THE CODE: A SHORT QUESTIONNAIRE FOR IDENTIFYING CO-DEPENDENCY AMONG UNIVERSITY STUDENTS

Abstract

PURPOSE: the purpose of this study was to develop an instrument to predict co-dependency problems among university students to assist psychologists, counselors, health educators, and student personnel staff to determine if a student might have a family background of alcoholism without directly asking as some students are in denial concerning family alcoholism. METHODS: A literature search revealed 63 variables associated with those who had alcoholism in their immediate families sometimes called “co-dependents.” Individuals were classified as Co-Dependent if they indicated that either a parent or grandparent had sometimes or often drank too much. Individuals were classified as non-co-dependent if all relatives had never or only sometimes drank. Content validity for questionnaire development was determined by a jury of over 30 professions in the field of alcohol and drugs. Then the instrument was administered to an undergraduate class of 100 university students who were asked to comment on the items. Next 15 self-identified co-dependents attending a seminar on co-dependency were asked to comment on the items. A four-point Likert scale was developed. The finalized version was sent to four universities who volunteered assistance. After elimination of items due to non-significant differences between co- and non-co-dependents or items with reliability less than .3, eleven items remained. For these 11 items, a mean score was determined for each individual. Factor analysis revealed two factors which accounted for 59% of the variance. Since the reliability between both factors was .8, it was decided that one factor could be used for a total score. Spearman-Brown split half technique for internal consistency found a reliability coefficient of .87 and the Cronbach alpha test for homogeneity resulted in an alpha of .89. Discriminate analysis correctly classified individuals as co-depended 69% of the time. Cross validation between two forms of the check list found no significant difference between any items. A highly significant difference was found between co- and non-co-dependents on both forms. RESULTS AND CONCLUSION: These procedures indicated that this eleven items instrument was highly reliable, had high internal consistency and could correctly predict a person’s co-dependency - or family background of alcoholism - status approximately 70% of the time. The instrument also could be used intact or embedded with other items and is useful to identify familial alcoholism.