Estimation across multiple models with application to Bayesian computing and software development

Statistical models are sometimes incorporated into computer software for making predictions about future observations. When the computer model consists of a single statistical model this corresponds to estimation of a function of the model parameters. This paper is concerned with the case that the computer model implements multiple, individually-estimated statistical sub-models. This case frequently arises, for example, in models for medical decision making that derive parameter information from multiple clinical studies. We develop a method for calculating the posterior mean of a function of the parameter vectors of multiple statistical models that is easy to implement in computer software, has high asymptotic accuracy, and has a computational cost linear in the total number of model parameters. The formula is then used to derive a general result about posterior estimation across multiple models. The utility of the results is illustrated by application to clinical software that estimates the risk of fatal coronary disease in people with diabetes.     

Populism and the Rise of the Reform Party in Alberta*

L'augmentation rapide du soutien populaire accordé au Reform Party of Canada est abordée du point de vue de deux théories déjà anciennes de mobilisation populiste-la première axée sur la position de classe et la seconde privilégiant le nativisme. Les données issues d'une enquête effectuêe dans toute la province en 1991 révèlent que les agriculteurs, un segment particulier de la petite bourgeoisie, ont tendance plus que tout autre groupe à accorder leur vote au Reform Party au cours d'une élection fédérate. Les Albertains qui se décrivent comme «Canadiens» plutôt qu'appartenant à un groupe ethnique ou racial particulier et, à un degré moindre, les répondants d'origine anglo-saxonne ou celtique, semblent un peu plus enclins à soutenir le Reform Party. De plus, les Albertains qui ont de fortes convictions sur certains dossiers sociaux et politiques (multicultu-ralisme, égalité des sexes, statut distinct du Québec, indulgence envers les criminels) sont également attirés par le Reform Party, ainsi que les personnes qui se sentent généralement aliénées sur le plan politique. The early rise in popular support for the Reform Party of Canada is discussed from the vantage point of two long-standing theories of populist mobilization, the first focussing on class position and the second emphasizing nativism. Data from a province-wide survey of Alberta residents conducted in 1991 reveal that farmers, a particular segment of the petite bourgeoisie, were more likely than any other class group to state that they would vote Reform in a federal election. Albertans who identified themselves as Canadians rather than identifying with a particular ethnic or racial group, and to a lesser extent those of Anglo-Saxon/Celtic origin, were somewhat more likely to support the Reform Party. In addition, Albertans with strong feelings on a number of specific social and political issues (multiculturalism, gender equality, distinct status for Quebec, lenient treatment of criminals) were also attracted to the Reform Party, as were those who felt generally politically alienated.     

Union decertification election outcomes: Bargaining unit characteristics and union resources

Previous analysis using aggregate data has concluded that union decertification activity is “exclusively a product of market conditions.” We employ data disaggregated by local (county) labor market which permits a preliminary investigation of the importance of potential nonmarket influences that are not measurable using aggregate data, namely, the type of bargaining unit, the type of union, the type of employer, and proxies for union resources. It also allows a more precise specification of the labor market characteristics previously found to influence decertification. Our results confirm the importance of market conditions in decertification activity and outcomes, but they also suggest that the likelihood of decertification is much greater in independent unions and in industry-county combinations in which the typical employee works part-time and has limited alternative income opportunities. Institutional characteristics of the union and employer, individual characteristics of bargaining unit members, and local economic conditions are important determinants of decertification.     

Popular Moralities and Institutional Rationalities in Malawi's Struggle Against AIDS

The AIDS epidemic in sub‐Saharan Africa is fertile ground for examining how moral evaluations evolve over time and across different settings. We compare the discourse on AIDS in Malawi as presented in the media with that of everyday conversations. Drawing on two sets of texts, newspaper articles and conversational journals, produced over a ten‐year period from 1999 through 2008, we analyze their moral injunctions, or what individuals should or should not do in response to the AIDS epidemic. The predominant injunctions in the early years in both sources were premarital abstinence and marital fidelity. Following the introduction of HIV testing and treatment, however, the discourses diverged. Discourse in the media turned toward moral injunctions requiring individuals to interact with authoritative institutions; thus the conceptual center of gravity of the struggle against AIDS in the media shifted from collectivities of individuals to collectivities of institutions. Rural Malawians, however, received these moral injunctions with great skepticism and continued to advise each other to prevent death by controlling one's sexual appetites.     

Developmental Idealism and Cultural Models of the Family in Malawi

This paper examines the extent to which developmental idealism has been disseminated in Malawi. Developmental idealism is a set of beliefs and values about development and the relationships between development and family structures and behavior. Developmental idealism states that attributes of societies and families defined as modern are better than attributes defined as traditional, that modern societies help produce modern families, that modern families facilitate the achievement of modern societies, and that the future will bring family change in the direction of modernity. Previous research has demonstrated that knowledge of developmental idealism is widespread in many places around the world, but provides little systematic data about it in sub-Saharan Africa or how knowledge of it is associated with certain demographic characteristics in that region. In this paper, we address this issue by examining whether ordinary people in two settings in Malawi, a sub-Saharan African country, have received and understood messages that are intended to associate development with certain types of family forms and family behaviors. We then examine associations between demographic characteristics and developmental idealism to investigate possible mechanisms linking global discourse about development to the grassroots. We analyze data collected in face-to-face surveys from two samples of Malawian men in 2009 and 2010, one rural, the other in a low-to-medium income neighborhood of a city. Our analysis of these survey data shows considerable evidence that many developmental idealism beliefs have been spread in that country and that education has positive effects on beliefs in the association between development and family attributes. We also find higher levels of developmental idealism awareness in the urban sample than we do in the rural sample, but once dissimilarities in education and wealth between the two samples are controlled, awareness levels no longer differed between urban and rural respondents. We explore how these beliefs intersect with longstanding local values and beliefs in Malawi.     

The impact of incomplete geocoding on small area population estimates

Small-area population estimates are often made using geocoded address data in conjunction with the housing-unit method. Previous research, however, suggests that these data are subject to systematic incompleteness that biases estimates of race, ethnicity, and other important demographic characteristics. This incompleteness is driven largely by an inability to complete georeference address-based datasets. Given these challenges, small-area demographers need further, and to date largely unavailable, information on the amount of error typically introduced by using incompletely geocoded data to estimate population. More specifically, we argue that applied demographers should like to know if these errors are statistically significant, spatially patterned, or systematically related to specific population characteristics. This paper evaluates the impact of incomplete geocoding on accuracy in small-area population estimates, using a Vintage 2000 set of block-group estimates of the household population for the Albuquerque, NM metro area. Precise estimates of the impact of incomplete geocoding on the accuracy of estimates are made, associations with specific demographic characteristics are considered, and a simple potential remediation based on Horvitz-Thompson theory is presented. The implications of these results for the practice of applied demography are reviewed.     

School Progress Among Children of Same-Sex Couples

This study uses logit regressions on a pooled sample of children from the 2012, 2013, and 2014 American Community Survey to perform a nationally representative analysis of school progress for a large sample of 4,430 children who reside with same-sex couples. Odds ratios from regressions that compare children between different-sex married couples and same-sex couples fail to show significant differences in normal school progress between households across a variety of sample compositions. Likewise, marginal effects from regressions that compare children with similar family dynamics between different-sex married couples and same-sex couples fail to predict significantly higher probabilities of grade retention for children of same-sex couples. Significantly lower grade retention rates are sometimes predicted for children of same-sex couples than for different-sex married couples, but these differences are sensitive to sample exclusions and do not indicate causal benefits to same-sex parenting.