Reliability measures in distorting body-image.

This study is the first to make a direct comparison between the distorting mirror and the distorting videocamera with the same group of nonclinical subjects. It also establishes the internal consistency, the test-retest reliability and the convergent validity of both techniques. Analysis indicates, as with other studies, that perceived image is consistently over-estimated and ideal image is consistently under-estimated by both techniques. With the distorting mirror, the frontal orientation is repeatable over four days and with the video the profile is the more reliable orientation. A comparison between profile and frontal orientations on any single test occasion (internal consistency) indicates that the results correlate well so either orientation is acceptable for experimental purposes. The extent to which both techniques were measuring the same aspect (convergent validity) was more marked for the ideal image. A researcher using the distorting mirror can be confident that the ideal image is reliable over time using either orientation but that the perceived image requires a frontal orientation. The frontal or profile orientation is acceptable for the ideal image using the distorting video, but the profile orientation is advisable if one wishes a reliable perceived image. The researcher should also be cautious in assuming that both techniques are measuring an identical body-image construct.