The development and validation of the Stirling Eating Disorder Scales
The thesis addresses three main research issues in eating disorders. Study One compared anorexic, bulimic, obese dieters, non-obese dieters and normal controls on measures of eating behaviour, perceived control, assertiveness, self-esteem, and self-directed hostility. Results indicated that eating disorder patients could be differentiated from both dietary/weight concern groups and normal controls. Study Two developed and validated a new primary eating disorder assessment measure - the Stirling Eating Disorder Scales (SEDS). The SEDS were developed according to the Thurstone Method of scale construction and fully standardised. Results indicated that the SEDS are internally consistent, between group and concurrently valid, reliable, and are not subject to gender or response bias. Study Three assessed the SEDS in terms of sensitivity to detect change in patients over treatment time. Eating disorder patients undergoing treatment, completed the SEDS and two other standardised measures on three occasions over six months. Results indicated that the SEDS are sensitive and detect change in the patients dietary/behaviours and cognitive/emotions over treatment time. Study Four compared eating disorder groups with depressed and panic disorder patients and normal controls on the SEDS and three other psychological measures. Results indicated that eating disorder patients can be differentiated from panic disorder and controls on all scales, but are similar to depressed in terms of perceived external control, low assertiveness, and low self-esteem. Differences/links between eating disorders and other psychological groups and the criterion validity of the SEDS are discussed.