Gregory C. Elliott

Mattering: Empirical Validation of a Social-Psychological Concept

Mattering is the extent to which we make a difference in the world around us. People matter simply because: others attend to them (awareness), invest themselves in them (importance), or look to them for resources (reliance). We construct and validate a mattering index, using confirmatory factor analysis. We establish the discriminant validity of the mattering index, using self-consciousness, self-esteem, self-monitoring, alienation, and perceived social support. Finally, we discuss the potential importance of mattering as a dimension of the self-concept.

Understanding stigma: Dimensions of deviance and coping

The present paper deals with stigma and its effects on social interaction. The major thesis is that stigma is a form of deviance that leads others to judge individuals as illegitimate for participation in an interaction, because they are incompetent, unpredictable, inconsistent, or a threat to the interaction. Illegitimacy places the stigmatized beyond the protection of a number of implicit social norms that govern any interaction. The disruptive impact of stigma depends on its classification along several dimensions: visibility, pervasiveness, clarity, centrality, relevance, salience, responsibility for acquisition, and removability. Strategies used to cope with stigma can be understood with reference to their focus (biographical identity, situated identity, or performance), target (stigmatized or normals), impact on the social‐psychological distance between normals and the stigmatized (assimilative or contrastive), and purpose (concealment, reduction of salience, or redefinition of the situation).

Self-esteem and self-presentation among the young as a function of age and gender

This study investigated the effect of self-esteem on the tendency to convey a false impression to others by hiding ones feelings behind a facade. Utilizing a sample of youth (ages 8–19), a multiple regression analysis revealed that the lower the self-esteem, the greater the tendency to present a false front. In addition, other aspects of the self-concept, including vulnerability to criticism, self-consciousness, and the tendency to fantasize affect the tendency to present a false front. Furthermore, the nature of the self-esteem/ fabrication relationship is a joint function of age and gender. As age increases, the direct effect of self-esteem on fabrication loses statistical significance, but this loss occurs differently for boys and girls. For boys, the direct effect is significant in pre-adolescence (8–11), but not in early or late adolescence (12–14 and 15–19). For girls, the direct effect is significant in pre- and early adolescence (8–11 and 12–14), but not in late adolescence (15–19). Reasons for the differing patterns for boys and girls are discussed.

Child Physical Abuse and Self-Perceived Social Isolation Among Adolescents

This research examines the connection between physical abuse and social isolation. Using data from the National Youth Survey, a measure of self-perceived social isolation was constructed indicating the extent to which respondents feel detached from their friends and from school. Those who had experienced violence were predicted to be more isolated than those who had not. Results strongly supported the hypothesis, controlling for theoretically relevant variables. Explanation is provided in terms of damage to attachment skills, social competence, and self-esteem concomitant to being a victim of abuse. Males were more socially isolated than females, and Hispanics more than Whites. Children with involved parents were less socially isolated; those whose parents experienced normlessness were more isolated. Children who recently experienced a stressful event or were from riskier neighborhoods were more isolated. The number of children in the family was positively related to isolation. Social isolation decreases between seventh and eighth grades.

Achieving fairness in the face of competing concerns: The different effects of individual and group characteristics.

This study investigates preferences among allocation procedures available for use by those distributing rewards for individual performance in a group effort. Our first hypothesis is that allocators try to balance the payoff distribution in response to all available information about the individual members. We manipulated a target persons level of contribution to the group effort and level of financial need. Results showed a main effect for both individual characteristics, but no interaction between them, thereby supporting our hypothesis of independent additive effects. Second, we argued that group characteristics would serve only to modify the effects of the individual characteristics. We manipulated two group characteristics: morale and task outcome. As predicted, no main effects for either group level variable emerged. Further, the effect of contribution was greater under failure than success and under low than high morale. The effect of need was greater under low morale, but only for female allocators.

Self-Esteem and Self-Consistency: A Theoretical and Empirical Link Between Two Primary Motivations

This study investigates the relationship between two primary motivational components of the self-concept: self-esteem and self-consistency. Research has shown that high self-e.steem is associated with greater consistency. Developed from theories ofthe self-concept, a structural model relating these two constructs was posited such that; (1) self-esteem is causally prior to self-consistency and (2) the effect of self-esteem is (at least partially) mediated hy other components of the self-concept (self-consciousness. the tendency to fantasize and the tendency to present a false front by hiding ones feelings). An analysis of covariance structures was applied to a structural equation model wilh unobserved variables, using data from a sample of youth (ages 8-19). Mulligroup analyses ofthe sample broken down by developmental stage (pre-. early and late adolescence) revealed that the direct effect of self-esteem on self-consiste ncy and the mediational properties of the other self-concept components varied across age. These differences are interpreted in tight of theories of cognitive and social development.

Interpreting higher order interactions in log-linear analysis.

In the present article, I call for a change in focus in the log-linear analysis of multiway contingency tables. Up to this point, researchers have concentrated largely on procedures for finding the model that most adequately fits the observed data. In many cases, the best fitting model contains one or more complex, higher order interactions involving several variables. It is often desirable to understand the pattern of these interactions. In this article, I present a procedure for substantively interpreting such higher order interactions. This procedure is based on the analogy between log-linear analysis and the general linear model. An example taken from the occupational aspirations literature illustrates the application of the procedure. With the development of the log-linear method, the analysis of nominal and ordinal variables in multiway contingency tables has attained a sophistication comparable to that found in the use of the general linear model to analyze interval and ratio variables. Complex conditional relations can now be discerned, and the procedure allows one to select a model containing such relations that best fits the observed data. (See Bishop, Feinberg, & Holland, 1975; Feinberg, 1981; or Goodman & Magidson, 1978, for a complete presentation and examples of the procedure.) Recently, psychologists have begun to make more extensive use of log-linear analysis (Bonett & Bentler, 1983; Kriska M to determine whether the hypotheses are supported in confirmatory analysis or to extend ones theory in exploratory analysis, one must understand the pattern of the higher order interaction. In the present article, I present a procedure for interpreting such higher order interactions. To illustrate the procedure, I reanalyze a five-dimensional contingency table taken from an investigation of occupational aspirations. Consider Table 1, taken from Sewell and Orensteins (1965) investigation of occupational aspirations of Wisconsin high school seniors. Staring at the table does relatively little to help one identify the patterns of association in the table, especially when one admits that these associations might be quite complex. In applying log-linear analysis to this table, contingent relations that escaped prior analysis come to light, revealing new insights into the determinants of occupational aspirations. Further, the procedure described below provides comprehensible and theoretically meaningful interpretations for any such higher order interactions.

Counting the costs: equity and the allocation of negative group products

Des vignettes decrivant un groupe hypothetique de quatre personnes ont ete presentees a des sujets a qui on a demande de distribuer une somme de 1000 dollars au groupe en modulant les allocations suivant le comportement et les contributions positives ou negatives du groupe

Components of Pacifism

From the philosophical and social science literature on pacifism, four components were derived as characterizing the pacifist: physical nonviolence, psychological non-violence, active value orientation, and internal locus of control. Scales to measure attitudes toward these components were developed and validated. Each scale evinced a strong level of internal consistency. Confirmatory factor analysis revealed that the four components are distinct but related and that each scale is a satisfactory indicator of its intended component. The usefulness of the scales in future research on pacifism is discussed.

Perceived Mattering to the Family and Physical Violence Within the Family by Adolescents

Mattering is the extent to which people believe they make a difference in the world around them. This study hypothesizes that adolescents who believe they matter less to their families will more likely threaten or engage in intrafamily physical violence. The data come from a national sample of 2,004 adolescents. Controlling for respondents’ age, gender, race, religiosity, and family socioeconomic status, structure, and size, logistic regression reveals that mattering to family is a strong contributor to violence. The effect of mattering is mediated by self-esteem and attitude toward violence. Females are more violent than males. Compared with the average respondent, Hispanics are less likely than Whites to commit violence. Children from larger families increasingly use violence. Religiosity diminishes family violence. Children whose responding parent did not finish high school are less likely to turn to violence compared with those whose parent did postcollege study.