- Email: [email protected]
Body Shape Questionnaire: Psychometric properties of the short version (BSQ-8C) and norms from the general Swedish population
Body Image 9 (2012) 547–550
Contents lists available at SciVerse ScienceDirect
Body Image journal homepage: www.elsevier.com/locate/bodyimage
Brief research report
Body Shape Questionnaire: Psychometric properties of the short version (BSQ-8C) and norms from the general Swedish population Elisabeth Welch, Marcus Lagerström, Ata Ghaderi ∗ Uppsala University, Department of Psychology, Uppsala, Sweden
a r t i c l e
i n f o
a b s t r a c t
Article history: Received 23 November 2011 Received in revised form 21 April 2012 Accepted 24 April 2012 Keywords: Body dissatisfaction Body Shape Questionnaire BSQ-8C
There is a need for brief, sensitive, and reliable instruments for contextual assessment of body dissatisfaction. Suitable selection and optimal use of psychological assessment instruments requires appropriate psychometric as well as normative data. Two studies were conducted to investigate the psychometric properties and collect general population norms for a short, eight item version of the Body Shape Questionnaire (BSQ-8C). In the first study, a sample of undergraduate students (N = 182) was recruited to investigate the psychometric properties of the BSQ-8C, while in the second study a sample from the general population of Swedish women aged 18–30 years (N = 747) was used to collect norms. The test–retest reliability, internal consistency, and convergent validity for BSQ-8C were high. Given the questionnaire’s briefness and good psychometric properties, particularly when used in conjunction with norms, the BSQ8C is considered a valuable instrument for measuring body shape dissatisfaction among young adult women. © 2012 Elsevier Ltd. All rights reserved.
Introduction Over-evaluation of shape and weight constitutes the core psychopathology of eating disorders (Fairburn & Harrison, 2003) and consequently it has been found that increased body dissatisfaction is a risk factor for the development of eating disorders (e.g., Ghaderi, 2003; Ghaderi & Scott, 2001). Effective prevention plans or treatments for body dissatisfaction require sensitive and reliable assessment instruments that can measure the magnitude of change in body dissatisfaction over time. Information from such instruments can be further enhanced by collecting norms from the general population which provide contextual interpretation of the data. A significant number of questionnaires have been developed, each focusing on various aspects of body image or body shape (Brown, Cash, & Mikulka, 1990; Cash, Phillips, Santos, & Hrabosky, 2004). These aspects incorporate behavioral (e.g., Reas, Whisenhunt, Netemeyer, & Williamson, 2002) as well as perceptual, evaluative, affective, and cognitive components (Allen, Byrne, McLean, & Davis, 2008). One of the most commonly used questionnaires that taps into these components is the Body Shape Questionnaire (BSQ), developed by Cooper, Taylor, Cooper, and Fairburn (1987). The BSQ measures concerns with body shape and is designed to capture the phenomenological experience of “feeling
∗ Corresponding author at: Department of Psychology, P.O. Box 1225, 75142 Uppsala, Sweden. Tel.: +46 184717986. E-mail addresses: [email protected] (E. Welch), [email protected] (A. Ghaderi). 1740-1445/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bodyim.2012.04.009
fat.” It consists of 34 questions to which the reader responds on a 6-point Likert-like scale from “never” to “always.” Although the BSQ has proved to be a valid and reliable measure (Cooper et al., 1987; Rosen, Jones, Ramirez, & Waxman, 1996), it is a fairly long questionnaire and is less practical to use if administered repeatedly or together with several other questionnaires. In addition, given its unidimensional nature, Evans and Dolan (1993) suggested that it might be unnecessarily long for use in studies not focusing mainly on body disparagement. This unidimensionality of BSQ has been found in several other studies of participants with languages and cultural contexts other than English or Caucasian (e.g., Ghaderi & Scott, 2004; Mumford, Whitehouse, & Platts, 1991). Therefore, a brief version of the BSQ with the same excellent psychometric properties as of the original version would be very useful in both research and clinical practice. Several shorter versions of the BSQ have been proposed, but their psychometric properties are in most cases unknown as they are generally derived from the administered full version, and not administered as the brief version per se, although some exceptions exist (e.g., Dowson & Henderson, 2001). Pook, Tuschen-Caffier, and Brähler (2008) conducted a confirmatory factor analysis of different versions of the BSQ to identify the psychometrically best version. Eight forms of the BSQ where included in the analysis where three of the derivations showed a reasonable fit and only one version, the so called “BSQ-8C”, showed high sensitivity to change. The authors concluded that research on the BSQ derivations as stand-alone questionnaires is needed. Consequently, we conducted two studies to address this call. The aim of the first study was to investigate the psychometric proper-
548
E. Welch et al. / Body Image 9 (2012) 547–550
Table 1 Item means, standard deviations and corrected item-total correlation coefficients. Items included in the BSQ-8C
1. Have you been afraid that you might become fat (or fatter)? 2. Has feeling full (e.g., after eating a large meal) made you feel fat? 3. Has thinking about your shape interfered with your ability to concentrate (e.g., while watching television, reading, listening to conversations)? 4. Have you imagined cutting off fleshy areas of your body? 5. Have you felt excessively large and rounded? 6. Have you thought that you are in the shape you are because you lack self-control 7. Has seeing your reflection (e.g., in a mirror or shop window) made you feel bad about your shape 8. Have you been particularly self-conscious about your shape when in the company of other people
Time 1
Time 2
Mean(SD) Total sample
Mean(SD) Men
Mean (SD) Women
Corrected item-total correlation coefficient
Mean(SD) Total sample
Mean(SD) Men
Mean(SD) Women
Corrected item-total correlation coefficient
2.92(1.42)
2.07(1.02)
3.30(1.41)
.85
2.81(1.33)
2.02(0.91)
3.17(1.33)
.86
2.63(1.26)
2.05(1.06)
2.89(1.26)
.80
2.63(1.22)
1.95(0.91)
2.94(1.22)
.75
1.82(1.08)
1.35(0.78)
2.03(1.14)
.79
1.85(1.08)
1.38(0.73)
2.06(1.15)
.80
1.26(0.73)
1.05(0.30)
1.36(0.84)
.52
1.25(0.80)
1.09(0.55)
1.32(0.88)
.56
2.13(1.29)
1.64(1.06)
2.36(1.33)
.78
2.13(1.29)
1.55(0.79)
2.40(1.38)
.85
2.51(1.37)
2.31(1.09)
2.60(1.48)
.73
2.59(1.36)
2.16(1.03)
2.78(1.45)
.77
2.34(1.21)
1.69(0.86)
2.63(1.24)
.82
2.34(1.26)
1.67(0.88)
2.64(1.29)
.86
2.87(1.16)
2.27(1.11)
3.14(1.09)
.73
2.74(1.18)
2.13(0.98)
3.02(1.16)
.81
Note. The respondent is asked “Over the past four weeks . . .” The questions are answered on a 6-point scale, from “never” to “always.”
ties of the BSQ-8C as a stand-alone questionnaire (i.e., test–retest reliability, internal consistency, and validity). The aim of the second study was to obtain norms from the general population in order to provide a more informative and contextually sensitive interpretation of scores on the BSQ-8C. Data from norms is important as it enables analysis of whether observed changes over time might be clinically significant. Given the higher prevalence of body dissatisfaction and significantly higher occurrence of eating disorders and disordered eating among women compared to men, a sample from the general Swedish population of young adult females (18–30 years) was used for the second study.
Method Participants Study 1. A sample of undergraduate students at a large Swedish university was recruited to investigate the psychometric properties of the BSQ-8C as a stand-alone questionnaire. A total of 182 undergraduate students (69.2% females and 30.8% males) from a number of different academic disciplines participated in the test–retest part of the study. Study 2. A random sample of 1501 females (18–30 years old) from the general Swedish population was used in the study. A total of 747 participants (51%) with a mean age of 23.9 years (SD = 3.9 years) and a mean body mass index (BMI) of 22.9 (SD = 3.9) completed the questionnaire package. The BMI values were calculated based on self-reported weight and height. The total number of participants excludes incorrectly listed addresses (n = 27), individuals being abroad at the time of the survey (n = 11), as well as those unable to complete the questionnaire due to mental disability (n = 1).
Procedure Study 1. Undergraduate students were recruited among university students at a large university in Sweden during the spring semester (more specifically, March–April). An e-mail explaining the purpose of the study was sent to course leaders at the university lecturing in five different disciplines (Psychology, Gender Research, Information Technology, Law, and Pharmacy). After receiving approval from the course leaders, two dates (approximately 2 weeks apart) were chosen for the first and second data collection (Time 1 and Time 2, respectively). The students were approached at the start of lectures and were informed about the purpose of the study and its voluntary and anonymous nature. Those that chose to participate in the study were given the BSQ-8C and a set of six questions to aid in matching the questionnaires from the first and second assessment and to retain anonymity. Specifically, the students were asked to list their favorite (1) color, (2) book, (3) movie, (4) food, and (5) leisure time activity, and (6) state their gender. The second assessment was done approximately two weeks after the first assessment (M = 15.1 days, SD = 4.3 days, Range: 11–21 days). Ethical approval for the study was obtained by the regional ethics committee. Study 2. For recruiting a representative sample of females aged 18–30 years, a random selection procedure was conducted by the Swedish Population Address Register (SPAR). A sample including only females in this particular age range was chosen due to young women’s over-representation in eating pathology and body image concerns. A questionnaire package including demographic questions, the BSQ-8C, the Eating Disorders Examination Questionnaire and the Clinical Impairment Assessment Questionnaire was sent out to all potential participants. Data from these questionnaires are also used in Welch, Birgegård, Parling, and Ghaderi (2011). Three reminder letters were sent out, and incentives to complete the questionnaire (gift certificate lotteries, movie tickets, etc.) were
E. Welch et al. / Body Image 9 (2012) 547–550
549
Table 2 Mean, median, mode and the 5th, 10th, 25th, 50th, 75th, 90th and 95th percentiles of total BSQ-8C scores for women between the ages of 18 and 30 from the Swedish general population. Age
N
Mean
SD
Median
Mode
Percentiles (‰) 5th
18 19 20 21 22 23 24 25 26 27 28 29 30 All ages
67 61 56 49 59 60 46 55 63 53 60 61 51 747
22 22 20 22 19 20 20 20 21 18 18 19 20 20
11 12 11 10 10 10 10 9 11 9 9 10 10 10
21 18 19 21 17 19 16 18 18 15 16 16 17 18
12 13 11 26 8 8 8 18 18 8 9 8 M 8
8 9 8 9 8 8 8 10 9 8 8 8 8 8
10th 9 10 9 9 8 8 8 10 10 8 9 8 9 9
15th
50th
75th
90th
95th
12 13 11 15 11 12 12 14 12 11 10 11 11 12
21 18 19 21 17 19 16 18 18 15 16 16 17 18
30 34 27 27 25 26 27 24 27 26 24 27 24 26
38 41 36 40 35 35 37 35 40 34 30 33 37 36
44 43 40 42 41 42 41 38 43 38 34 36 41 40
Note. M, multiple modes.
Results
included in the two last reminder letters to maximize the response rate. The study was approved by the regional ethics committee. Study 1 Instruments BSQ-8C – brief form of the Body Shape Questionnaire. The full version BSQ is a self-report questionnaire, originally comprising 34 questions measuring the extent of psychopathology of concerns about body shape. The questions refer to the subject’s state over the past four weeks and are answered on a 6-point scale, from “never” to “always.” Higher values on the BSQ indicate more body dissatisfaction. The BSQ has been shown to have good concurrent and discriminative validity (Cooper et al., 1987). The BSQ-8C is the short version of the BSQ and consists of eight items extracted from the full version according to Evans and Dolan (1993). The BSQ-8C, and not the other short versions of the BSQ, was used in the present study due to its high sensitivity to change in the course of therapy (Pook et al., 2008). The items of the BSQ-8C are shown in Table 1. Since BSQ-8C was given to Swedish participants in both Study 1 and Study 2, a translation was required. The full version BSQ has previously been translated into Swedish and has shown excellent psychometric properties (Ghaderi & Scott, 2004). Specifically, the BSQ was translated into Swedish and then back-translated to English by independent translators who had Swedish or English as their native language. After the back-translation to English, the questionnaire was found to be nearly identical to the original version, with only the word order being changed in three of the items. Eating Disorders Examination Questionnaire (EDE-Q). The EDE-Q is a 28-item self-report measure adapted from the Eating Disorders Examination (Fairburn & Cooper, 1993), which is a semistructured interview for assessing eating disorders. It focuses on the past 28 days and provides a global score as well as four subscales: Eating Concern, Shape Concern, Weight Concern, and Restraint. It also provides data on the frequency of core eating disorder behaviors, including binge eating and compensatory behaviors. In Study 2 the Shape Concern subscale of the EDE-Q, version 6.0 (Fairburn & Beglin, 2008), was used to investigate the convergent validity of the BSQ-8C.
The BSQ-8C showed high test–retest reliability (r = .95) in the undergraduate sample. Significant differences between the two assessments were also investigated and no significant differences emerged (t(181) = .79, p = .43). The internal consistency of the BSQ8C measured with Cronbach’s alpha was .92 at the Time 1 and .93 at Time 2. Item means, standard deviations, and corrected item-total correlation coefficients at Time 1 and Time 2 of the test–retest are presented in Table 1. Study 2 Using data from the general population, we investigated the convergent validity of the BSQ-8C by correlating it to the Shape Concern subscale of the EDE-Q and to body mass index (BMI). The emerging correlation to the Shape Concern subscale was high (r = .90, p < .001) and the correlation to BMI was fair (r = .39, p < .001). Cronbach’s alphas for the BSQ-8C and the Shape Concern subscale of the EDE-Q were .94 and .93 respectively. The mean of the BSQ-8C for the population-based sample was 20 (SD = 10) while the median was 18. The highest possible score in the BSQ-8C is 48. An age breakdown of the means, standard deviations, medians, modes, and percentile ranks for the BSQ-8C are shown in Table 2. Finally, to investigate whether the items in the Swedish version of the BSQ-8C load on a single factor, as is the case in the English version, we conducted an exploratory factor analysis, which returned one factor that explained 59.5% of the variance. A confirmatory factor analysis using linear structural equation modeling in LISREL (Jöreskog & Sörbom, 1993) with all items loading on a single factor resulted in a poor fit (2 = 138.14, p < .00001, RMSEA = .089). Using modification indices in LISREL to investigate the lack of fit, more acceptable fit indices were achieved (e.g., 2 = 92.3, p < .0001, RMSEA = .072) by letting the error covariance between the first and second items on the BSQ be freely estimated. Discussion
Statistical Analyses Test–retest was analyzed using both Pearson’s r and Student’s t-test. The internal consistency of the BSQ-8C and the Shape Concern subscale of the EDE-Q were analyzed using Cronbach’s alpha. Norms for the general population sample in Study 2 are presented using descriptive statistics.
The results indicate that the BSQ-8C is a one-dimensional instrument with high internal consistency, excellent test–retest reliability and high convergent validity. In previous studies utilizing the BSQ-8C by Evans and Dolan (1993), as well as by Pook et al. (2008), the conclusions had to be viewed with caution as the instrument was an unverified extraction from the full version of the BSQ.
550
E. Welch et al. / Body Image 9 (2012) 547–550
The present study contributes to the body of knowledge on the measures of body dissatisfaction by showing that the BSQ-8C possesses excellent psychometric properties when administered as a stand-alone instrument. Given its low number of items and good psychometric properties, the BSQ-8C is a valuable instrument for measuring body shape dissatisfaction. Another potential advantage of the BSQ-8C is its gender non-specificity. The wording of some of its items in the original version of the BSQ makes it less appropriate for use with males, whereas the BSQ-8C does not contain any items that might be exclusively relevant for females. Thus future studies might be able to investigate the level of body shape dissatisfaction in male populations using the BSQ-8C. Availability of norms for young adult females makes it significantly easier for clinicians and researchers to investigate the occurrence and change in body dissatisfaction among their patients or samples by interpreting the scores in a context. In addition, given its brief format and focus on attitudinal (i.e., more cognitive and emotional) aspects of body shape dissatisfaction, the BSQ-8C provides an opportunity to investigate the interaction between these aspects and its behavioral representations (e.g., body checking) in the emergence and maintenance of body shape dissatisfaction. There are two major limitations with the present study. The first is the high drop-out rate in Study 2, although the total response rate was acceptable. The nearly identical mean ages between the sample and the total population of women in the same age group support the assumption that the norms can be viewed as representative for young adult women in Sweden. Lack of data on male participants is the second shortcoming of the present study; however the present availability of a short and psychometrically robust instrument provides a good opportunity for future research to obtain more data on males as well. Further research on the BSQ8C in other age groups, cultures, and languages would enhance its generalizability. Acknowledgment The Swedish Research Council (42120042614) is acknowledged for financial support for the data collection.
References Allen, K. L., Byrne, S. M., McLean, N. J., & Davis, E. A. (2008). Overconcern with weight and shape is not the same as body dissatisfaction: Evidence from a prospective study of pre-adolescent boys and girls. Body Image, 5, 261–270. Brown, T. A., Cash, T. F., & Mikulka, P. J. (1990). Attitudinal body-image assessment: Factor analysis of the Body-Self Relations Questionnaire. Journal of Personality Assessment, 55, 135–144. Cash, T. F., Phillips, K. A., Santos, M. T., & Hrabosky, J. I. (2004). Measuring negative body image: Validation of the Body Image Disturbance Questionnaire in a nonclinical population. Body Image, 1, 363–372. Cooper, P., Taylor, M., Cooper, Z., & Fairburn, C. (1987). The development and validation of the Body Satisfaction Questionnaire. International Journal of Eating Disorders, 6, 485–494. Dowson, J., & Henderson, L. (2001). The validity of a short version of the Body Shape Questionnaire. Psychiatry Research, 102, 263–271. Evans, C., & Dolan, B. (1993). Body Shape Questionnaire: Derivation of shortened alternate forms. International Journal of Eating Disorders, 13, 315–321. Fairburn, C. G., & Beglin, S. J. (2008). Eating Disorder Examination Questionnaire (EDE-Q 6.0). In C. G. Fairburn (Ed.), Eating Disorder Examination Questionnaire (EDE-Q 6.0) (pp. 309–313). New York: Guilford Press. Fairburn, C. G., & Cooper, Z. (1993). The Eating Disorder Examination. In C. G. Fairburn & G. T. Wilson (Eds.), The Eating Disorder Examination (pp. 317–360). New York: Guilford Press. Fairburn, C. G., & Harrison, P. J. (2003). Eating disorders. The Lancet, 361, 407–416. Ghaderi, A. (2003). Structural modeling analysis of prospective risk factors for eating disorder. Eating Behaviors, 3, 387–396. Ghaderi, A., & Scott, B. (2001). Prevalence, incidence and prospective risk factors for eating disorders. Acta Psychiatrica Scandinavica, 104, 122–130. Ghaderi, A., & Scott, B. (2004). The reliability and validity of the Swedish version of the Body Shape Questionnaire. Scandinavian Journal of Psychology, 45, 319–324. Jöreskog, K., & Sörbom, D. (1993). LISREL 8: Structural Equation Modeling with the SIMPLIS Command Language. Chicago: Scientific Software International. Mumford, D., Whitehouse, A., & Platts, M. (1991). Eating disorders among Asian schoolgirls in Bradford: Sociocultural correlates. British Journal of Psychiatry, 158, 222–228. Pook, M., Tuschen-Caffier, B., & Brähler, E. (2008). Evaluation and comparison of different versions of the Body Shape Questionnaire. Psychiatry Research, 158, 67–73. Reas, D. L., Whisenhunt, B. L., Netemeyer, R., & Williamson, D. A. (2002). Development of the Body Checking Questionnaire: A self-report measure of body checking behaviors. The International Journal of Eating Disorders, 31, 324–333. Rosen, J. C., Jones, A., Ramirez, E., & Waxman, S. (1996). Body Shape Questionnaire: Studies of validity and reliability. The International Journal of Eating Disorders, 20, 315–319. Welch, E., Birgegård, A., Parling, T., & Ghaderi, A. (2011). Eating Disorder Examination Questionnaire and Clinical Impairment Assessment Questionnaire: General population and clinical norms for young adult women in Sweden. Behaviour Research and Therapy, 49, 85–91.
Contents lists available at SciVerse ScienceDirect
Body Image journal homepage: www.elsevier.com/locate/bodyimage
Brief research report
Body Shape Questionnaire: Psychometric properties of the short version (BSQ-8C) and norms from the general Swedish population Elisabeth Welch, Marcus Lagerström, Ata Ghaderi ∗ Uppsala University, Department of Psychology, Uppsala, Sweden
a r t i c l e
i n f o
a b s t r a c t
Article history: Received 23 November 2011 Received in revised form 21 April 2012 Accepted 24 April 2012 Keywords: Body dissatisfaction Body Shape Questionnaire BSQ-8C
There is a need for brief, sensitive, and reliable instruments for contextual assessment of body dissatisfaction. Suitable selection and optimal use of psychological assessment instruments requires appropriate psychometric as well as normative data. Two studies were conducted to investigate the psychometric properties and collect general population norms for a short, eight item version of the Body Shape Questionnaire (BSQ-8C). In the first study, a sample of undergraduate students (N = 182) was recruited to investigate the psychometric properties of the BSQ-8C, while in the second study a sample from the general population of Swedish women aged 18–30 years (N = 747) was used to collect norms. The test–retest reliability, internal consistency, and convergent validity for BSQ-8C were high. Given the questionnaire’s briefness and good psychometric properties, particularly when used in conjunction with norms, the BSQ8C is considered a valuable instrument for measuring body shape dissatisfaction among young adult women. © 2012 Elsevier Ltd. All rights reserved.
Introduction Over-evaluation of shape and weight constitutes the core psychopathology of eating disorders (Fairburn & Harrison, 2003) and consequently it has been found that increased body dissatisfaction is a risk factor for the development of eating disorders (e.g., Ghaderi, 2003; Ghaderi & Scott, 2001). Effective prevention plans or treatments for body dissatisfaction require sensitive and reliable assessment instruments that can measure the magnitude of change in body dissatisfaction over time. Information from such instruments can be further enhanced by collecting norms from the general population which provide contextual interpretation of the data. A significant number of questionnaires have been developed, each focusing on various aspects of body image or body shape (Brown, Cash, & Mikulka, 1990; Cash, Phillips, Santos, & Hrabosky, 2004). These aspects incorporate behavioral (e.g., Reas, Whisenhunt, Netemeyer, & Williamson, 2002) as well as perceptual, evaluative, affective, and cognitive components (Allen, Byrne, McLean, & Davis, 2008). One of the most commonly used questionnaires that taps into these components is the Body Shape Questionnaire (BSQ), developed by Cooper, Taylor, Cooper, and Fairburn (1987). The BSQ measures concerns with body shape and is designed to capture the phenomenological experience of “feeling
∗ Corresponding author at: Department of Psychology, P.O. Box 1225, 75142 Uppsala, Sweden. Tel.: +46 184717986. E-mail addresses: [email protected] (E. Welch), [email protected] (A. Ghaderi). 1740-1445/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bodyim.2012.04.009
fat.” It consists of 34 questions to which the reader responds on a 6-point Likert-like scale from “never” to “always.” Although the BSQ has proved to be a valid and reliable measure (Cooper et al., 1987; Rosen, Jones, Ramirez, & Waxman, 1996), it is a fairly long questionnaire and is less practical to use if administered repeatedly or together with several other questionnaires. In addition, given its unidimensional nature, Evans and Dolan (1993) suggested that it might be unnecessarily long for use in studies not focusing mainly on body disparagement. This unidimensionality of BSQ has been found in several other studies of participants with languages and cultural contexts other than English or Caucasian (e.g., Ghaderi & Scott, 2004; Mumford, Whitehouse, & Platts, 1991). Therefore, a brief version of the BSQ with the same excellent psychometric properties as of the original version would be very useful in both research and clinical practice. Several shorter versions of the BSQ have been proposed, but their psychometric properties are in most cases unknown as they are generally derived from the administered full version, and not administered as the brief version per se, although some exceptions exist (e.g., Dowson & Henderson, 2001). Pook, Tuschen-Caffier, and Brähler (2008) conducted a confirmatory factor analysis of different versions of the BSQ to identify the psychometrically best version. Eight forms of the BSQ where included in the analysis where three of the derivations showed a reasonable fit and only one version, the so called “BSQ-8C”, showed high sensitivity to change. The authors concluded that research on the BSQ derivations as stand-alone questionnaires is needed. Consequently, we conducted two studies to address this call. The aim of the first study was to investigate the psychometric proper-
548
E. Welch et al. / Body Image 9 (2012) 547–550
Table 1 Item means, standard deviations and corrected item-total correlation coefficients. Items included in the BSQ-8C
1. Have you been afraid that you might become fat (or fatter)? 2. Has feeling full (e.g., after eating a large meal) made you feel fat? 3. Has thinking about your shape interfered with your ability to concentrate (e.g., while watching television, reading, listening to conversations)? 4. Have you imagined cutting off fleshy areas of your body? 5. Have you felt excessively large and rounded? 6. Have you thought that you are in the shape you are because you lack self-control 7. Has seeing your reflection (e.g., in a mirror or shop window) made you feel bad about your shape 8. Have you been particularly self-conscious about your shape when in the company of other people
Time 1
Time 2
Mean(SD) Total sample
Mean(SD) Men
Mean (SD) Women
Corrected item-total correlation coefficient
Mean(SD) Total sample
Mean(SD) Men
Mean(SD) Women
Corrected item-total correlation coefficient
2.92(1.42)
2.07(1.02)
3.30(1.41)
.85
2.81(1.33)
2.02(0.91)
3.17(1.33)
.86
2.63(1.26)
2.05(1.06)
2.89(1.26)
.80
2.63(1.22)
1.95(0.91)
2.94(1.22)
.75
1.82(1.08)
1.35(0.78)
2.03(1.14)
.79
1.85(1.08)
1.38(0.73)
2.06(1.15)
.80
1.26(0.73)
1.05(0.30)
1.36(0.84)
.52
1.25(0.80)
1.09(0.55)
1.32(0.88)
.56
2.13(1.29)
1.64(1.06)
2.36(1.33)
.78
2.13(1.29)
1.55(0.79)
2.40(1.38)
.85
2.51(1.37)
2.31(1.09)
2.60(1.48)
.73
2.59(1.36)
2.16(1.03)
2.78(1.45)
.77
2.34(1.21)
1.69(0.86)
2.63(1.24)
.82
2.34(1.26)
1.67(0.88)
2.64(1.29)
.86
2.87(1.16)
2.27(1.11)
3.14(1.09)
.73
2.74(1.18)
2.13(0.98)
3.02(1.16)
.81
Note. The respondent is asked “Over the past four weeks . . .” The questions are answered on a 6-point scale, from “never” to “always.”
ties of the BSQ-8C as a stand-alone questionnaire (i.e., test–retest reliability, internal consistency, and validity). The aim of the second study was to obtain norms from the general population in order to provide a more informative and contextually sensitive interpretation of scores on the BSQ-8C. Data from norms is important as it enables analysis of whether observed changes over time might be clinically significant. Given the higher prevalence of body dissatisfaction and significantly higher occurrence of eating disorders and disordered eating among women compared to men, a sample from the general Swedish population of young adult females (18–30 years) was used for the second study.
Method Participants Study 1. A sample of undergraduate students at a large Swedish university was recruited to investigate the psychometric properties of the BSQ-8C as a stand-alone questionnaire. A total of 182 undergraduate students (69.2% females and 30.8% males) from a number of different academic disciplines participated in the test–retest part of the study. Study 2. A random sample of 1501 females (18–30 years old) from the general Swedish population was used in the study. A total of 747 participants (51%) with a mean age of 23.9 years (SD = 3.9 years) and a mean body mass index (BMI) of 22.9 (SD = 3.9) completed the questionnaire package. The BMI values were calculated based on self-reported weight and height. The total number of participants excludes incorrectly listed addresses (n = 27), individuals being abroad at the time of the survey (n = 11), as well as those unable to complete the questionnaire due to mental disability (n = 1).
Procedure Study 1. Undergraduate students were recruited among university students at a large university in Sweden during the spring semester (more specifically, March–April). An e-mail explaining the purpose of the study was sent to course leaders at the university lecturing in five different disciplines (Psychology, Gender Research, Information Technology, Law, and Pharmacy). After receiving approval from the course leaders, two dates (approximately 2 weeks apart) were chosen for the first and second data collection (Time 1 and Time 2, respectively). The students were approached at the start of lectures and were informed about the purpose of the study and its voluntary and anonymous nature. Those that chose to participate in the study were given the BSQ-8C and a set of six questions to aid in matching the questionnaires from the first and second assessment and to retain anonymity. Specifically, the students were asked to list their favorite (1) color, (2) book, (3) movie, (4) food, and (5) leisure time activity, and (6) state their gender. The second assessment was done approximately two weeks after the first assessment (M = 15.1 days, SD = 4.3 days, Range: 11–21 days). Ethical approval for the study was obtained by the regional ethics committee. Study 2. For recruiting a representative sample of females aged 18–30 years, a random selection procedure was conducted by the Swedish Population Address Register (SPAR). A sample including only females in this particular age range was chosen due to young women’s over-representation in eating pathology and body image concerns. A questionnaire package including demographic questions, the BSQ-8C, the Eating Disorders Examination Questionnaire and the Clinical Impairment Assessment Questionnaire was sent out to all potential participants. Data from these questionnaires are also used in Welch, Birgegård, Parling, and Ghaderi (2011). Three reminder letters were sent out, and incentives to complete the questionnaire (gift certificate lotteries, movie tickets, etc.) were
E. Welch et al. / Body Image 9 (2012) 547–550
549
Table 2 Mean, median, mode and the 5th, 10th, 25th, 50th, 75th, 90th and 95th percentiles of total BSQ-8C scores for women between the ages of 18 and 30 from the Swedish general population. Age
N
Mean
SD
Median
Mode
Percentiles (‰) 5th
18 19 20 21 22 23 24 25 26 27 28 29 30 All ages
67 61 56 49 59 60 46 55 63 53 60 61 51 747
22 22 20 22 19 20 20 20 21 18 18 19 20 20
11 12 11 10 10 10 10 9 11 9 9 10 10 10
21 18 19 21 17 19 16 18 18 15 16 16 17 18
12 13 11 26 8 8 8 18 18 8 9 8 M 8
8 9 8 9 8 8 8 10 9 8 8 8 8 8
10th 9 10 9 9 8 8 8 10 10 8 9 8 9 9
15th
50th
75th
90th
95th
12 13 11 15 11 12 12 14 12 11 10 11 11 12
21 18 19 21 17 19 16 18 18 15 16 16 17 18
30 34 27 27 25 26 27 24 27 26 24 27 24 26
38 41 36 40 35 35 37 35 40 34 30 33 37 36
44 43 40 42 41 42 41 38 43 38 34 36 41 40
Note. M, multiple modes.
Results
included in the two last reminder letters to maximize the response rate. The study was approved by the regional ethics committee. Study 1 Instruments BSQ-8C – brief form of the Body Shape Questionnaire. The full version BSQ is a self-report questionnaire, originally comprising 34 questions measuring the extent of psychopathology of concerns about body shape. The questions refer to the subject’s state over the past four weeks and are answered on a 6-point scale, from “never” to “always.” Higher values on the BSQ indicate more body dissatisfaction. The BSQ has been shown to have good concurrent and discriminative validity (Cooper et al., 1987). The BSQ-8C is the short version of the BSQ and consists of eight items extracted from the full version according to Evans and Dolan (1993). The BSQ-8C, and not the other short versions of the BSQ, was used in the present study due to its high sensitivity to change in the course of therapy (Pook et al., 2008). The items of the BSQ-8C are shown in Table 1. Since BSQ-8C was given to Swedish participants in both Study 1 and Study 2, a translation was required. The full version BSQ has previously been translated into Swedish and has shown excellent psychometric properties (Ghaderi & Scott, 2004). Specifically, the BSQ was translated into Swedish and then back-translated to English by independent translators who had Swedish or English as their native language. After the back-translation to English, the questionnaire was found to be nearly identical to the original version, with only the word order being changed in three of the items. Eating Disorders Examination Questionnaire (EDE-Q). The EDE-Q is a 28-item self-report measure adapted from the Eating Disorders Examination (Fairburn & Cooper, 1993), which is a semistructured interview for assessing eating disorders. It focuses on the past 28 days and provides a global score as well as four subscales: Eating Concern, Shape Concern, Weight Concern, and Restraint. It also provides data on the frequency of core eating disorder behaviors, including binge eating and compensatory behaviors. In Study 2 the Shape Concern subscale of the EDE-Q, version 6.0 (Fairburn & Beglin, 2008), was used to investigate the convergent validity of the BSQ-8C.
The BSQ-8C showed high test–retest reliability (r = .95) in the undergraduate sample. Significant differences between the two assessments were also investigated and no significant differences emerged (t(181) = .79, p = .43). The internal consistency of the BSQ8C measured with Cronbach’s alpha was .92 at the Time 1 and .93 at Time 2. Item means, standard deviations, and corrected item-total correlation coefficients at Time 1 and Time 2 of the test–retest are presented in Table 1. Study 2 Using data from the general population, we investigated the convergent validity of the BSQ-8C by correlating it to the Shape Concern subscale of the EDE-Q and to body mass index (BMI). The emerging correlation to the Shape Concern subscale was high (r = .90, p < .001) and the correlation to BMI was fair (r = .39, p < .001). Cronbach’s alphas for the BSQ-8C and the Shape Concern subscale of the EDE-Q were .94 and .93 respectively. The mean of the BSQ-8C for the population-based sample was 20 (SD = 10) while the median was 18. The highest possible score in the BSQ-8C is 48. An age breakdown of the means, standard deviations, medians, modes, and percentile ranks for the BSQ-8C are shown in Table 2. Finally, to investigate whether the items in the Swedish version of the BSQ-8C load on a single factor, as is the case in the English version, we conducted an exploratory factor analysis, which returned one factor that explained 59.5% of the variance. A confirmatory factor analysis using linear structural equation modeling in LISREL (Jöreskog & Sörbom, 1993) with all items loading on a single factor resulted in a poor fit (2 = 138.14, p < .00001, RMSEA = .089). Using modification indices in LISREL to investigate the lack of fit, more acceptable fit indices were achieved (e.g., 2 = 92.3, p < .0001, RMSEA = .072) by letting the error covariance between the first and second items on the BSQ be freely estimated. Discussion
Statistical Analyses Test–retest was analyzed using both Pearson’s r and Student’s t-test. The internal consistency of the BSQ-8C and the Shape Concern subscale of the EDE-Q were analyzed using Cronbach’s alpha. Norms for the general population sample in Study 2 are presented using descriptive statistics.
The results indicate that the BSQ-8C is a one-dimensional instrument with high internal consistency, excellent test–retest reliability and high convergent validity. In previous studies utilizing the BSQ-8C by Evans and Dolan (1993), as well as by Pook et al. (2008), the conclusions had to be viewed with caution as the instrument was an unverified extraction from the full version of the BSQ.
550
E. Welch et al. / Body Image 9 (2012) 547–550
The present study contributes to the body of knowledge on the measures of body dissatisfaction by showing that the BSQ-8C possesses excellent psychometric properties when administered as a stand-alone instrument. Given its low number of items and good psychometric properties, the BSQ-8C is a valuable instrument for measuring body shape dissatisfaction. Another potential advantage of the BSQ-8C is its gender non-specificity. The wording of some of its items in the original version of the BSQ makes it less appropriate for use with males, whereas the BSQ-8C does not contain any items that might be exclusively relevant for females. Thus future studies might be able to investigate the level of body shape dissatisfaction in male populations using the BSQ-8C. Availability of norms for young adult females makes it significantly easier for clinicians and researchers to investigate the occurrence and change in body dissatisfaction among their patients or samples by interpreting the scores in a context. In addition, given its brief format and focus on attitudinal (i.e., more cognitive and emotional) aspects of body shape dissatisfaction, the BSQ-8C provides an opportunity to investigate the interaction between these aspects and its behavioral representations (e.g., body checking) in the emergence and maintenance of body shape dissatisfaction. There are two major limitations with the present study. The first is the high drop-out rate in Study 2, although the total response rate was acceptable. The nearly identical mean ages between the sample and the total population of women in the same age group support the assumption that the norms can be viewed as representative for young adult women in Sweden. Lack of data on male participants is the second shortcoming of the present study; however the present availability of a short and psychometrically robust instrument provides a good opportunity for future research to obtain more data on males as well. Further research on the BSQ8C in other age groups, cultures, and languages would enhance its generalizability. Acknowledgment The Swedish Research Council (42120042614) is acknowledged for financial support for the data collection.
References Allen, K. L., Byrne, S. M., McLean, N. J., & Davis, E. A. (2008). Overconcern with weight and shape is not the same as body dissatisfaction: Evidence from a prospective study of pre-adolescent boys and girls. Body Image, 5, 261–270. Brown, T. A., Cash, T. F., & Mikulka, P. J. (1990). Attitudinal body-image assessment: Factor analysis of the Body-Self Relations Questionnaire. Journal of Personality Assessment, 55, 135–144. Cash, T. F., Phillips, K. A., Santos, M. T., & Hrabosky, J. I. (2004). Measuring negative body image: Validation of the Body Image Disturbance Questionnaire in a nonclinical population. Body Image, 1, 363–372. Cooper, P., Taylor, M., Cooper, Z., & Fairburn, C. (1987). The development and validation of the Body Satisfaction Questionnaire. International Journal of Eating Disorders, 6, 485–494. Dowson, J., & Henderson, L. (2001). The validity of a short version of the Body Shape Questionnaire. Psychiatry Research, 102, 263–271. Evans, C., & Dolan, B. (1993). Body Shape Questionnaire: Derivation of shortened alternate forms. International Journal of Eating Disorders, 13, 315–321. Fairburn, C. G., & Beglin, S. J. (2008). Eating Disorder Examination Questionnaire (EDE-Q 6.0). In C. G. Fairburn (Ed.), Eating Disorder Examination Questionnaire (EDE-Q 6.0) (pp. 309–313). New York: Guilford Press. Fairburn, C. G., & Cooper, Z. (1993). The Eating Disorder Examination. In C. G. Fairburn & G. T. Wilson (Eds.), The Eating Disorder Examination (pp. 317–360). New York: Guilford Press. Fairburn, C. G., & Harrison, P. J. (2003). Eating disorders. The Lancet, 361, 407–416. Ghaderi, A. (2003). Structural modeling analysis of prospective risk factors for eating disorder. Eating Behaviors, 3, 387–396. Ghaderi, A., & Scott, B. (2001). Prevalence, incidence and prospective risk factors for eating disorders. Acta Psychiatrica Scandinavica, 104, 122–130. Ghaderi, A., & Scott, B. (2004). The reliability and validity of the Swedish version of the Body Shape Questionnaire. Scandinavian Journal of Psychology, 45, 319–324. Jöreskog, K., & Sörbom, D. (1993). LISREL 8: Structural Equation Modeling with the SIMPLIS Command Language. Chicago: Scientific Software International. Mumford, D., Whitehouse, A., & Platts, M. (1991). Eating disorders among Asian schoolgirls in Bradford: Sociocultural correlates. British Journal of Psychiatry, 158, 222–228. Pook, M., Tuschen-Caffier, B., & Brähler, E. (2008). Evaluation and comparison of different versions of the Body Shape Questionnaire. Psychiatry Research, 158, 67–73. Reas, D. L., Whisenhunt, B. L., Netemeyer, R., & Williamson, D. A. (2002). Development of the Body Checking Questionnaire: A self-report measure of body checking behaviors. The International Journal of Eating Disorders, 31, 324–333. Rosen, J. C., Jones, A., Ramirez, E., & Waxman, S. (1996). Body Shape Questionnaire: Studies of validity and reliability. The International Journal of Eating Disorders, 20, 315–319. Welch, E., Birgegård, A., Parling, T., & Ghaderi, A. (2011). Eating Disorder Examination Questionnaire and Clinical Impairment Assessment Questionnaire: General population and clinical norms for young adult women in Sweden. Behaviour Research and Therapy, 49, 85–91.