Validation of an interval scaling: the sickness impact profile

Abstract

The Sickness Impact Profile (SIP) is a measure of sickness-related behavioral dysfunction consisting of 189 items in 14 topic categories. To increase its discrimination, precision, and sensitivity in accounting for variance, the decision was made to scale the instrument. A two-step direct scaling procedure was used in order to avoid the monumental scaling tasks required by indirect procedures that guarantee equal-interval results; but because an equal-interval scale was needed, it was necessary to validate the scale values obtained and investigated the equal-intervval properties of the obtained scale. A three-stage validation process is described, consisting of an initial scaling by a group of 25 health professionals and students in 1973, a second scaling by 108 members of a prepaid group health plan in 1975, and an investigation of the metric properties of the resulting scale values. In addition, the concept of dysfunction underlying the SIP was validated. SIP scores from a field trial were compared with mean ratings of severity of dysfunction represented by the combinations of checked items from which the scores were derived.