Sie sind auf Seite 1von 11

Development and Validation of a New Figural Drawing Scale for Body-Image Assessment: The BIAS-BD

Rick M. Gardner and Leah M. Jappe


University of Colorado, Denver m m

Lisa Gardner
Denver, Colorado

This study describes the development and validation of a new and improved body-image assessment tool, the BIAS-BD, based on known body dimensions. The scales construction consisted of 17 male and 17 female contour-line drawings that used known anthropometric body dimensions of shoulder, chest, waist, hip breadth, thigh breadth, and upper leg breadth. The gural drawings correspond to a series of body weights ranging from 60% below the known average to 140% above average. Differences between gural drawings represented a 5% change in body weight. Participants were 207 undergraduates, including 66 men and 141 women, who selected drawings that reected their perceived size and their ideal size. Retesting occurred after a 2-week interval and resulted in testretest reliability values of r 5 .86 for actual perceived size, r 5 .72 for ideal size, and r 5 .76 for body dissatisfaction (po.005). There were no signicant differences in reliability values between genders. Mean differences in perceived size, ideal size, and body dissatisfaction between the two test administrations were small. Concurrent validity, measured as the correspondence between perceived and reported size, was r 5 .76 (po.005). Participants slightly overestimated their perceived body size, with women overestimating signicantly more. Unlike existing scales, the present scale uses gural drawings based on known body dimensions and has better reliability and validity. It avoids several problems inherent in existing scales, including scale coarseness, the presence of ethnic facial and body features, and the

Rick M. Gardner and Leah M. Jappe, Department of Psychology, University of Colorado, Denver. Lisa Gardner is an architect at Slater-Paul Architects, Denver, CO. The authors thank Elizabeth Ann Gardner and Margie Krest for their editorial assistance. Correspondence concerning this article should be addressed to: Rick M. Gardner, Department of Psychology, University of Colorado, Campus Box 173, P.O. Box 173364, Denver, CO 80217; e-mail: rick.gardner@ucdenver.edu

JOURNAL OF CLINICAL PSYCHOLOGY, Vol. 65(1), 113--122 (2009) & 2008 Wiley Periodicals, Inc. Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jclp.20526

114

Journal of Clinical Psychology, January 2009

lack of documented reliability and validity values. & 2008 Wiley Periodicals, Inc. J Clin Psychol 65:113122, 2009. Keywords: body image; BID; gural scales; contour line drawings; BMI; obesity; body size distortion; body dissatisfaction

The investigation of body image disturbance (BID) has increased in recent years, largely due to the widespread acknowledgment of its critical role in eating disorders and obesity. BID involves some degree of dissatisfaction with appearance. Thompson (1995) dened BID as . . . any form of affective, cognitive, behavioral, or perceptual disturbance that is directly concerned with an aspect of physical appearance (p. 120). Both researchers and clinicians now agree that the assessment and evaluation of BID is crucial for treatment programs that target obesity (Thompson, 1995). A variety of measures that assess the different aspects of BID include gural scales, questionnaires, and examination of cognitive, affective, and behavioral functioning. Figural stimuli, also referred to as silhouette or contour line drawings, are the most widely adopted measures of one aspect of BID; namely, dissatisfaction with body size (Thompson, 1995). A variety of questionnaires have been developed to measure other aspects of BID, such as dissatisfaction with individual body regions or with body shape (Thompson & Van Den Berg, 2002). Figural drawings typically consist of a set of discrete schematic gures, numbering between 5 and 12, that represent the frontal view of both male and female gures and that display body sizes ranging from very thin to very wide. To test BID, participants select the gures that represent their perceived current size as well as their ideal size. The difference between the perceived size and the ideal size constitutes the discrepancy index commonly used to represent body size dissatisfaction. Thompson and Gray (1995) and Thompson, Heinberg, Altabe and Tantleff-Dunn (1999) reviewed the most widely used and psychometrically sound gural rating scales, including information about reliability and standardization samples if available. More recently, Gardner (2001) conducted a similar review of instruments designed specically for children and adolescents. Thompson and Gray (1995) reviewed 22 gural drawing scales and found that validity values were given for only three scales, with concurrent validity demonstrated for only two scales. The Contour Drawing Rating Scale (Thompson & Gray, 1995) was the only scale designed for adults that assessed whether the participants actual body size corresponds with participants perceived body size. In addition, numerous methodological problems exist with the use of gural drawing scales. Thompson and Gray (1995) and Gardner, Friedman, & Jackson (1998) discussed several of these problems. First, scale coarseness exists when the test requires participants to select from only a small number of drawings. This limits precision in measurement and can result in lost information (Gardner et al., 1998). This methodological issue causes articial increases or decreases in moderated regression effect size. Gardner et al. also demonstrated that most participants select from a small subset of the gures available. This restriction in the range of responses can potentially inate testretest reliability scores and further exacerbate the issue of scale coarseness. Another criticism of contour line drawings involves the method of presentation. Usually, all gural stimuli are presented on a single sheet of paper with the gures arranged in ascending size from left to right. As mentioned previously, participants
Journal of Clinical Psychology DOI: 10.1002/jclp

New Figural Drawing Scale

115

typically mark the gural drawings that they think represent their actual and ideal sizes; however, Gardner et al. (1998) argued that such techniques will likely produce spuriously high testretest reliability estimates because the participants have little difculty remembering which gures they marked previously, particularly when they are shown in an ascending order. Only a few investigators have randomly arranged the presentation order. Doll, Ball, and Willows (2004) recently demonstrated that these two methodologies give differing selections of perceived and ideal size when presenting gures as an ordered or unordered array. Participants judged their perceived size as smaller when viewing the gures in ascending order of size. Doll et al. concluded that the manner of gure presentation might be responsible, in part, for the discordant results reported in the body image assessment literature. Furthermore, a recent study by Nicholls, Orr, Okubo, and Loftus (2006) on the effect of spatial biases on responses to Likert-type scales found a bias to respond to the left of the scales midpoint when values are placed in an ascending order. When gural drawing scales are presented in this fashion, they are always arranged with smaller gures to the left and increasingly larger gures to the right. This raises the distinct possibility that participants will respond with a bias towards thinner gures. Additional bias can occur when gural stimuli scales are drawn with facial and body features that reect obvious Caucasian ethnicity. Thompson (1996) and Altabe (1996) recommended gural line drawings that omit these details make the scales more appropriate for a broad range of ethnic groups. Another difculty with all existing scales, and one that this study specically targets, concerns the level of distortion that does not correspond with the known changes in body size that occur when an individual gains or loses weight. Artists have constructed almost all existing gural scales to correspond with what they subjectively believe represent a variety of weights rather than known body dimensions for varying weights. This format makes it difcult to assess the validity of the scales. Thompson and Gray (1995) noted that several scales display disproportionate arms and legs, differential thickness between right and left arms, or a lack of separation between the arms and bodies in obese drawings. In this study, we examine the reliability and validity of a new gural scale named the Body Image Assessment ScaleBody Dimensions (BIAS-BD) that uses known anthropometric physical dimensions of adult men and women. It specically addresses the methodological shortcomings mentioned previously that are present in currently published gural drawing scales. This scale is designed to allow investigators to measure body dissatisfaction by examining the discrepancy between perceived body size and self-determined ideal body size. Furthermore, it allows for the measurement of the individuals body-size distortion by examining the discrepancy between self-reported body size and perceived body size.

Method Participants Two hundred seven undergraduate university students (141 women, 66 men) volunteered for the study, without compensation, and completed two testing sessions. The mean body mass index (BMI) was 22.57 (SD 5 3.93) for women and 23.49 (SD 5 4.50) for men. Demographic data were not collected to ensure anonymity.
Journal of Clinical Psychology DOI: 10.1002/jclp

116

Journal of Clinical Psychology, January 2009

Materials A set of 17 gural line drawings representing adult human male and female gures was generated from anthropometric surveys and stereophotometric data developed by the United States Air Force Materiel Command. The Air Force Research Laboratory developed a computer program called GEBOD (Generator of Body Data) that produces human body-dimension data for nearly any size man or woman based on individual height and weight (Cheng, Obergefell, & Rizer, 1994). For any given height and weight, the program uses regression equations to generate 29 body dimensions, including dimensions such as head length, neck circumference, shoulder breath, waist breadth, and so on. The regression equations have been developed from the data collected by the Air Force from 4,325 participants. No breakdown on the number of male and female participants was provided. To illustrate, Table 1 shows sample data for six body regions of an adult woman of average height for weights up to 30% below and above average. To construct the present scale, we used a frontal view photograph (Carter & Heath, 1990) of an adult man and woman with height and weight equal to reported mean values provided by the most current National Center for Health Statistics (NCHS; 2004). According to this report, the average man aged 20 to 74 years is 69.5 in. tall with a weight of 191 lb, or a BMI of 27.8. A woman in the same age range has a mean height of 64 in. with a weight of 164.3 lb, or a BMI of 28.2. A contour line drawing of these prototypical adults was generated from the photographs. The dimensions of shoulder, chest, waist, and hip breadth as well as thigh and upper leg circumference (converted to leg diameter) were then used to adjust the line drawing to known dimensions generated by the GEBOD program. The gural drawings were created using the digital drafting program AutoCAD. The contour line drawing of the average-size person was scanned and imported as an image into the AutoCAD program, scaled to accurate size, and then traced as a continuous contour drawing. Having established the contour of the average person, guidelines were drawn across the gure at the shoulders, waist, hips, and thighs. These lines established common points of manipulation so that all gures generated from the average-size person were scaled at the same place. The contour lines were then stretched at these set locations to the dimensions provided by the GEBOD data. Minor morphing adjustments around these points were required to create a smooth body shape. This method allowed for gures to reect accurate shapes of key dimensions rather than enlarging or shrinking the entire image by a scale percentage, as some previous investigators have done (Gardner, Stark, Jackson, & Friedman, 1999). Each of the four body-area dimensions were increased or decreased by Table 1
Body Dimensions in Inches for Body Regions of a Female With Average Height at Weights up to 730% Average Weight
Weight (kg) 115.0 131.4 147.9 164.3 180.7 197.2 213.6 (52.2) (59.6) (67.1) (74.5) (82.0) (89.5) (96.9) %7 Avg. 30 20 10 0 110 120 130 Shoulder breadth 13.96 14.18 14.41 14.63 14.86 15.08 15.31 Chest breadth 10.58 11.16 11.74 12.31 12.89 13.46 14.04 Waist breadth 9.03 9.65 10.28 10.90 11.52 12.15 12.77 Hip breadth 13.24 13.94 14.65 15.35 16.06 16.76 17.47 Thigh breadth 6.56 7.07 7.59 8.42 8.61 9.12 9.63 Upper leg breadth 5.47 6.04 6.22 5.59 6.97 7.35 7.70

Journal of Clinical Psychology

DOI: 10.1002/jclp

New Figural Drawing Scale

117

Figure 1. The Body Image Asessment Scale-Body Dimensions (BIAS-BD) with gural line drawings representing adult men and women with weights ranging from 60% below to 140% above average. Dimensions of the shoulder, chest, waist, hip breadth, thigh breadth, and upper leg breadth were determined from anthropometric data by Cheng et al. (1994) at the U.S. Air Force Materiel Command.

successive 5% amounts to create each level of distortion. The resulting gural drawings are illustrated in Figure 1. Figural drawings were constructed for weight values ranging from 60% below average to 140% above average. Corresponding BMI values for these weight percentages ranged from 16.68 to 38.92 for men and 16.92 to 39.48 for women. Clothing, hair, and facial features were deliberately omitted from the drawings to avoid possible ethnic identication and to focus the observers attention on body shape and size. Procedure Data collection took place in various undergraduate psychology classes. Participants were informed of the purpose of the research project and its approval by the universitys human subjects research committee. Students also were informed that participation was voluntary and that all data would be kept condential. During class time, students were handed a copy of the 17 gural drawings appropriate for their gender. The drawings were randomly arranged by size into two horizontal rows on a single 8.5 11 in. sheet of paper, with the same random order used for all participants. Participants selected and marked the gural drawings that they felt best represented their actual size (marked by the letter A) and the size they would like to be ideally (marked by the letter I). Participants also were asked to list their height and weight on the scale for assessing validity. A four-digit identication number was included so the experimenter could return in 2 weeks to collect additional data to match scores for testretest reliability. The same procedure was followed during the second administration. A different random order of the gural drawings for all participants was used for the second phase of the study. Results Data were screened for outliers using the NCSS software (Hintze, 1996). Figural rating choices were converted to BMI values for all analyses. BMI is found by BMI 5 weight (kg)/height (m)2 or BMI 5 weight (lb)/[height (in.)]2 703.
Journal of Clinical Psychology DOI: 10.1002/jclp

118

Journal of Clinical Psychology, January 2009

To examine testretest reliability, the BMI values corresponding to the gural drawings that the participants selected as their perceived size were compared for the two test administrations. For all participants, the testretest correlation was r 5 0.86 (po.005) with a condence interval (CI) at .95 5 .82r.86r.89. The obtained r has a value higher than the minimum value of .80 recommended by Carmines (1990). The coefcients were almost identical for women, r 5 .85 (CI .95 5 .80r.85r.89) and men, r 5 .88 (CI .95 5 .81r.88r.92). A Z test between independent group correlations (Kanji, 1993) revealed no signicant difference between genders (Z 5 .79, p4.05). Testretest reliability also may be examined by comparing the differences in perceived BMI values that participants selected between the two test administrations at Testing Sessions 1 and 2. The average discrepancy was M 5 .01 (SE 5 .24), indicating that participants were very consistent in judging their perceived BMI size on the two test administrations. Mean values were almost identical for men (M 5 0.0, SE 5 .40) and women (M 5 .02, SE 5 .30). Testretest reliability of the BMI values corresponding to the gural drawing selected as the ideal size also was compared for the two test administrations. For all participants, r 5 .72, po.005 (CI .95 5 .65r.72r.78), a value lower than the recommended minimum. The coefcient for men, r 5 .75 (CI .95 5 .62 r.75r.84) was only slightly higher and not signicantly different than the coefcient for women, r 5 .64 (CI .95 5 .53r.64r.73; Z 5 1.41, p4.05). This tendency may reect previously reported ndings that the stability of body-size perceptions is typically higher in men than in women, particularly women with eating disorder tendencies (Melnyk, Cash, & Janda, 2004; Rudiger, Cash, Roehrig, & Thompson, 2007). Comparing the discrepancies in ideal size, BMI selected for all participants between administration of the two tests gave an average mean difference of M 5 .45, SE 5 .26. Similar values were obtained for men (M 5 .58, SE 5 .52) and women (M 5 .40, SE 5 .29). Testretest reliability for body dissatisfaction (perceived BMIideal BMI) was r 5 .81, po.005 (CI .95 5 .76r.81r.85). Only a small and nonsignicant difference was found between men, r 5 .85 (CI .95 5 .77r.85r.91), and women, r 5 .79 (CI .95 5 .72r.79r.84; Z 5 1.22, p4.05). A comparison of differences in body dissatisfaction measured in BMI units for all participants between administration of the two tests gave an average of M 5 .33, SE 5 .43. A difference existed between men (M 5 .88, SE 5 .44) and women (M 5 .14, SE 5 .45). Overall, the scale shows an acceptably high testretest reliability for perceived actual size and body dissatisfaction. This is true when examining both testretest correlations as well as the magnitude of differences between the two test administrations. For ideal body size, the reliability only approached the acceptable value of .80. To test for concurrent validity, the BMI values for gural drawings selected as the participants perceived actual body size were correlated with the participants selfreported BMI. If the scale is valid, one would expect a correspondence between the observers judgments of their body size as reected in the BMI of the gural drawing they selected and their reported BMI. For all participants, this correlation was r 5 .76, po.005 (CI .95 5 .70r.76r81). For women, the correlation was r 5 .77, po.005 (CI .95 5 .71r.77r.82). For men, the correlation was r 5 .80, po.005 (CI .95 5 .69r.80r.87), a difference which was not signicant (Z 5 .52, p4.05). For both men and women, this result demonstrates the scales acceptably high concurrent validity.
Journal of Clinical Psychology DOI: 10.1002/jclp

New Figural Drawing Scale

119

The present scale also permits an analysis of perceived body-size distortion by comparing the self-reported BMI values with the BMI values that correspond to the gural drawings participants chose as reecting their actual size. The difference between these two values was M 5 2.21 (SD 5 3.82), showing that overall, participants overestimated their body size by slightly more than 2 BMI values. Women overestimated their size to a greater extent (M 5 2.93, SD 5 3.64) than did men (M 5 .69, SD 5 3.78), a difference that proved signicant, t(205) 5 2.84, po.01. This overestimation of size in men and women is in general agreement with measures of body-size distortion obtained with several studies in our laboratory using the video distortion methodology (for a review, see Gardner & Boice, 2004), although a signicant difference between genders has not usually been found. Finally, the relationship between the actual perceived size and the ideal perceived size was examined. It could be expected that as actual perceived size increases, a concomitant increase in ideal size also occurs. The correlation for all participants was r 5 .43, po.005 (CI .95 5 .29r.43r.51). Similar relationships were found for men, r 5 .44, po.005 (CI .95 5 .16r.44r.58), and for women, r 5 .54, po.005 (CI .95 5 .41r.54r.65), a difference that was not signicant (Z 5 .87, p4.05). These ndings indicate that as participants viewed their perceived body as larger, their expressed ideal size also increased. Discussion Overall, the BIAS-BD demonstrates good testretest reliability of perceived body size both when measured as a function of the correspondence of scores between the two test administrations as well as when examining score differences between test administrations. The testretest reliability of ideal scores is somewhat lower. The higher testretest values obtained when judging perceived size, as compared to ideal size, is not unexpected. Numerous transient factors such as recent exposure to thinideal media messages, mood state, and so on can momentarily affect ones judgment of ideal size (Thompson et al., 1999). The present scale also shows satisfactory concurrent validity, with correlations reaching or closely approximating the .80 standard for both men and women. In addition, the scale demonstrates that both men and women overestimate their body size, an outcome that concurs with other measures of body-size distortion such as the video-distortion technique. A surprising result is that women signicantly overestimate their size to a greater extent, a nding that has not been consistently observed in our previous studies using the video-distortion technique. In a planned future study, we will make a direct comparison of mens and womens overestimations using both measurement techniques. The present gural drawing scale offers several advantages over existing scales. Most existing scales have no testretest or validity ratings, including several that are commonly cited in body-image studies. Based on the review by Thompson and Gray (1995) of 22 existing scales, only 6 presented data on testretest reliability. Of those six, only three have coefcients approaching that of the presented scale. Validity data are presented for only 3 of the 22 scales reviewed. The Thompson and Gray body-image assessment scale, which also used self-reported height and weight, is the only existing scale to present both reliability and concurrent validity data for a scale designed for adults. The testretest reliability of the Thompson and Gray scale was r 5 .78 over a 1-week period while concurrent validity was found to be r 5 .71. Furthermore, these data are based on a relatively small sample size of 32
Journal of Clinical Psychology DOI: 10.1002/jclp

120

Journal of Clinical Psychology, January 2009

participants. Schonbuch and Schell (1967) and Singh (1993) demonstrated the validity of their scales by reporting the number of raters who were able to correctly rank order the size of the gures. Four of ve were able to do so with the Schonbuch and Schell scale while all 69 raters were able to do so with the Singh scale. However, both of these scales have limited use in that the Schonbuch and Schell scale represents men only while the Singh scale consists only of female drawings. Concurrent validity was chosen as a better measure of validity for the current scale. The BIAS-BD scale demonstrates greater reliability and concurrent validity with a signicantly larger number of participants than do existing scales to date. In addition, gural drawings representing both men and women are given. The scale offers several other advantages as well. Since it is based on known physical changes that are present at various weights, it provides a more realistic representation of the size and shape of the human body at different weights. Following the recommendation of Thompson (1996), the drawings show no hair, facial features, or clothing items, unlike most previous scales, and therefore focus the observers attention primarily on the size and shape of the drawing. These drawings also should aid in making the scale more appropriate across differing age and ethnic groups (Altabe, 1996; Thompson, 1996). Even though some investigators believe that such additional accents make the drawings more realistic, the high reliability and validity of the present scale clearly speaks against the necessity of those features. An additional advantage of the present scale is the presence of a larger number of drawings representing a greater range of body sizes than those in all but one existing published scale. The 17 drawings represent a range of body sizes that present equal weight increments from 60 to 140% of average size. This range should easily encompass the variety of weights found in most studies and address the problem of scale coarseness described earlier (Gardner et al., 1998). Another advantage, not possessed by existing scales, is the ability to determine perceived body-size distortion. Since each of the gural drawings corresponds to a specic BMI and weight, the respondents choice of gures can be compared to their self-reported and/or actual BMI and weight to determine an indication of body-size over- or underestimation. Because the participant must choose 1 of 17 discrete gures, the body-size-distortion measurements likely will be less precise than those obtained with video-distortion methods where participants adjust their perceived size along a continuous scale. Nevertheless, body-size distortion with this scale generally agrees with measurements found by the TVvideo techniques. Unlike many previous studies, the gural stimuli were presented to participants in random order, as recommended by Gardner et al. (1998) and Doll et al. (2004). As noted earlier, presenting stimuli in an ascending order of size may inate testretest scores (Gardner et al., 1998) and can signicantly affect test results (Doll et al., 2004). One limitation facing the present study relates to self-estimates of body weight and height given by participants. It has been well documented that self-estimates may be somewhat inaccurate (see meta-analysis by Bowman & DeLucia, 1992), such that individuals typically overestimate height and underestimate weight. In addition, Villanueva (2001) showed that numerous variables affect such estimates, including sex, age, race/ethnicity, marital status, income, and activity level, and that these variables often differently affect men and womens estimates. In a study using a large sample of college students, Imrhan, Imrhan, and Hart (1996) also found that males and females overestimated height and underestimated weight, but concluded that the discrepancies were sufciently small to allow using self-estimates in place of
Journal of Clinical Psychology DOI: 10.1002/jclp

New Figural Drawing Scale

121

measurements. Due to the conclusions of Imrhan et al., this study based its validity results on self-estimates of body size correlated with each individuals perceived body size. Nevertheless, inaccuracy of self-estimates and inaccuracy in individual perception could affect the validity of the present results, and a subsequent study will examine these factors. The lack of demographic information collected, for purposes of maintaining anonymity, is another limitation facing the present study. Subsequent research will examine any changes that exist in reliability and validity scores between various age, ethnic, and socioeconomic groups. Such information may help strengthen the argument regarding the ability to generalize from a gural scale absent of facial or body features. Finally, one limitation of the validity data is that the correlations reect only the relationship between perceived and self-reported body size. This is not as direct a measurement as would have occurred if actual height and weight dimensions had been taken. In addition, perceived body size is a psychological construct that is inuenced by numerous psychological variables that could potentially impact validity measures. In conclusion, the investigators believe that the present scale is superior to all existing scales used in body-image research. It has superior reliability and validity, permits measures of body-size distortion, and is the only scale based on known physical dimensions of body regions in adults at varying weight levels. The scale will be a useful research tool for investigators interested in studying a variety of aspects of adult body image and BID.

References
Altabe, M. (1996). Issues in the assessment and treatment of body image disturbance in culturally diverse populations. In J.K. Thompson, (Ed.), Body image, eating disorders, and obesity (pp. 129148). Washington, DC: American Psychological Association. Bowman, R.L., & DeLucia, J.L. (1992). Accuracy of self-reported weight: A meta-analysis. Behavior Therapy, 23, 637655. Carmines, E.G. (1990). The statistical analysis of overidentied linear recursive models. Quality and Quantity, 24, 6585. Carter, J.E.L., & Heath, B.H. (1990). SomatotypingDevelopment and applications. Cambridge, England: Cambridge University Press. Cheng, H., Obergefell, L., & Rizer, A. (1994). Generator of Body Data (GEBOD) Manual, Armstrong Laboratory Report No. AL/CF-TR-19940051, Wright-Patterson Air Force Base, OH. Doll, M., Ball, G.D.C., & Willows, N.D. (2004). Rating of gures used for body image assessment varies depending on the method of gure presentation. International Journal of Eating Disorders, 35, 109114. Gardner, R.M. (2001). Assessment of body image disturbance in children and adolescents. In J.K. Thompson & L. Smolak, (Ed.), Body image, eating disorders, and obesity in youth (pp. 193213). Washington, DC: American Psychological Association. Gardner, R.M., & Boice, R. (2004). A computer program for measuring body size distortion and body dissatisfaction. Behavior Research Methods, Instruments, & Computers, 36(1), 8995. Gardner, R.M., Friedman, B.N., & Jackson, N.A. (1998). Methodological concerns when using silhouettes to measure body image. Perceptual and Motor Skills, 86, 387395.
Journal of Clinical Psychology DOI: 10.1002/jclp

122

Journal of Clinical Psychology, January 2009

Gardner, R.M., Stark, K., Jackson, N.A, & Friedman, B.N. (1999). Development and validation of two new scales for assessment of body-image. Perceptual and Motor Skills, 89, 981993. Hintze, J.L. (1996). Number Cruncher Statistical System, Version 6.0.2.1. Kaysville, UT: NCSS. Imrhan, S.N., Imrhan, V., & Hart, C. (1996). Can self-estimates of body weight and height be used in place of measurement for college students? Ergonomics, 39, 14451453. Kanji, G.K. (1993). 100 statistical tests. Thousand Oaks, CA: Sage. Melnyk, S.E., Cash, T.F., & Janda, L.H. (2004). Body image ups and downs: Prediction of intra-individual level and variability of womens daily body image experiences. Body Image, 1, 225236. National Center for Health Statistics. (2004). Mean body weight, height, and body mass index, United States 19602002. Advanced Data No 347, PHS 20051250 (pp. 118). Hyattsville, MD: Author. Nicholls, M.E.R., Orr, C.A., Okubo, M., & Loftus, A. (2006). Satisfaction guaranteed: The effect of spatial biases on responses to Likert scales. Psychological Science, 17, 10271028. Rudiger, J.A., Cash, T.F., Roehrig, M., & Thompson, J.K. (2007). Day-to-day body-image states: Prospective predictors of intra-individual level and variability. Body Image, 4, 110. Schonbuch, S.S., & Schell, R.E. (1967). Judgments of body appearance by fat and skinny male college students. Perceptual and Motor Skills, 24, 9991002. Singh, D. (1993). Adaptive signicance of female physical attractiveness: Role of waist-to-hip ratio. Journal of Personality and Social Psychology, 65, 293307. Thompson, J.K. (1995). Assessment of body image. In D.B. Allison, (Ed.), Handbook of assessment methods for eating behaviors and weight related problems (pp. 119144). Thousand Oaks, CA: Sage. Thompson, J.K. (1996). Assessing body image disturbance: Measures, methodology, and implementation. In J.K. Thompson, (Ed.), Body image, eating disorders, and obesity (pp. 4981). Washington, DC: American Psychological Association. Thompson, M., & Gray, J. (1995). Development and validation of a new body image assessment scale. Journal of Personality Assessment, 64, 258269. Thompson, J.K., Heinberg, L.J., Altabe, M., & Tantleff-Dunn, S. (1999). Exacting beauty: Theory, assessment, and treatment of body image disturbance. Washington, DC: American Psychological Association. Thompson, J.K., & Van Den Berg, P. (2002). Measuring body image attitudes among adolescents and adults. In T.F. Cash & T. Pruzinsky, (Ed.), Body image: A handbook of theory, research, and clinical practice (pp. 142154). New York: Guilford Press. Villanueva, E.V. (2001). The validity of self-reported weight in U.S. adults: A population based cross-sectional study. BMC Public Health, 1(6), 110.

Journal of Clinical Psychology

DOI: 10.1002/jclp