A Graphical Tool for Interpreting Regression Coefficients of Trinomial Logit Models

Multinomial logit (also termed multi-logit) models permit the analysis of the statistical relation between a categorical response variable and a set of explicative variables (called covariates or regressors). Although multinomial logit is widely used in both the social and economic sciences, the interpretation of regression coefficients may be tricky, as the effect of covariates on the probability distribution of the response variable is nonconstant and difficult to quantify. The ternary plots illustrated in this article aim at facilitating the interpretation of regression coefficients and permit the effect of covariates (either singularly or jointly considered) on the probability distribution of the dependent variable to be quantified. Ternary plots can be drawn both for ordered and for unordered categorical dependent variables, when the number of possible outcomes equals three (trinomial response variable); these plots allow not only to represent the covariate effects over the whole parameter space of the dependent variable but also to compare the covariate effects of any given individual profile. The method is illustrated and discussed through analysis of a dataset concerning the transition of master’s graduates of the University of Trento (Italy) from university to employment.     

Sektion Frauen- und Geschlechterforschung

Ohne Zusammenfassung     

The Impact of Personal Finance Education Delivered in High School and College Courses

This study investigates the impact of personal finance education delivered in high school and college. Outcomes of interest were investment knowledge and household savings rates measured years after the financial education was delivered. A web-based survey with questions about participation in financial education, financial experiences, income and inheritances, and demographic characteristics was administered to 1,039 alumni from a large midwestern university. Participation in a college level personal finance course was associated with higher levels of investment knowledge. Experience with financial instruments appeared to explain more of the variance in both investment knowledge and savings rates. No significant relationship between taking a high school course and investment knowledge was found. Financial experiences were found to be positively associated with savings rates.

Veranstaltungshinweise

Veranstaltungshinweise

Veranstaltungshinweise     

Do Reports of Age and Circumstances of First Intercourse Differ in a Birth Cohort When Asked Seventeen Years Apart?

Conclusions about temporal changes in age and circumstances of first intercourse are generally derived from retrospective reports by people of various ages in cross-sectional studies, with an inherent assumption of no bias stemming from time since the event. We examined this assumption through repeated questions on age and circumstances of first heterosexual intercourse (FHI) at ages 21 and 38 in a birth cohort. Despite considerable movement in individual reports, there was no bias in reported age of FHI. However, a greater proportion of both men and women stated at the later assessment both partners had been equally willing (versus persuading or persuaded). The distribution of current views of the appropriateness of the timing did not differ markedly between assessments, although there were many individual changes. Reports of contraceptive usage were similar at the two assessments for men but differed among women, mainly through more reporting that they could not remember. These findings imply that among cohorts born in the 1970s, there is no bias in reports of age of FHI many years after the event, and views on the appropriateness of timing persist. However, time biases reports in favor of a more mutual willingness.     

Separations: A Personal Account of Bowen Family Systems Theory

This paper is a personal account of my own family of origin research. It explores the impact of separations from parents, nuclear family, and extended family through a Bowen family systems theory perspective using concepts pertaining to Bowen theory such as chronic anxiety, differentiation of self, multigenerational family process, and the emotional system. An outline of the process of doing research in ‘vivo’ with my mother as well as conversations with my supervisor are included. Theoretical differences between individual and system models are discussed. A Bowen theoretical approach to the anxiety of separations is investigated. The efficacy of engaging in family of origin work and the effects of thinking systems is examined in light of how it assisted me to view family members, family system disturbances, and clients’ emotional systems more objectively.     

Exposure to occupational violence and the buffering effects of intra-oranizational support

This paper presents a conceptually distinct analysis of the data first reported by Leather et al. Specifically, it investigates the effects of exposure to a variety of forms of work-related violence upon work attitudes and general well-being within a sample of UK public house licensees—individuals who manage public houses and hold the licence permitting the sale of alcoholic drinks on the premises. In addition, it examines the role of social support in moderating such effects. Based upon a sample of 242 licensees, it reports a consistent interaction between exposure to such violence and the availability of perceived intra-organizational support in determining the size of any negative effects upon individual well-being, job satisfaction and organizational commitment. It is argued that exposure to all forms of work-related violence, including intimidation, verbal abuse and threat, should be seen as a potential stressor within the work environment, the negative effects of which are buffered by perceived support from within the organization, but not by that perceived to be available from informal sources such as family and friends.     

Exponential Family Models for sampled and census Network data

Much progress has been made on the development of statistical methods for network analysis in the past ten years, building on the general class of exponential family random graph (ERG) network models first introduced by Holland and Leinhardt (1981) . Recent examples include models for Markov graphs, "p*" models, and actor‐oriented models. For empirical application, these ERG models take a logistic form, and require the equivalent of a network census: data on all dyads within the network. In a largely separate stream of research, conditional log‐linear (CLL) models have been adapted for analyzing locally sampled ("egocentric") network data. While the general relation between log‐linear and logistic models is well known and has been exploited in the case of a priori blockmodels for networks, the relation for the CLL models is different due to the treatment of absent ties. For fully saturated tie independence models, CLL and ERG are equivalent and related via Bayes' rule. For other tie independence models, the two do not yield equivalent predicted values, but we show that in practice the differences are unlikely to be large. The alternate conditioning in the two models sheds light on the relationship between local and complete network data, and the role that models can play in bridging the gap between them.