A study of marital and family boundaries and their relationship to symptomatic behavior in families with substance abusing adolescents

Kevin Marvell Marett, Purdue University

Abstract

This study explored the assumption from structural family therapy that a relationship exists between family/marital boundaries and the symptomatic behavior of adolescents. The nature of this relationship, i.e. linear or curvilinear, was also explored. This research employed a survey design using six self-report measures involving the perspectives of mother, father, and a substance abusing adolescent child. The three measures of the independent variable of boundary stability, or the degree to which systems or subsystems are distinguishable from other systems or subsystems, included the Dyadic Formation Inventory (Dyadic Inclusiveness and Dyadic Exclusiveness Subscales), the Family Adaptability and Cohesion Evaluation Scales (FACES) III (Real Version), Cohesion scale, and the Parent-Adolescent Communication form. The three measures used for the dependent variable of symptomatic behavior, or the extent to which adolescents were violating established norms, consisted of the Polydrug Use History Questionnaire, the Family Problems Assessment Scale, and the School and Work Information form. Each independent variable (IV) instrument, which included scores from mother, father, and adolescent, was regressed with each dependent variable (DV) instrument, resulting in nine multiple regression models. The individual perspective scores i.e. mother, father, adolescent, from each IV measure were also regressed on the Family Problems Assessment Scale (FPAS) thereby producing three more regression models. Of the twelve regression models, only two FACES III regressed on FPAS and the father's perspective from each of the IV instruments regressed on the FPAS, showed significant support for the assumption that a relationship exists between family boundaries and problematic behavior in adolescents. Due to contradictory findings, the nature of this relationship is still in question as evidence supporting both linear and curvilinear distributions was found--scatterplots failed to show any indication of curvilinearity, but FACES III showed the sample used for this study to be truncated, with virtually no families in the enmeshed category.

Degree

Ph.D.

Advisors

Sprenkle, Purdue University.

Subject Area

Families & family life|Personal relationships|Sociology

COinS