This article describes the user-written program margeff, which enables the fast estimation of (average) marginal effects. Besides describing the program, this article offers a new discussion of some problems that are related to computation of marginal effects. I will argue that (1) marginal effects computed at means are not good approximations of average marginal effects, computed as means of marginal effects evaluated at each observations, if some of the parameter estimates are large; (2) both average marginal effects and marginal effects computed at means might produce wrong estimates for dummies that are part of a set of indicator variables indicating different categories of a single underlying variable; and (3) the use of marginal effects computed at means is preferred if some of the regressors are mathematical transformations of other regressors.
By examining social attitudes on same-sex adoption in 28 European countries, we highlighted individual and country-level factors that can determine the level of social acceptance or rejection of this specific kind of adoption. This article contributes to the literature on social acceptance of lesbian women, gay men, and their adoption practices in Europe and directs attention to several previously under-researched aspects of social attitudes on same-sex parenting rights. The empirical base of this study was the fourth round of the European Values Study, conducted in 2008-2010. Using ordered logistic regressions, we examined the impact of several individual and country-level characteristics on the agreement level with the statement that "Homosexual couples should be able to adopt children." We found strong relationships between social attitudes towards adoption by same-sex couples and the existence of legislation permitting same-sex adoption practices at the country-level, as well as some individual attitudes, including those related to traditional family formation practices, "justification of homosexuality," and (non-) preference for homosexual neighbors. Our findings indicate a shift within the potential interpretational contexts of adoption by same-sex couples from a narrow sexuality-based framework to a different and possibly much wider context of family and parenting practices.
Multilevel multiprocess models are simultaneous equation systems that include multilevel hazard equations with correlated random effects. Demographers routinely use these models to adjust estimates for endogeneity and sample selection. In this article, I demonstrate how multilevel multiprocess models can be fit with the gsem command. I distinguish between two classes of multilevel multiprocess models: nonrecursive systems of hazard equations without observed endogenous variables and recursive systems that include a hazard equation with observed endogenous qualitative variables. I illustrate the estimation of both classes of models using sample datasets shipped with the statistical software aML. I pay special attention to identifying structural coefficients in nonrecursive simultaneous systems.
Multilevel multiprocess hazard models are routinely used by demographers to control for endogeneity and selection effects. These models consist of multilevel proportional hazards equations, and possibly probit equations, with correlated random effects. Although Stata currently lacks a specialized command for fitting systems of multilevel proportional hazards models, systems of seemingly unrelated lognormal survival models can be fit with the user-written cmp command (Roodman 2011, Stata Journal 11: 159-206). In this article, we describe multiprocess survival models and demonstrate theoretical and practical aspects of estimation. We also illustrate the application of the cmp command using examples related to demographic research. The examples use a dataset shipped with the statistical software aML.
It is well known that participation in education is incompatible with the transition to motherhood. However, enrolment is overwhelmingly treated as a single status even though participation in education may be combined with employmentresulting in double-status positions, and the fertility implications of double-status positions are less clear-cut. Relying on normative and economic approaches, we develop original and competing hypotheses regarding the effect of double-status positions on the transition to motherhood. We also speculate on how the postcommunist transition and institutional context might influence the hypothesised effects. The hypotheses are tested using event history data from the Hungarian Generations and Gender Survey. We employ event history methods, which take into account the potential endogeneity of employment and enrolment decisions. We find robust evidence that first birth rates are higher among women in double-status positions than among women who are merely enrolled, but that difference is smaller in younger cohorts than in older ones. We also find some evidence that first birth rates are lower in double-status positions than among women who are employed but not enrolled. Our findings suggest that the conflict between participation in education and motherhood is mitigated in double-status positions, especially among members of the oldest cohort. Since double status is prevalent in modern societies, but has different meanings in different contexts according to educational system and welfare state, we argue for future research on this issue.
This paper addresses two questions: (1) Do informal job searchers find good jobs in Hungary? (2) Do social resources theory and the theory of employee referrals explain the conditions under which informal job searchers find good jobs? The questions are examined using a rather unique dataset which was collected among people who completed secondary vocational education in 1998. It is found that the use of informal methods in itself rarely promotes one's chances of finding a good job. Rather, good jobs can be accessed through either high status contacts or employee referrals. These findings are consistent with both social resources theory and the theory of employee referrals.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.