
21. I pull my clothes as tightly as possible around 1 2 3 4 5
myself to see how I look.
22. I compare myself to models on TV or in 1 2 3 4 5
magazines.
23. I pinch my cheeks to measure fatness. 1 2 3 4 5

Note: The BCQ is in questionnaire format and can therefore be administered in either individual or group
settings. Completion time for the measure is approximately 5-10 minutes. To score the total BCQ, simply sum all
the items. To calculate the overall appearance scale, sum the following items: 3, 5, 8, 11,12, 13, 15, 17, 21, 22 . To
calculate the speci®c body parts scale, sum the following items: 1, 2, 6, 9,10, 14, 16, 19. To calculate the
idiosyncratic checking scale, sum the following items: 4, 7, 18, 20,23.