Dimension of familism: A study of roles and relationships
Purpose. The major focus of this study was to clarify the problem proposed by Nye and Rushing (1969) of conceptualization and measurement of integration into one's family. Familism and integration into one's kin-network were identified as one and the same concept. The dimensions of familism were specified as attitudes, perceptions, and behaviors (the same as Nye and Rushing's use of the terms consensual, affectual, and association integration), and the sum of these or cumulative familism.
Methodology. A mailed questionnaire was used to collect data from residents in one Texas community. The sampling method chosen was systematic random sampling using the Denton telephone directory as the sampling frame. Data were collected during July and August 1979. The two weighted samples yielded a 51 percent return rate. The final sample on which the study was based consisted of 145 persons. While not representative of the ethnic or sexual characteristics of the community, the sample approximated the modal characteristics present.
Hypotheses and Findings. The hypotheses were relational and non-directional in nature. Hypothesis I. "Age and role-position will be associated with the dimensions of familism" was in general substantiated. Hypothesis II. "A relationship will exist between Heller's familism scale and a derived attitudinal scale" was also supported, although the derived attitudinal scale was more highly associated with and predictive of other components of familism than was Heller's scale. Hypothesis III. "A relationship will exist between the familistic dimensions" was also supported. Attitudes and behaviors together were found to be predictive of r('2) .88 or almost 90 percent of the variation in cumulative familism. Perceptions, while related to both perceptions and behaviors, were not highly predictive of cumulative familism.
In addition, data used to analyze Research Question I. "What will be the pattern of familism among the sample," revealed that the study sample reflects patterns of the companionate family (Burgess) and the conjugal family (Goode).
Summary and Conclusions. The model was supported. Familism appears to be both a composite of several dimensions and to consist of separate components. The interactive effect between attitudes, perceptions, and behaviors was demonstrated, and the multivariate analyses were used to form a heuristic model of familism. Familism appears to be embedded within a network of family relationships, and while changes in components of familism change throughout the life cycle, cumulative familism appears to be age independent.