Boyd C. Rollins’s research while affiliated with Brigham Young University and other places

The parent—child relationship initiates a child into the social world and reshapes components of the adult self-concept into identification with parental roles. Much of what occurs between parents and children transforms a biological organism into a human being and confronts adults with a new set of experiences and responsibilities. Through this facet of the socialization process, parents and children acquire the knowledge, attitudes, skills, values, and expectations that allow them to become increasingly integrated into new social relationships.

Two contemporary theoretical explanations of adolescent self-esteem, symbolic interaction and social learning, were investigated and compared. Special attention focused on the relative effect of selected variables, representing each explanation, on four dimensions of self-esteem. A stratified random sample of 184 families with adolescents provided self-report data. Multiple regression and bivariate analysis resulted in evidence for the general conclusions that: (1) adolescent self-esteem was more a function of the reflected appraisal of the parents than it was of adolescents modeling their parents' self-esteem; and (2) female adolescents were more likely to be influenced by their parents than were male adolescents. In addition, the study suggests that when researchers investigate adolescent self-esteem, it is essential that they take into account its various dimensions, as well as the sex of the parent and the-adolescent.

Compared 2 contemporary theoretical explanations of adolescent self-esteem: symbolized interaction and social learning. Special attention focused on the relative magnitude of effect that a selected cluster of variables, representing each explanation, had on 4 dimensions of self-esteem: self-esteem worth, self-derogation, positive self-esteem, and self-esteem power. Self-report data were gathered from a stratified random sample of 184 families with 368 adolescents. Multiple regression and the multiple partial correlation coefficient were used to analyze the data. Findings suggest that in 9 of 12 dyad-dimension comparisons wherein the desired level of significance was obtained, both symbolic interaction and social learning variables simultaneously contributed to the overall variance of adolescent self-esteem. It is concluded that an examination of the antecedents of adolescent self-esteem from one theoretical perspective, while excluding the other, limits understanding of the phenomenon. Therefore, if observed variation is to be accounted for, a middle-range theoretical framework must be constructed. (35 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)

Philosophical issues are raised about the diverse and complex methods used in studying marriage and family phenomena. Research methods in this field are recognized as having much in common with the more general research strategies of the social and behavioral sciences. Advances in methodological approaches and tools are reviewed, emphasizing design, measurement, and analytic developments that have been most important to marriage and family research. Several emergent issues are highlighted because they seem especially likely to influence the future of research in this field. These issues include assessing the fit between empirical methods and theories, operationalizing group properties, increasing the use of secondary analysis of existing data, and the role of ethics and values in marriage and family research.

In a simulation game designed for families to plan adolescents' career goals, 183 families were studied for evidence that traditional sex-role attitudes were reflected in family decisions regarding the career goals of teenagers. Special attention was focused upon 96 of these families, each having both a male and female adolescent. Results indicated that family decisions favored the career goals of adolescent males over adolescent females by a 4 to 1 ratio. That is, the differential preference given to adolescents' goals seemed to be determined by the adolescents' gender. Furthermore, this variable retained its explanatory efficacy despite analyses for the effects of such alternative variables as the ordinal position of the adolescent, religious preference of parents, religiosity of parents, education of fathers, fathers' income, employment status of mothers, and whether the daughters desired an occupational or homemaking career as a primary life goal. Another important finding was that daughters were divided about equally in their choices of an occupational versus homemaking career. Generational differences were also prevalent, because fathers clearly preferred domestic goals for their daughters more than adolescent females chose these goals for themselves.

The study empirically combines two commonly used measures of adolescent self-esteem, the Rosenberg Self-Esteem Scale and a modified version of the Osgood Semantic Differential, with special attention focusing on the degree of association between the two measures. A crucial aspect of the study focused upon ascertaining the degree of association between four suggested independent dimensions of self-esteem derived from the two instruments. In addition, the study utilized sex and generation variables in assessing structural equivalence. A stratified random sample of 184 families with adolescents was taken. Self-report data were gathered from this sample. Factor analysis with varimax rotation and the Cronbach Reliability Coefficient were the statistical procedures employed. The findings of the study suggest two important considerations for self-esteem research and application: first, the results indicate that four valid and reliable dimensions of self-esteem can be derived from the two instruments; and second, the dimensions derived demonstrate structural equivalence.

Attempts to clarify conceptualizations in the general area of marital power. The concept of power is differentiated from control attempt, control, resources, and authority. A theory is developed which specifies the relationships among these 5 variables. It is assumed that (a) power and control are social interaction constructs rather than attributes of individual persons; (b) power and control are relevant constructs only when a conflict exists between the goals of marriage partners; and (c) authority, resources, and power do not exist independently of perceptions. The theory clarifies a number of conceptual ambiguities and appears to be useful for integrating empirical research in this area. (40 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)

While a variety of scales and instruments to measure parental support exist, most of these have not received adequate analysis. Using data from the studies of J. M. Colonico and D. L. Thomas (1973) and C. Allen (1975), the Parental-Child Interaction Rating Scale, the Cornell Parent Behavior Description scale, and the Parent Behavior Inventory were examined to determine whether (a) the 3 measures interrelate in meaningful ways, (b) support is a unidimensional construct, and (c) assuming multidimensionality, subscales with adequate reliability and validity can be created. The factor analytic results extracted multiple dimensions of parental support, thus supporting the assumption of conceptual differentiation within the central construct. They further indicate similarity between the first 2 scales and less similarity between the first and last scales. The various subscales of parental support were shown to have adequate alpha coefficients of internal consistency and fair discriminate and construct validity. (16 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)

The present study investigates the effects of crisis on conjugal power using SIMFAM, a laboratory method. Research on non-family groups has shown that during an unsolved crisis a leader tends to decrease in power and to be replaced. The relevance of these ideas for marital dyads was tested, as well as the idea that the more equalitarian the conjugal power structure during a non-crisis situation, the more likely a change in relative conjugal power will occur during a crisis. Hypothesis based on these ideas were supported with a sample of 50 married couples.