Estimation of generalized linear latent variable models

Summary.  Generalized linear latent variable models (GLLVMs), as defined by Bartholomew and Knott, enable modelling of relationships between manifest and latent variables. They extend structural equation modelling techniques, which are powerful tools in the social sciences. However, because of the complexity of the log-likelihood function of a GLLVM, an approximation such as numerical integration must be used for inference. This can limit drastically the number of variables in the model and can lead to biased estimators. We propose a new estimator for the parameters of a GLLVM, based on a Laplace approximation to the likelihood function and which can be computed even for models with a large number of variables. The new estimator can be viewed as an M -estimator, leading to readily available asymptotic properties and correct inference. A simulation study shows its excellent finite sample properties, in particular when compared with a well-established approach such as LISREL. A real data example on the measurement of wealth for the computation of multidimensional inequality is analysed to highlight the importance of the methodology.     

Individualism vs Collectivism in Different Cultures: A cross-cultural study

In this study, individualism vs collectivism, in line with the work of Triandis in 1985, was assessed in two groups of male and female students, in Egypt and Germany. The results confirm the effect that cultural background has on individualist vs collectivist orientations in both of these cultures for male and female students. Men and women scored higher on individualism in Germany than in Egypt, whereas collectivism scores were higher in Egypt than in Germany. These findings are discussed in terms of general recommendations for intercultural interventions that discourage viewing people solely in terms of group membership, and instead as distinct individuals.     

Social influences on sorority and fraternity member smoking

Objective: To understand how the Greek system uniquely influences smoking attitudes, beliefs, and behaviors among college students who belong to a social sorority or fraternity. Participants: Active members of sororities (n = 16) and fraternities (n = 17) were interviewed between February 2013 and October 2015. Methods: In-depth interviews were conducted examining the social influences of the Greek system on member smoking. Transcribed interviews were coded and analyzed for themes. Results: Sorority and fraternity members experienced different social influences on smoking. Sorority stigmatization of smoking outside bars or parties caused sorority members to hide their smoking. Fraternity members had no social penalties for smoking and allowed smoking at the fraternity house. Fraternity members influenced both fraternity members and sorority members to smoke by sharing cigarettes and smoking at parties. Conclusions: Sororities and fraternities have created social environments that both promote smoking and stigmatize it, presenting unique challenges in college student smoking prevention and cessation efforts.     

A decade of changes: Within-group analysis of LGBTQ individuals' perceptions of their community context and the relevance for social service providers

The previous decade has seen a drastic shift in the social environment for lesbian, gay, bisexual, transgender, and queer (LGBTQ) people. It is not surprising then to find that the perceived “best” and “worst” aspects of life for nonmetropolitan LGBTQ individuals have also changed. We use data from two cohorts of the Rainbow Illinois survey to examine differences between responses in 2000 and 2011. Furthermore, we examine demographic differences in the perception of what is “best” and “worst” among the 2011 cohort. Results suggest that although the general social climate has improved for some, it remains problematic for others. Implications for social service providers are thoroughly discussed.     

Treatment Evaluation in the Presence of Sample Selection

Sample selection and attrition are inherent in a range of treatment evaluation problems such as the estimation of the returns to schooling or training. Conventional estimators tackling selection bias typically rely on restrictive functional form assumptions that are unlikely to hold in reality. This paper shows identification of average and quantile treatment effects in the presence of the double selection problem into (i) a selective subpopulation (e.g., working—selection on unobservables) and (ii) a binary treatment (e.g., training—selection on observables) based on weighting observations by the inverse of a nested propensity score that characterizes either selection probability. Weighting estimators based on parametric propensity score models are applied to female labor market data to estimate the returns to education.     

Infants’ Lexical Processing Efficiency is Related to Vocabulary Size by One Year of Age

By 15–18 months, infants’ skill in interpreting familiar words, or lexical processing efficiency (LPE), improves substantially and is correlated with vocabulary size concurrently and several months later. Prior to this age LPE is quite poor, and to date there is little evidence that it is related to vocabulary size. If this relation only emerges once infants have relatively good LPE, and also know a substantial number of words, it could suggest that the processes that support the rapid growth in vocabulary commonly observed as infants approach age 2 may not yet be functional in the earlier stages of lexical development. However, using a modified LPE task we found that 12‐month‐olds with better LPE understood more words at that age and also produced more words several months later. These results suggest that meaningful individual differences in LPE are already emerging by 12 months and may support lexical development across the second year.     

Multiple hypothesis tests based On conditional differences in means

Many hypothesis tests are univariate tests and cannot cope with multiple hypothesis without an auxiliary procedure as e. g. the Bonferroni-Holm-procedure. At the same time, there is an urgent need for testing multiple hypothesis due to the very simple existing methods as the Bonferroni-correction or the Bonferroni-Holm-procedure, which suffers from a very small local significance level to detect statistical inferences or the drawback that logical and statistical dependencies among the test statistics are not used, whereby its detection is NP-hard. In honour of this occasion, we present a multiple hypothesis test for i.i.d. random variables based on conditional differences in means, which is capable to cope with multiple hypothesis and does not suffer on such drawbacks as the Bonferroni-correction or the Bonferroni-Holm-procedure. Thereby, the computation time can be neglected.     

Interval censored and truncated data: Rate of convergence of NPMLE of the density

We consider survival data that are both interval censored and truncated. Under appropriate assumptions on the involved distributions, the censoring, truncation and survival, we prove the consistency of the NPMLE of the density of the survival, and give the rate of convergence. Finally, we give an example where the joint law of the censoring and truncation can be explicitly computed.