Bayesian inference for dynamic social network data

We consider a continuous-time model for the evolution of social networks. A social network is here conceived as a (di-) graph on a set of vertices, representing actors, and the changes of interest are creation and disappearance over time of (arcs) edges in the graph. Hence we model a collection of random edge indicators that are not, in general, independent. We explicitly model the interdependencies between edge indicators that arise from interaction between social entities. A Markov chain is defined in terms of an embedded chain with holding times and transition probabilities. Data are observed at fixed points in time and hence we are not able to observe the embedded chain directly. Introducing a prior distribution for the parameters we may implement an MCMC algorithm for exploring the posterior distribution of the parameters by simulating the evolution of the embedded process between observations.     

NOTE ON POINT AND INTERVAL ESTIMATION AND PREDICTION FOR NORMAL EQUICORRELATED DATA

For normal homoscedastic equicorrelated data the parameters μ, σ and p are identifiable, but it is shown that no consistent estimators exist. Confidence intervals for μ and prediction intervals with known confidence coefficients can, however, be obtained.     

Intercultural qualitative research and Ph.D. students

The educational environment for postgraduate health professionals from developing countries in contemporary western universities is an intermediate zone between home and host culture. In this zone, knowledge is shaped through the development of concepts within the limitations of (often) pre‐fluent language capacity. It is characterized by the mutual exchange of knowledge between western teachers and their developing‐country students, who will return home to leadership roles working with marginalized individuals, communities and populations. This qualitative research aims to improve health education for global health professionals from developing countries, and in so doing it illustrates many of the difficulties and benefits involved in intercultural qualitative research for Ph.D. students.     

The value of the second income to two-earner families with children

An increasing number and percentage of families in the United States endeavor to improve family income by placing two adults in the paid work force. This study examines the impact of the second earner on family income for 1,624 married-couple families with children, including 871 one-earner and 753 two-earner families. After-tax income is regressed on one- or two-earner status as well as covariates which confound the income-earner relationship for families of various income levels. The actual after-tax income differential of $7,172 is reduced to $6,076 in the regression analysis. Further, income for each family type is estimated by applying the regression coefficients from one-earner households to the characteristics of two-earner households and the regression coefficients from two-earner households to one-earner household characteristics. Actual incomes for two-earner families are found to be higher than those of one-earner families. However, income differences are reduced from 34% higher actual before-tax income to an estimated 14% lower after-tax income. The findings have important implications for families selecting two earners solely for the purpose of increasing after-tax income.This study received funding from the Graduate School Fund at Northern Illinois University. An earlier version of this paper was presented at the Family Economics/Home Management Conference, Indianapolis, IN, June 1987.Mary Pritchard is an Assistant Professor, Department of Human and Family Resources, Northern Illinois University, DeKalb, IL 60115. She received her Ph.D. from Purdue University and includes family economic well-being, family income and expenditures, and spending patterns of adolescents among her research interests.     

WHO'S PUBLISHING IN JMFT? AN EXAMINATION OF AUTHOR CHARACTERISTICS

This article examines the gender, setting, and degree of the first authors of published articles in JMFT from 1990 to 1995. Findings reveal that men are the first authors of 67% of the articles and women, 33%. PhDs are first authors on 58% of the articles, and authors in university settings write 53% of the published articles. Thirty-nine percent of the articles published during this period were research; 26%, clinical; 16%, theory; and 14%, training. Surprising findings were that MSW authors had a higher acceptance rate than PhDs, and authors in training institutes had a higher acceptance rate than those in university settings.     

Engaging the Process of Transformation

The contributions included in this issue demonstrate the variety of ways in which participatory and interpretive methods are transforming the field of psychology and clarify the integration of the personal and political commitments of the authors within their research. A number of issues remain to be negotiated, however, before these methods take their full place in the field. This conclusion considers a few of these issues, including an examination of how these approaches demand a fundamental reassessment of validity in psychological research. It also considers the critical issue of the ownership of knowledge, and proposes a reframing of the ethical guidelines for research in the field. Finally, it invites further discussion and debate around the issues raised in this issue.     

The Evolution of Organizations: Suggestions from Complexity Theory About the Interplay Between Natural Selection and Adaptation

There has been much debate in the managementliterature between neo-Darwinists (who believe in thenatural selection of populations of organizations) andadaptationists (who contend that changes in organization structure and behavior occur in response to theenvironment). The general thesis of neo-Darwinism isthat species are blindly selected for survival by theenvironment. The latest empirical support for the dominant neo-Darwinism perspective adopted bymost biologists is based primarily on the experimentsconducted by Salvador Luria who claims to haveconclusively demonstrated that genes mutate randomly.Recently, however, biologists have re-examined Luria sresearch methods and, after replications of hisexperiments, now question some aspects of the validityof his results. Moreover, there is now new researchwhich provides support for the earlier adaptationistposition, namely, the existence of evolutionary driversand directors existing within self-organizing systems.Of particular importance to the present study is the experimental indication thatself-organizing systems play a conscious role in theirown evolution. We propose that similar mechanisms orprocesses operate in organizational adaptation, thuspointing toward a theoretical modification ofneo-Darwinism that embraces both adaptation and naturalselection in a general, unified theory.     

Changes in Income Concentration: Taxes or Macroeconomic Conditions?

We examine the relative importance of tax rates and macroeconomic fluctuations in explaining the share of national adjusted gross income (AGI) reported by the top 0.5% of all taxpayers. Results indicate that cutting the top income or capital gains tax rate would increase the top AGI share but not by enough to increase revenues. The preponderance of evidence suggests that the top AGI share is affected more by the capital gains tax rate than by the income tax rate but that real gross domestic product fluctuations have even larger effects.     

Estimation of regression models with equi-correlated responses when some observations on the response variable are missing

The present article deals with the problem of estimation of parameters in a linear regression model when some data on response variable is missing and the responses are equi-correlated. The ordinary least squares and optimal homogeneous predictors are employed to find the imputed values of missing observations. Their efficiency properties are analyzed using the small disturbances asymptotic theory. The estimation of regression coefficients using these imputed values is also considered and a comparison of estimators is presented.