Psychological Reports, 1991, 68, 491-499.

@ Psychological Reports 1991

STABLE VS UNSTABLE W E I G H T HISTORY, BODY-IMAGE, AND W E I G H T CONCERN I N WOMEN OF AVERAGE BODY W E I G H T ' CHERYL D. THOMAS University of Windsor Summary.-The focus of this study was the effect of unstable weight history on current body-image among women. Participants were 45 college women of average weight (median age = 23 yr.); 15 had stable weight histories, 15 had past lowest weights 15% or more below current weight, and 15 had past highest weights 15% or more above current weight. The women were assessed on perceptual and attitudinal measures of body-image and a self-report measure of preoccupation with weight. I n comparison to women with stable weight histories, those with unstable weight histories indicated greater dissatisfaction with the shape and size of their bodies, expressed more negative body attitudes, and reported increased concern about weight. Women with high weight histories overestimated their frontal body-size and expressed greater hssatisfaction with frontal size than women with low weight histories, but the two groups did not differ significantly on any of the other dependent measures. Possible explanations for increased body-image disturbance and weight concern among women with unstable weight histories are presented.

Current body weight has been related to body-size judgements and body attitudes among women with and without clinical eating disorders (e.g., Birtchnell, Lacey, & Harte, 1985; Cash & Green, 1986; Hesse-Biber, ClaytonMatthews, & Downey, 1987; Steiger, Fraenkel, & Leichner, 1989). However, weight history variables have received comparatively Little attention despite the fact that some data suggest a link among instabhty of weight, eating disturbance, and body-image. Crisp and Kalucy (1974) reported that anorexic patients overestimated their body-size and that the extent of overestimation was related to premorbid weight. Moreover, among the normal controls in their study, those who showed the greatest overestimation had experienced recent weight losses. Consistent with Stunkard's earlier work on the effects of juvenile obesity (see Stunkard, 1976), Crisp and Kalucy concluded that overestimation might reflect a failure to adapt body-size perceptions to conform to changes in body weight and shape. According to this explanation, size overestimation occurs in individuals who do not adjust their perceptions to accommodate decreases in body-size or shape; size underestimation reflects a failure to adjust perceptions to conform to real increases in body-size or shape.

'Preparation of this paper was supported by a University of Windsor Research Board Grant. Please direct correspondence to Cheryl D. Thomas, Ph.D., Department of Psychology, Univenicy of Windsor, 401 Sunset Avenue, Windsor, Ontario N9B 3P4, Canada.

492

C. D. THOMAS

Some recent results obtained from samples of bulimic women (e.g., Count & Adams, 1985; Freeman, Thomas, Solyom, & Koopman, 1985), normal dieters, and "restrained" normals (e.g., Count & Adams, 1985) suggest a relationshp between weight history and size overestimation that is consistent with "adaptive failure." However, other findings (e.g., Gorham & Hundleby, 1988; Slade, 1977) d o not fit the explanation proposed by Crisp and Kalucy. Slade (1977) found an inverse relationship between size overestimation and weight increase over the course of normal pregnancy. H e suggested that overestimation was due to "abnormal body-size sensitivity," a condition that might be associated with any significant change in body-size or shape (i.e., weight increase or decrease). The "abnormal sensitivity" and "adaptive failure" hypotheses have not been adequately tested to date; the few studies that d o report pertinent data d o not provide consistent support for either hypothesis. As with other questions related to body-image (see Cash & Brown, 1987), use of heterogeneous samples and measures of uncertain validity has impeded understanding of how weight history relates to current body-image and weight concerns. In the present study, the "abnormal sensitivity" and "adaptive failure" hypotheses were tested in a nonclinical sample of college women. The sample was homogeneous with respect to age and current average weight, but the women differed with respect to adult weight history. Measures with established psychometric support were employed in the assessment of both perceptual and attitudinal aspects of body-image. The "abnormal sensitivity" hypothesis predicts that both previously t h n n e r and previously heavier women will overestimate their body-size, whereas the "adaptive failure" hypothesis predicts that previously heavier women will overestimate and previously thinner women will underestimate. Both hypotheses predict that women with stable weight histories wdl estimate their body-size with reasonable accuracy. I n addition to inaccurate body-size judgements, women with unstable weight histories were expected to express more negative body attitudes and to report greater preoccupation with weight and dieting than women with stable adult weights.

Participants Participants were 45 undergraduate women (median age = 23 yr.) of average weight who reported that they had never been diagnosed or treated for anorexia nervosa, bulimia, or obesity. Average weight was defined as a weight within 10% of the midpoint of the average range for sex, age, height, and medium body frame (Metropolitan Life Insurance Company, 1983). Groups were formed on the basis of reported adult weight history (highest and lowest weights since Age 16, excluding pregnancy). Women with stable

STABILITY O F WEIGHT HISTORY, CONCERN, BODY IMAGE

493

weight histories (n = 15) reported highest and lowest adult weights that fell within 5% of their current weight. Women with high weight histories (n = 15) reported a highest past weight that was 15% or more above their current weight (M = 17%) and a lowest past weight that fell within 5% of current weight. Women with low weight histories (n = 15) reported a lowest past adult weight that was 15% or more below their current weight (M= -21%) and a highest past weight that fell within 5% of their current weight. All participants were treated according to ethical standards for research with human subjects (American Psychological Association, 1990). Measures and Procedure Participants were individually assessed in the laboratory by the female investigator. Measures were administered according to three different sequences to control for order effects. A general research questionnaire was used to gather demographic and weight history information, including previous diagnosis of or treatment for eating disorder. The women were also assessed on the following measures. Eating Disorder Inventory (EDI).-This measure (Garner & Olmstead, 1984) is a 64-item self-report inventory that assesses disturbed eating attitudes, eating behaviors, and personality features common among individuals with clinical eating disorders. Extensive empirical support for the utility of the inventory for assessing symptoms of eating disorder in clinical and at-risk populations is reported in the manual (see Garner & Olmstead, 1984). The inventory yields scores on eight scales, however, only two scales were of interest in the current study, namely, (a) Drive for thinness, which measures preoccupation with weight and dieting, and (b) Body dissatisfaction, which measures preoccupation with the size and shape of body parts. Body-esteem Scale (BES).-This 35-item scale was developed and validated as a multidimensional measure of body satisfaction (see Franzoi & Shields, 1984; Franzoi & Herzog, 1986; Thomas & Freeman, 1990). The subscales for women assess self-reported satisfaction with body parts and functions related to sexual attractiveness, weight concern, and physical condition. Higher scores reflect greater satisfaction. Figure Ratings (FR).-The stimuli for t h s rating task were developed by Stunkard, Sorenson, and Schulsinger (1983) and consist of nine line-drawn female figures that range ordinally from 1 (very thin) to 9 (very heavy). Consistent with previous use of Figure Ratings (e.g., Fallon & Rozin, 1985; Thompson & Psaltis, 1988), participants indicated the figures that best approximated their current and their ideal body shapes. Larger discrepancies between current and ideal figures reflect greater body-shape dissatisfaction. Videocamera Assessment (VCA).-Participants estimated their own bodysize according to this phototechnical procedure (Freeman, Thomas, Solyom, & Hunter, 1984). The subject's own full-length image, visible on a video

494

C. D. THOMAS

monitor, is varied along a continuous horizontal axis from 80% (thin) to 140% (heavy) of actual size. Current and ideal size estimates are obtained for frontal and profile body views and converted to Body Perception Indices (Slade & Russell, 1973) according to the formula estimated size/actual size x 100. Scores above, at, and below 100 reflect overestimation, accurate estimation, and underestimation, respectively. Discrepancies between current and ideal estimates reflect body-size dissatisfaction. The vahdity of the procedure is good (see Cash & Brown, 1987; Thomas, submitted). At the close of the testing session, the height and weight of each woman was measured; actual body widths at the shoulders, waist, and hips were assessed using standard calipers.

Age, Wcrght, and Weight-history Measures A series of one-way analyses of variance were performed to u~ in age, weight and weight history; see Table assess b e t ~ e e n - ~ r odifferences 1. The groups did not differ significantly with respect to age, height, actual body widths, current weight, current dieting, or amount of weight gained or lost during the previous six months. Women with low weight histories had achieved their lowest weights at a younger age than women with high weight histories (p

Stable vs unstable weight history, body-image, and weight concern in women of average body weight.

The focus of this study was the effect of unstable weight history on current body-image among women. Participants were 45 college women of average wei...
331KB Sizes 0 Downloads 0 Views