MIMCA: multiple imputation for categorical variables with multiple correspondence analysis

We propose a multiple imputation method to deal with incomplete categorical data. This method imputes the missing entries using the principal component method dedicated to categorical data: multiple correspondence analysis (MCA). The uncertainty concerning the parameters of the imputation model is reflected using a non-parametric bootstrap. Multiple imputation using MCA (MIMCA) requires estimating a small number of parameters due to the dimensionality reduction property of MCA. It allows the user to impute a large range of data sets. In particular, a high number of categories per variable, a high number of variables or a small number of individuals are not an issue for MIMCA. Through a simulation study based on real data sets, the method is assessed and compared to the reference methods (multiple imputation using the loglinear model, multiple imputation by logistic regressions) as well to the latest works on the topic (multiple imputation by random forests or by the Dirichlet process mixture of products of multinomial distributions model). The proposed method provides a good point estimate of the parameters of the analysis model considered, such as the coefficients of a main effects logistic regression model, and a reliable estimate of the variability of the estimators. In addition, MIMCA has the great advantage that it is substantially less time consuming on data sets of high dimensions than the other multiple imputation methods.     

The impact of child support on child health

The broad goals of child support policy are to keep children in single-parent families out of poverty and to make sure that their material needs are met. One potentially important, but relatively understudied, set of measures of child well-being are health outcomes. A fixed-effects analysis of data from the Child and Young Adult file of the 1979 National Longitudinal Survey of Youth shows that, conditional upon receipt of some amount of child support, higher payment levels are associated with significantly greater odds of having private health insurance coverage and significantly lower odds of poor or declining health status. These effects persist even after controlling for other factors that are likely to be correlated with child support payments, including total family income and paternal visitation patterns.     

Does Sport Celebrity Advocacy Work? Testing the Potential for Endorsements to Backfire

Fandom and psychological reactance theory, a framework for understanding reactions to messages perceived to limit an individual’s freedom to choose freely, are used to examine the influence of sports celebrities. An experiment was conducted (n = 283) to examine the impact of source similarity and controlling language use on reactance levels and message evaluations. Findings suggest that sources perceived to be similar generated less reactance than dissimilar sources. Similar source messages were also linked to greater potential for information-seeking behavior. High- versus low-controlling language use was examined in conjunction with similarity; low-controlling language was found to generate the least reactance regardless of source similarity. Implications for the use of celebrity advocacy in marketing and public relations are discussed.     

Dyadic Analyses of Preschool‐aged Children's Friendships: Convergence and Differences between Friendship Classifications from Peer Sociometric Data and Teacher's Reports

Peer sociometrics and teachers' friendship reports were compared in 2179 preschool dyads. One hundred twenty of 306 reciprocated friend dyads from peer sociometric data were also identified as good friends by their classroom teachers, and 301 of 600 of non‐reciprocated dyads in peer data were named as friends by one or both classroom teachers (overall kappa = .16). Friendship classifications from both peer and teacher data had significant relations with variables relevant to peer interactions, social skills, peer acceptance, and teacher‐rated scales (six of seven tests significant for peer data; five of eight significant for teacher data). Multilevel analyses indicated that friendship status effects were not qualified by classroom‐level differences. Findings suggest that sociometric tasks can identify preschoolers' peer friendships and that the range of correlates may be broader in peer‐choice data than in teachers' friendship evaluations.     

An Extended Single‐index Model with Missing Response at Random

An extended single‐index model is considered when responses are missing at random. A three‐step estimation procedure is developed to define an estimator for the single‐index parameter vector by a joint estimating equation. The proposed estimator is shown to be asymptotically normal. An algorithm for computing this estimator is proposed. This algorithm only involves one‐dimensional nonparametric smoothers, thereby avoiding the data sparsity problem caused by high model dimensionality. Some simulation studies are conducted to investigate the finite sample performances of the proposed estimators.     

The Relation Between Therapeutic Alliance and Treatment Motivation in Residential Youth Care: A Cross-Lagged Panel Analysis

This study investigated the longitudinal relation between therapeutic alliance and treatment motivation in a sample of 174 adolescents receiving residential treatment in the Netherlands. Structural equation modeling with a cross-lagged panel design was used to examine the relation between therapeutic alliance and treatment motivation up to 9 months of treatment. Results revealed that autoregressive associations between initial therapeutic alliance and alliance at subsequent time points were significant, whereas for treatment motivation a significant association was found after 6 months, but not after 9 months. Results also showed that a higher level of therapeutic alliance after 3 months was predictive of a higher level of treatment motivation after 6 months. Furthermore, a higher level of therapeutic alliance after 6 months was predictive of a higher level of treatment motivation after 9 months. Implications for research and clinical practice are discussed.     

A Narrative Approach to Pregnancy‐related Discrimination and Leader–follower Relationships

This study sheds light on women's perceptions of pregnancy‐related discrimination committed by the discriminated woman's immediate leader. The purpose is to show how women describe this perception during pregnancy and after their return to working life after maternity leave. Further, the study explores the ways in which the quality of the leader–follower relationship is revealed during the narration process. The research draws on five individual stories, collected in two rounds of interviews. Representations of several kinds of discriminatory practices can be found in the narratives and reveal that even being a partner in a good‐quality leader–follower relationship does not necessarily protect a woman from pregnancy‐related discrimination. However, women involved in a high‐quality leader–follower relationship described their perceptions in a more positive manner than those in a low‐quality relationship. The contribution of the article to the literature is twofold: firstly, the results challenge the traditional view that high‐quality leader–member exchange relationships lead to positive outcomes. Secondly, examining the different ways in which people describe their perceptions of discrimination deepens our understanding of the equality problems that working women face.     

Disruption in Parental Co-habitation and its Effects on Short-Term,Medium-Term,and Long-Term Outcomes of Adolescents

In this paper the relationships between a disruption in parental co-habitation and various categories of adolescent outcomes over multiple time horizons are explored. Using data from the National Longitudinal Study of Adolescent Health (Add Health), we estimated the effects of a change from living with both parents to just one, on academic and employment outcomes, the likelihood to indulge in risky behaviors, mental health outcomes and body mass index measures, from less than 1 year to over 14 years after the change. Propensity score matching methods were used to control for individual characteristics and pre-existing differences in the family environment that may increase the chances of separation, and the results are compared to those obtained using ordinary least squares or probit methods. Results showed evidence of adverse effects of living with one parent in the short term, medium term and long term. Adolescents living with one parent had lower academic achievement in all term lengths, poor mental health in the short to medium term, and were more likely to engage in risky behaviors in the medium to long term.     

The FastHCS algorithm for robust PCA

Principal component analysis (PCA) is widely used to analyze high-dimensional data, but it is very sensitive to outliers. Robust PCA methods seek fits that are unaffected by the outliers and can therefore be trusted to reveal them. FastHCS (high-dimensional congruent subsets) is a robust PCA algorithm suitable for high-dimensional applications, including cases where the number of variables exceeds the number of observations. After detailing the FastHCS algorithm, we carry out an extensive simulation study and three real data applications, the results of which show that FastHCS is systematically more robust to outliers than state-of-the-art methods.