Multiple-group Structural Modeling Research Articles

Balancing typological and dimensional approaches: Assessment of adult attachment styles with Factor Mixture Analysis.

Many studies show the importance of adult attachment styles and their impact on social and emotional adaptation in adulthood. However, there is no agreement about whether attachment should be regarded as typological or dimensional, and some authors have proposed reconciling both options, so that adult attachment styles can be assessed more accurately and realistically. In this study we have adopted this comprehensive view and used Factor Mixture Analysis, the most appropriate model for assessing this mixture view. More specifically, we attempted to determine the nature and types (if any) of attachment styles that can be assessed with the Adult Attachment Questionnaire (CAA), using this mixture view. A total of 515 adults from Spain took part, with ages between 18 and 56 years old. In addition to the CAA questionnaire, they completed the Parental Bonding Instrument and the emotional stability subscale of the Overall Personality Assessment Scale. On the basis of the CAA scores, the results show that only two profiles-insecure attachment and normal-range-can be univocally differentiated. Furthermore, the results of a full multiple-group structural model show that each of these profiles has a different pattern of validity relations with the external variables maternal care, maternal overprotection and emotional stability. These differential validity results reinforce the general hypothesis that two differentiated clusters of individuals can be distinguished on the basis of the responses to the CAA items.

Managing E-commerce Platform Quality and its Performance Implication: A Multiple-Group Structural Model Comparison

ABSTRACTThis study develops a comprehensive e-service quality measure of e-commerce platforms as intermediaries, which considers both the functionality of the website itself and the performance assessment of independent sellers. The study further investigates the joint impacts of both the attributes of e-commerce platforms and the performance of independent sellers on users’ online shopping experience: perceived online transaction value and the ensuing satisfaction. Finally, the study examines the effects of e-service quality on the outcome variables across two major e-commerce platforms as intermediaries, eBay and Amazon. The multiple-group structural model analysis provides important insights to understanding differing impacts of different quality components on shopping experiences across various online marketplaces. Theoretical and managerial implications are discussed.

Testing Self-Segregation: Multiple-Group Structural Modeling of College Students’ Interracial Friendship by Race

Using structural equation modeling, this study examined the effects of peer environments on collegiate interracial friendship and how such effects vary by students’ race. The results show that the peer environment of Greek life mediated the relationship between structural diversity and interracial friendship in college, in that students attending institutions with greater structural diversity were less likely to frequently interact with peers from Greek life, which had a positive effect on interracial friendship. This mediation effect was consistent for all four racial groups of the study. This study also uncovers unique findings related to Latino/as and interracial friendship: that structural diversity has an indirect effect on interracial friendship via participation in ethnic student organizations for Latino/a students and that participation in ethnic student organizations is also directly and positively linked to interracial friendship for this group. The study discusses implications for understanding intergroup relations and patterns of interracial friendship.

Abstract CN11-02: Models of the pathways between socioeconomic status and health

Abstract CN11-02: Models of the pathways between socioeconomic status and health

Moderators of the relation between substance use level and problems: Test of a self-regulation model in middle adolescence.

The authors tested predictions, derived from a self-regulation model, about variables moderating the relationship between level of substance use (tobacco, alcohol, and marijuana) and problems associated with use. Data were from two independent studies of adolescents, with mean ages of 15.4 and 15.5 years (Ns = 1,699 and 1,225). Factor analysis indicated correlated dimensions of control problems and conduct problems. Protective moderation was found for variables indexing good self-control; risk-enhancing moderation was found for variables indexing poor self-control. These effects were generally independent of deviance-prone attitudes and externalizing symptomatology. Multiple-group structural modeling indicated moderation occurred for paths from life stress and coping motives and for paths from level to control and conduct problems. Moderation effects were also found for parental variables, peer variables, and academic competence.

Sex differences in self-reported delinquency and official records: A multiple-group structural modeling approach

There has been a large increase in research in the last 10 years or so on the nature, extent, and causes of female delinquency, especially on how patterns differ from those of male delinquency. Few research efforts, however, have systematically examined sex differences in the reliability and construct validity of the most common technique used in causal research-the self—report method. The present study addresses this issue by empirically examining male-female differences in self-report reliabilities and the effect of the prevalence and incidence of delinquent behavior on official contact (i.e., arrest, court referrals). Multiple-group covriance structure models are employed to test hypotheses of measurement and structural invariance. The major conclusion is that while self-report measurement parameters (e.g., reliabilities) may differ, the structural link between the prevalence and incidence of delinquent behavior and official contact is generally invariant with respect to sex.