Family Cohesion and Enmeshment: Different Constructs, Different Effects

Abstract

Some past conceptualizations in family research have synthesized the constructs of family cohesion and enmeshment by placing enmeshment at the high extreme end of cohesion. In this article, we argue that, theoretically, the 2 are different constructs-cohesion is a measure of supportive interaction, and enmeshment is a measure of psychological control. We examine this hypothesis by testing the associations between adolescent reports of family cohesion and enmeshment and several measures of adolescent problem behaviors using a sample of 471 students in preadolescence and early and middle adolescence from a suburb of a city in the South. Results show that cohesion is associated negatively with both internalizing and externalizing adolescent problem behaviors. Enmeshment is related positively with youth problems, and more strongly with internalizing problems. Furthermore, different patterns of interaction emerge among the 2 family variables and the adolescents' grade in school and sex. Key Words: adolescence, cohesion, enmeshment, family func- . tioning, psychological control. This article considers family cohesion and enmeshment as separate aspects of family functioning and examines their differential relations with one specific facet of individual functioning within families-adolescent problem behaviors. Family cohesion is defined as shared affection, support, helpfulness, and caring among family (Barbarin, 1984; Moos, 1974). Enmeshment is defined as family patterns that facilitate psychological and fusion among family members, potentially inhibiting the individuation process and the development and maintenance of psychosocial maturity (Barbarin, 1984; Barber, Olsen, & Shagle, 1994; Greenberger & Sorensen, 1974). Examination of several prevailing models of family functioning reveals that family cohesion and enmeshment are two important aspects of family life (Beavers, Lewis, Gossett, & Phillips, 1975; Bloom, 1985; Epstein, Bishop, & Baldwin, 1982; Moos & Moos, 1976; Olson, Russell, & Sprenkle, 1983; Skinner, Steinhauer, & SantaBarbara, 1983). However, the relationship between these aspects of family interaction has been controversial. Some family scholars have synthesized the constructs by suggesting that a high level of cohesion represents enmeshment in families (Epstein et al., 1982; Olson et al., 1983; Skinner et al., 1983). Based extensively but not exclusively on the clinical writings of Bowen (1960), Minuchin (1974), and Stierlin (1974), Olson et al. defined family cohesion as the emotional bonding that family have toward one another (p. 70). In this conceptualization, there are varying levels of cohesion, ranging from disengaged (very low) to enmeshed (very high). Olson et al. suggested that the mid-range levels of cohesion (i.e., separated and connected) are most viable for family functioning, whereas the extreme ends are problematic for most families. From this perspective, highly cohesive families are enmeshed and, therefore, discourage efforts toward individuation through communication patterns that are psychologically and emotionally inhibitive or intrusive. This conceptual synthesis of family supportiveness and psychological intrusiveness is evident also in the models developed by Epstein et al. (1982) and by Skinner et al. (1983). For example, in the McMaster Model of Family Functioning, Epstein et al. included a concept labeled that is similar to the cohesion concept defined by Olson et al. (1983). Epstein and colleagues defined affective involvement as the extent to which the family shows interest in and values the particular activities and interests of individual family members (p. 127). Six levels of involvement were specified: lack of involvement, involvement devoid of feelings, narcissistic involvement, empathic involvement, overinvolvement, and symbiotic involvement. Empathic involvement is considered optimal for health. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 284 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.