Family and sociodemographic influences on patterns of leaving home in Postwar Britain

We identify child-level and parent-level characteristics associated with children’s patterns of leaving home. We use a multilevel discrete-time hazards model to examine the impact of family and demographic factors at both levels, and utilize the Alternating Conditional Expectation algorithm optimally to transform the dependent and independent variables. We find that measured variables at both the child and the parent level have important influences, as do period and cohort factors. However, unmeasured parent-level factors have an influence on the departure of children that is broadly similar in magnitude to measured factors.     

Optimal Bayesian Randomization

Randomization is a puzzle for Bayesians. The intuitive need for randomization is clear, but there is a standard result that Bayesians need not randomize. In this paper we propose a model in which randomization is a strictly optimal procedure. The most important aspect of our model is that there are several parties who make different decisions and observe different data. The result also sheds light on the ethical considerations involving randomization in a clinical trial.     

Services to prevent child placing-out for cocaine-affected and non-affected families

This study sought to evaluate the effectiveness of the'In-Home Family Care' programme of Children's Home Society of California in Alameda County, particularly regarding the emphasis and impact of services for cocaine-affected families. The family characteristics, problems addressed and service provision in cases served over a 3-year period were assessed in terms of their relation to placement outcome for cocaine-affected and non-affected families. This evaluation found that, despite serious and often chronic problems among cocaine-affected families, their child removal rate was no higher than that for non-affected families. Implications for risk assessment and placement decisions are discussed.     

Child Anger Regulation, Parental Responses to Children's Anger Displays, and Early Child Antisocial Behavior

The display and regulation of child anger in family interaction was coded in a sample of 240 boys and girls at child age 6, and coded using the Specific Affect Coding System. Child antisocial behavior was longitudinally assessed, beginning in kindergarten. Pooled‐ and family‐level analyses were used to assess hazard rates for child anger. Parents’ ability to modulate their own emotions and negative behavior, and children's ability to down‐regulate anger were associated with increased latency for child anger. Hazard for child anger increased as parents’ insensitive and negative responses toward the child cumulated during family interaction. Macro‐level, non‐hazard analyses indicated that chronic levels of child antisocial behavior were associated with the frequency of parental negative behavior, but not with the frequency of child anger. Micro‐level hazard analyses indicated that children's ability to regulate anger was related to chronic levels of child covert but not overt antisocial behavior.     

The importance of preservation of the ethical principle of equipoise in the design of clinical trials: relative impact of the methodological quality domains on the treatment effect in randomized controlled trials

Previous research has identified methodological problems in the design and conduct of randomized trials that could, if left unaddressed, lead to biased results. In this report we discuss one such problem, inadequate control intervention, and argue that it can be by far the most important design characteristic of randomized trials in overestimating the effect of new treatments. Current guidelines for the design and reporting of randomized trials, such as the Consolidated Standards of Reporting Trials (CONSORT) statement, do not address the choice of the comparator intervention. We argue that an adequate control intervention can be selected if people designing a trial explicitly take into consideration the ethical principle of equipoise, also known as "the uncertainty principle."