Suchergebnisse

Examining the major approaches to hypothesis testing and model selection, this book blends statistical theory with recommendations for practice, illustrated with real-world social science examples. It systematically compares classical (frequentist) and Bayesian approaches, showing how they are applied, exploring ways to reconcile the differences between them, and evaluating key controversies and criticisms. The book also addresses the role of hypothesis testing in the evaluation of theories, the relationship between hypothesis tests and confidence intervals, and the role of prior knowledge in

Research has found that education is associated with more liberal opinions on a number of issues, but there is uncertainty about the scope & interpretation of these findings. This study investigates the effects of education using data from the 1990 World Values Survey, which includes 40 nations & covers a wide range of opinions. Multilevel models are used to allow for national variation in the effects of education, including interactions with economic development & Communist rule. Education influences most political opinions; the effects can be summarized by saying that it promotes individualist values. Education is associated with somewhat lower confidence in most institutions. The relationship between education & confidence becomes weaker with economic development, while the effects of education on a number of political views shift to the left with economic development. The results suggest that education deserves more attention in explanations of national differences & historical trends in opinion. 2 Tables, 1 Figure, 32 References. Adapted from the source document.

Debate surrounding the alleged decline of the influence of social class on voting behavior across the late-19th & 20th centuries is reviewed, & new evidence is presented to argue that class position continues to affect voting choices. Using individual-level data from the 1972-1994 General Social Surveys, class voting in the US is analyzed across three key sociodemographic factors: race, religious affiliation, & gender. The influence of other factors such as ideology, values, public opinion, birth cohort, material satisfaction, & party affiliation is also considered. Results indicate that such non-class-based attitudes & identities may override class divisions in voting; however, they do not do so consistently, & the influence of class sometimes coexists with these factors. 3 Tables, 2 Figures, 43 References. K. Hyatt Stewart

Debate surrounding the alleged decline of the influence of social class on voting behavior across the late-19th & 20th centuries is reviewed, & new evidence is presented to argue that class position continues to affect voting choices. Using individual-level data from the 1972-1994 General Social Surveys, class voting in the US is analyzed across three key sociodemographic factors: race, religious affiliation, & gender. The influence of other factors such as ideology, values, public opinion, birth cohort, material satisfaction, & party affiliation is also considered. Results indicate that such non-class-based attitudes & identities may override class divisions in voting; however, they do not do so consistently, & the influence of class sometimes coexists with these factors. 3 Tables, 2 Figures, 43 References. K. Hyatt Stewart

The Bayesian information criterion (BIC) has become a popular criterion for model selection in recent years. The BIC is intended to provide a measure of the weight of evidence favoring one model over another, or Bayes factor. It has, however, some important drawbacks that are not widely recognized. First, Bayes factors depend on prior beliefs about the expected distribution of parameter values, and there is no guarantee that the Bayes factor implied by the BIC will be close to one calculated from a prior distribution that an observer would actually regard as appropriate. Second, to obtain the Bayes factors that follow from the BIC, investigators would have to vary their prior distributions depending on the marginal distributions of the variables and the nature of the hypothesis. Such variations seem unwarranted in principle and tend to make the BIC inclined to favor excessively simple models in practice. These points are illustrated by the analysis of several examples, and alternatives to use of the BIC are discussed.