Synthetic data method to incorporate external information into a current study

We consider the situation where there is a known regression model that can be used to predict an outcome, Y, from a set of predictor variables X . A new variable B is expected to enhance the prediction of Y. A dataset of size n containing Y, X and B is available, and the challenge is to build an improved model for Y| X ,B that uses both the available individual level data and some summary information obtained from the known model for Y| X . We propose a synthetic data approach, which consists of creating m additional synthetic data observations, and then analyzing the combined dataset of size n + m to estimate the parameters of the Y| X ,B model. This combined dataset of size n + m now has missing values of B for m of the observations, and is analyzed using methods that can handle missing data (e.g., multiple imputation). We present simulation studies and illustrate the method using data from the Prostate Cancer Prevention Trial. Though the synthetic data method is applicable to a general regression context, to provide some justification, we show in two special cases that the asymptotic variances of the parameter estimates in the Y| X ,B model are identical to those from an alternative constrained maximum likelihood estimation approach. This correspondence in special cases and the method's broad applicability makes it appealing for use across diverse scenarios. The Canadian Journal of Statistics 47: 580–603; 2019 © 2019 Statistical Society of Canada     

DISSECTING THE MONTY HALL ANOMALY

We assess competing explanations of irrational behavior in the Monty Hall problem by creating new variants of the problem. Some variants employ a feature that automates the merging of probabilities, thus rendering transparent the probabilistic advantage of the rational choice. That feature also enables systematic variation in informational asymmetry, and in ordering of actions. Data from 77 subjects, each of whom makes 30 binary decisions, indicate that automated merging raises the fraction of rational choices from around 40% to over 80%. Other features examined have much less impact, indicating the importance of a Bayesian updating failure. (JEL C91, D02, D81, D83)     

Cognition in Situations

The symbolic interactionist tradition can contribute to advancing sociological studies of cognition by setting dual process models on more solid ground. I draw on Blumer's epistemological statements and the interactionist tradition more broadly to consider how dual process models of cognition could be applied to naturally occurring situations. I suggest that attending to the ways the past and the future are handled and modified within social interaction provides a usable inroad for the sociology of cognition to engage with situational analysis. I identify “resonance” and “iterative reprocessing” as concepts that are suitable to this end.     

Estimation of the Intraclass Correlation Coefficient for the Analysis of Covariance Model

For the unbalanced analysis of covariance model with one covariate, a simple formula is given for the intraclass correlation coefficient estimator that results from Henderson's Method 3 estimation of variance components. Example calculations and the corresponding interpretations are given for a study of the correlation of iron content among brothers. The example illustrates the manner in which the estimator depends on the pattern of correlation between the covariate and the variable under investigation.     

South Africa: Anti-Apartheid NGOs in Transition

Under apartheid, there were an ever-increasing number of nongovernmental organizations (NGOs) concerned about challenging the South African state and furthering a nonracial democratic society. In the 1990s, with the transition to an African National Congress-led democratic government, these organizations underwent profound changes. This article describes the key dynamics of this process, outlines the challenges currently confronting the new NGO sector, and concludes that the prospects for progressive NGO work in dealing with the poor and marginalized are constrained by the prevailing neoliberal economic climate.     

Considering the Risk of Infection by Cryptosporidium via Consumption of Municipally Treated Drinking Water from a Surface Water Source in a Southwestern Ontario Community

Through the use of case‐control analyses and quantitative microbial risk assessment (QMRA), relative risks of transmission of cryptosporidiosis have been evaluated (recreational water exposure vs. drinking water consumption) for a Canadian community with higher than national rates of cryptosporidiosis. A QMRA was developed to assess the risk of Cryptosporidium infection through the consumption of municipally treated drinking water. Simulations were based on site‐specific surface water contamination levels and drinking water treatment log₁₀ reduction capacity for Cryptosporidium. Results suggested that the risk of Cryptosporidium infection via drinking water in the study community, assuming routine operation of the water treatment plant, was negligible (6 infections per 10¹³ persons per day—5th percentile: 2 infections per 10¹⁵ persons per day; 95th percentile: 3 infections per 10¹² persons per day). The risk is essentially nonexistent during optimized, routine treatment operations. The study community achieves between 7 and 9 log₁₀Cryptosporidium oocyst reduction through routine water treatment processes. Although these results do not preclude the need for constant vigilance by both water treatment and public health professionals in this community, they suggest that the cause of higher rates of cryptosporidiosis are more likely due to recreational water contact, or perhaps direct animal contact. QMRA can be successfully applied at the community level to identify data gaps, rank relative public health risks, and forecast future risk scenarios. It is most useful when performed in a collaborative way with local stakeholders, from beginning to end of the risk analysis paradigm.     

Development of Ego‐resiliency: Relations to Observed Parenting and Polymorphisms in the Serotonin Transporter Gene During Early Childhood

We used observed parenting behaviors, along with genetic variants and haplotypes of the serotonin transporter gene (SLC6A4), as predictors of children's ego‐resiliency during early childhood (N = 153). The quality of mothers' parenting was observed at 18 months of age, and mothers' reports of ego‐resiliency were collected at six time points from 18 to 84 months. Genetic data were collected at 72 months. Observed parenting was positively associated with initial levels of children's ego‐resiliency. Furthermore, although individual genetic variants of the serotonin transporter gene (LPR, STin2) were not associated with ego‐resiliency, the S10 haplotype (that combines information from these two variants) was negatively associated with initial levels of ego‐resiliency. Both parenting and serotonin genetic variation uniquely predicted children's ego‐resiliency, suggesting an additive effect of genetic and parental factors.     

Correcting misconceptions about the development of social work in China: a response to Cunfu Jia

In this brief debate article we respond to Cunfu Jia's criticism of our article ‘Defining the Profession?’ (International Journal of Social Work 16: 382–389). We address the issue of ‘evidence’ and the issue of the applicability to China of the global definition of social work. We suggest that although debate articles by their very nature risk generating false dichotomies, they provide an important and welcome opportunity to explore commonality and difference and to develop our knowledge and thinking as a result of an international exchange of views.