Calibrated imputation in surveys under a quasi-model-assisted approach

Summary.  We propose to use calibrated imputation to compensate for missing values. This technique consists of finding final imputed values that are as close as possible to preliminary imputed values and are calibrated to satisfy constraints. Preliminary imputed values, potentially justified by an imputation model, are obtained through deterministic single imputation. Using appropriate constraints, the resulting imputed estimator is asymptotically unbiased for estimation of linear population parameters such as domain totals. A quasi-model-assisted approach is considered in the sense that inferences do not depend on the validity of an imputation model and are made with respect to the sampling design and a non-response model. An imputation model may still be used to generate imputed values and thus to improve the efficiency of the imputed estimator. This approach has the characteristic of handling naturally the situation where more than one imputation method is used owing to missing values in the variables that are used to obtain imputed values. We use the Taylor linearization technique to obtain a variance estimator under a general non-response model. For the logistic non-response model, we show that ignoring the effect of estimating the non-response model parameters leads to overestimating the variance of the imputed estimator. In practice, the overestimation is expected to be moderate or even negligible, as shown in a simulation study.     

The Development of Parent–Child Relationship Perceptions in Boys from Childhood to Adolescence: A Comparison Between Disruptive and Non-Disruptive Boys

The absence or weakness of the father role is considered by many authors as one of the principal explaining factors of disruptive behavior and other psychopathologies among children. However, empirical studies on the specific role of the father are rare. Boys with stable disruptive behavior from age 5 until age 15 were compared to boys who have never been disruptive during the same time period. The perceptions that boys have regarding their relationships with their fathers from childhood to adolescence was compared to, perceptions concerning their relationships with their mothers. The results indicate that boys from both groups generally feel less loved and less appreciated by their fathers than by their mothers in adolescence but not in childhood. The boys perceived that their two parents love and appreciate them a lot less at age 15, than at age 9 and age 11. They felt less loved and appreciated by their fathers than their mothers. At age 15, disruptive boys considered they were less loved by their two parents than non-disruptive boys did. The differences in the perceptions by the boys concerning the quality of relationships with their parents appear during adolescence.     

Long-Term Services and Supports for Older Adults: A Review of Home and Community-Based Services Versus Institutional Care

Despite a shift from institutional services toward more home and community-based services (HCBS) for older adults who need long-term services and supports (LTSS), the effects of HCBS have yet to be adequately synthesized in the literature. This review of literature from 1995 to 2012 compares the outcome trajectories of older adults served through HCBS (including assisted living [AL]) and in nursing homes (NHs) for physical function, cognition, mental health, mortality, use of acute care, and associated harms (e.g., accidents, abuse, and neglect) and costs. NH and AL residents did not differ in physical function, cognition, mental health, and mortality outcomes. The differences in harms between HCBS recipients and NH residents were mixed. Evidence was insufficient for cost comparisons. More and better research is needed to draw robust conclusions about how the service setting influences the outcomes and costs of LTSS for older adults. Future research should address the numerous methodological challenges present in this field of research and should emphasize studies evaluating the effectiveness of HCBS.     

A nonparametric procedure for the analysis of balanced crossover designs

The authors propose nonparametric tests for the hypothesis of no direct treatment effects, as well as for the hypothesis of no carryover effects, for balanced crossover designs in which the number of treatments equals the number of periods p, where p ≥ 3. They suppose that the design consists of n replications of balanced crossover designs, each formed by m Latin squares of order p. Their tests are permutation tests which are based on the n vectors of least squares estimators of the parameters of interest obtained from the n replications of the experiment. They obtain both the exact and limiting distribution of the test statistics, and they show that the tests have, asymptotically, the same power as the F‐ratio test.     

Probabilistic voting and platform selection in multi-party elections

The literature on stochastic voting to date has focused almost exclusively on models with only two candidates (or parties). This paper studies multiparty competition with stochastic voting. We look at two different models in which candidates aim to maximize their expected vote, as well as a model where the objective of candidates is rank minimization. The equilibria of these models are derived and characterized. We show that the properties of the equilibria are quite different from those derived in deterministic models. Furthermore, the analysis shows that deterministic voting models are not robust since the introduction of even a minute level of uncertainty leads to a drastic change in predictions. Consequently, we argue that the deterministic model provides a misleading benchmark. Stochastic models provide a much richer framework, and the nature of the uncertainty in voter choice is a key determinant of the qualtitative properties of the equilibria.The authors thank Martin Osborne and Maurice Salles for useful comments and suggestions. The first author would also like to thank the Bankard Fund for financial support.     

Testing independence with additional information

Let X = (X^j : j = 1,…, n) be n row vectors of dimension p independently and identically distributed multinomial. For each j, X^j is partitioned as X^j = (X^j₁, X^j₂, X^j₃), where p_i is the dimension of X^j_i with p₁ = 1,p₁+p₂+p₃ = p. In addition, consider vectors Y^j_i, i = 1,2j = 1,…,n_i that are independent and distributed as X¹_i. We treat here the problem of testing independence between X¹₁ and X¹₃ knowing that X¹₁ and X¹₂ are uncorrected. A locally best invariant test is proposed for this problem.