Racial diversity, minority concentration, and trust in Canadian urban neighborhoods

Using a sample of 42,329 respondents nested within 4254 Canadian urban neighborhoods, this study demonstrates the conceptual and empirical importance of making a distinction between neighborhood racial diversity and minority concentration, and examines how each is uniquely associated with trust. Our analysis shows that at a given level of racial minority concentration, Whites are more trusting when their minority neighbors are more evenly distributed across racial minority groups. Meanwhile, Whites are less trusting as the neighborhood share of racial minorities increases. Overall, the effect of racial minority concentration tends to prevail over that of racial diversity.     

Deconvolution methods for non-parametric inference in two-level mixed models

Summary.  We develop a general non-parametric approach to the analysis of clustered data via random effects. Assuming only that the link function is known, the regression functions and the distributions of both cluster means and observation errors are treated non-parametrically. Our argument proceeds by viewing the observation error at the cluster mean level as though it were a measurement error in an errors-in-variables problem, and using a deconvolution argument to access the distribution of the cluster mean. A Fourier deconvolution approach could be used if the distribution of the error-in-variables were known. In practice it is unknown, of course, but it can be estimated from repeated measurements, and in this way deconvolution can be achieved in an approximate sense. This argument might be interpreted as implying that large numbers of replicates are necessary for each cluster mean distribution, but that is not so; we avoid this requirement by incorporating statistical smoothing over values of nearby explanatory variables. Empirical rules are developed for the choice of smoothing parameter. Numerical simulations, and an application to real data, demonstrate small sample performance for this package of methodology. We also develop theory establishing statistical consistency.     

Clarifying the Association Between Mother–Father Relationship Aggression and Parenting

Although much research examines the association between fathers' relationship aggression and mothers' parenting, little attention is given to mothers' aggression, mutual aggression, or fathers' parenting. Using a sample of coresiding couples from the Fragile Families and Child Wellbeing Study (N = 973), the authors examine the association between mothers' and fathers' relationship aggression, measured as frequency and perpetration–victimization types (mutual, mother only, father only), and mothers' and fathers' parenting. Fixed effects regression models show that fathers' aggression is positively related to mothers' parenting stress, whereas father‐only or mother‐only aggression is related to fathers' stress. For both parents, aggression perpetration is negatively related to their own engagement with children. Mother‐only aggression is negatively related to mothers' spanking and positively related to fathers' spanking. These findings suggest the importance of examining both parents' aggression and perpetrators' as well as victims' parenting to better understand the link between relationship aggression and parenting.     

Relatives in Residence: Relatedness of Household Members Drives Schooling Differentials in Mozambique

Children typically receive investments from their fathers, but absent fathers often invest at low levels. In a father's absence, what types of nonfathers invest heavily in children? This article investigates educational participation as a reflection of childhood investments on Ibo Island, Mozambique, where only one third of school‐aged children live with their biological fathers. Father‐present children generally attended school at the highest rates. Stepchildren and father‐absent relatives (e.g., grandchildren, nieces) attended school at comparably high rates if any coresiding children were father‐present. This may signal high altruism among present fathers toward some nonoffspring. Consistent with this result, a fixed effects model indicates that, within the same household, adult males invested equally in their own children, relatives, and stepchildren. Prejudicially lower investments were made in children who were unrelated to the household's adult males, however; this result has strong negative implications for the well‐being of African children fostered by nonrelatives.     

Substance‐dependent women becoming mothers: breaking the cycle of adverse childhood experiences

Parenting may be particularly challenging for substance‐dependent mothers who have grown up with parents who themselves had substance use disorders (SUDs). The aim of this study was to explore how substance‐dependent mothers describe their childhood experiences with substance‐abusing parents and the association between these earlier experiences and their own role as caregivers. Using purposeful sampling, mothers admitted for 1 year to a family ward at a substance abuse clinic were approached. Through in‐depth, qualitative interviews, nine substance‐dependent mothers described their lives in the form of present, past and future tense. The findings indicate that substance‐dependent women, who have experienced SUDs in their families of origin, face several major challenges when they become mothers. Some describe having lived their whole lives ‘on the edge of society'. This makes their rehabilitation process more complex. All mothers work to abstain from substances, process traumatic experiences and integrate their family into society. They need help to build supportive social networks and to establish a safe and predictable family environment for themselves and their children. The therapeutic implications of these findings will be discussed.     

A Note on High-Dimensional Linear Regression With Interactions

The problem of interaction selection in high-dimensional data analysis has recently received much attention. This note aims to address and clarify several fundamental issues in interaction selection for linear regression models, especially when the input dimension p is much larger than the sample size n. We first discuss how to give a formal definition of “importance” for main and interaction effects. Then we focus on two-stage methods, which are computationally attractive for high-dimensional data analysis but thus far have been regarded as heuristic. We revisit the counterexample of Turlach and provide new insight to justify two-stage methods from the theoretical perspective. In the end, we suggest new strategies for interaction selection under the marginality principle and provide some simulation results.     

Notes on testing carry-over effects in continuous data under an incomplete block crossover design

We consider hypothesis testing and estimation of carry-over effects in continuous data under an incomplete block crossover design when comparing two experimental treatments with a placebo. We develop procedures for testing differential carry-over effects based on the weighted-least-squares (WLS) method. We apply Monte Carlo simulations to evaluate the performance of these test procedures in a variety of situations. We use the data regarding the forced expiratory volume in one second (FEV₁) readings taken from a double-blind crossover trial comparing two different doses of formoterol with a placebo to illustrate the use of test procedures proposed here.     

Decomposition of the main effects and interaction term by using orthogonal polynomials in multiple non symmetrical correspondence analysis

The multiple non symmetric correspondence analysis (MNSCA) is a useful technique for analyzing a two-way contingency table. In more complex cases, the predictor variables are more than one. In this paper, the MNSCA, along with the decomposition of the Gray–Williams Tau index, in main effects and interaction term, is used to analyze a contingency table with two predictor categorical variables and an ordinal response variable. The Multiple-Tau index is a measure of association that contains both main effects and interaction term. The main effects represent the change in the response variables due to the change in the level/categories of the predictor variables, considering the effects of their addition, while the interaction effect represents the combined effect of predictor categorical variables on the ordinal response variable. Moreover, for ordinal scale variables, we propose a further decomposition in order to check the existence of power components by using Emerson's orthogonal polynomials.     

Beyond interviewer effects in the standardized measurement of ego-centric networks

The effects which interviewers exert on the collection of ego-centric networks have recently come into the focus of methodological considerations. Studies consistently show that the size of networks varies depending on the interviewer. We would like to expand on this research strand by pointing to different aspects which have so far gone unremarked in the discussion. First, size is mainly analysed as a network measure which is influenced during data collection, while other common measures such as network density or composition have not received sufficient consideration. Second, large-scale surveys using face-to-face interviews usually allocate interviewers to a single sampling point. Differences between sampling points (locality effects) are attributed to interviewer effects. Hence, we disentangle the effects of the locality and interviewer. Third, the discussion on interviewer effects often follows an “actor-oriented” consideration of how data collection situations are structured by interviewers. Expanding this approach from a relational perspective, we consider the relationship between the interviewers and respondents and whether this relationship influences the collection of network data. To test our hypotheses about the influence of interviewers, the locality and the interviewer-respondent relationship on different network measures, we use data from the 2010 German General Social Survey (n = 2827 respondents, n = 220 interviewers). The multilevel analyses show that the relationship between the interviewer and the respondent is not very relevant. Furthermore, the analyses show that interviewers have an influence on the network size but not on measures of their composition. However, evidence on the prevalence of locality or interviewer effects is mixed. Finally, homophilous interviewer-respondent relationships have very little effect on network characteristics. We find evidence of training and fatigue effects on network size. However, much of the variation in network size caused by the interviewer still remains unexplained. We draw conclusions on how to organize interview situations in surveys.