Retest reliability of scores on objective and projective measures of dependency: relationship to life events and interest interval.

The retest reliabilities of widely used objective and projective measures of dependency were assessed in a mixed-sex sample of undergraduates (54 women and 34 men). Subjects completed Hirschfeld and colleagues' (1977) Interpersonal Dependency Inventory (IDI) and Masling, Rabie, and Blondheim's (1967) Rorschach Oral Dependency (ROD) scale on two occasions separated by 16, 28, or 60 weeks. The IDI and ROD scale showed good retest reliability over 16 weeks in both men and women. However, the ROD scale did not show adequate retest reliability over longer periods in subjects of either sex. IDI scores showed excellent long-term retest reliability in women, but poor long-term retest reliability in men. Subjects' self-reports and impact ratings of life events experienced during the intertest period were unrelated to changes in subjects' IDI and ROD scale scores from Time 1 to Time 2, regardless of the intertest interval used.