Effect of Gender Composition on Group Performance

The effect of gender composition on group performance was investigated over three years using the MARKSTRAT 2 simulation. Senior undergraduates formed groups and competed within five‐firm (group) industries over 10 weeks, as well as submitting group reports on their performances. Performance and report marks were independent of group size and individual performance in other course assessment. Women per group (numbers, percentage, gender ratio) was positively related to simulation performance, with few women‐dominated groups ranking lower than first or second within each industry. Very few all‐men groups achieved first or second rankings. There were no significant correlations between group gender compositions and marks for the group written report. Women's more interactive, people‐oriented and co‐operative work styles apparently facilitate the MARKSTRAT 2 group process and, when combined with men's more analytical decision‐making tendencies and competitive orientation, appear to explain the superior performances of mixed groups. The emergence and utilization of these tendencies within groups are discussed in terms of Social Contact Theory and Competition Theory. Although generalizations from student‐based studies to the workplace is problematic, the results indicate that groups may be more effective when women outnumber or equal men, especially in complex management activities requiring extensive information management and processing, planning and decision‐making over protracted periods.     

Who Lives Near Coke Plants and Oil Refineries? An Exploration of the Environmental Inequity Hypothesis

Facility-specific information on pollution was obtained for 36 coke plants and 46 oil refineries in the United States and matched with information on populations surrounding these 82 facilities. These data were analyzed to determine whether environmental inequities were present, whether they were more economic or racial in nature, and whether the racial composition of nearby communities has changed significantly since plants began operations.The Census tracts near coke plants have a disproportionate share of poor and nonwhite residents. Multivariate analyses suggest that existing inequities are primarily economic in nature. The findings for oil refineries are not strongly supportive of the environmental inequity hypothesis. Rank ordering of facilities by race, poverty, and pollution produces limited (although not consistent) evidence that the more risky facilities tend to be operating in communities with above-median proportions of nonwhite residents (near coke plants) and Hispanic residents (near oil refineries). Over time, the racial makeup of many communities near facilities has changed significantly, particularly in the case of coke plants sited in the early 1900s. Further risk-oriented studies of multiple manufacturing facilities in various industrial sectors of the economy are recommended.The authors are all affiliated with the Harvard Center for Risk Analysis     

Using a mixture model for multiple imputation in the presence of outliers: the 'Healthy for life' project

Summary.  We consider the problem of obtaining population-based inference in the presence of missing data and outliers in the context of estimating the prevalence of obesity and body mass index measures from the 'Healthy for life' study. Identifying multiple outliers in a multivariate setting is problematic because of problems such as masking, in which groups of outliers inflate the covariance matrix in a fashion that prevents their identification when included, and swamping, in which outliers skew covariances in a fashion that makes non-outlying observations appear to be outliers. We develop a latent class model that assumes that each observation belongs to one of K unobserved latent classes, with each latent class having a distinct covariance matrix. We consider the latent class covariance matrix with the largest determinant to form an 'outlier class'. By separating the covariance matrix for the outliers from the covariance matrices for the remainder of the data, we avoid the problems of masking and swamping. As did Ghosh-Dastidar and Schafer, we use a multiple-imputation approach, which allows us simultaneously to conduct inference after removing cases that appear to be outliers and to promulgate uncertainty in the outlier status through the model inference. We extend the work of Ghosh-Dastidar and Schafer by embedding the outlier class in a larger mixture model, consider penalized likelihood and posterior predictive distributions to assess model choice and model fit, and develop the model in a fashion to account for the complex sample design. We also consider the repeated sampling properties of the multiple imputation removal of outliers.     

Observing different orders of risk aversion

A decision maker's attitude towards risk is said to be of orderi, i=1, 2, if for every given riskē with expected value zero, the risk premium the decision maker is willing to pay to avoid the risktē goes witht to zero at the same order ast ⁱ. This article presents an experiment testing the order of decision makers' attitudes toward risk. Its major result is that both attitudes exist, each in significant proportions. Moreover, two classes of first-order behavior are defined. The rank-dependent model (Quiggin, 1982) belongs to one, the disappointment aversion model (Gul, 1991) to the other. We show that only the first of these two classes appears among our subjects.     

Validating Benefit and Cost Estimates: The Case of Airbag Regulation

Preregulation estimates of benefits and costs are rarely validated after regulations are implemented. This article performs such a validation for the mandatory automobile airbag requirement. We found that the original 1984 model used to estimate benefits became invalid when 1997 values were input into that 1984 model. However, using a published 1997 cost-effectiveness model, we demonstrate, by replacing the model inputs with the values from 1984, that the 1997 cost-effectiveness ratios, based on real-world crash data and tear-down cost data, are less attractive than what would have been originally anticipated. The three most important errors in the 1984 input values are identified: the overestimation of airbag effectiveness, the overestimation of baseline fatality/injury rates, and the underestimation of manual safety belt use. This case study, which suggests that airbags are a reasonable investment in safety, shows that the regulatory analysis tools do not always produce findings that are biased against health, safety, and environmental regulation. Future validation studies of health, safety, and environmental regulation should focus on validation of benefit and risk estimates, areas where we found significant error, as well as on cost estimates.     

The complexity of roles in community action projects: The example of the evaluation of ‘alternatives’

Evaluation of community interventions is a special form of evaluation involving many roles that are key to accomplishing both the intervention and the accompanying research. The present paper describes the roles involved in the evaluation of a community intervention (‘Alternatives’) aimed at safer use of alcohol and depressant medications by seniors.The three core roles were the community developer who provided leadership in the implementation of the project, the researcher/evaluator, and the community committee that sanctioned and facilitated the project. Other roles on the project included: the project coordinator, the nurse-interviewers who collected the survey data for the evaluation, volunteers from seniors groups, local health service providers, the media, the funders, and others on the project team who provided specific technical expertise.The paper describes the competing interests among these roles as well as conflicts that arose and how these conflicts were handled. Although specific roles are described, the types of roles and the potential for competing interests and conflict are generally relevant to community level evaluations.     

Estimation for hidden Markov random fields

A finite state Markov random field is noisily observed via a second finite state process. The parameters of the model are estimated, as well as the most likely signal given the observations.     

Social Isolation and Lethal Violence Across the Metro/Nonmetro Divide: The Effects of Socioeconomic Disadvantage and Poverty Concentration on Homicide*

Abstract This study extends the macro‐level criminological research tradition by examining the links between socioeconomic disadvantage, poverty concentration, and homicide in metropolitan and nonmetropolitan U.S. counties. Most research in this tradition has tested structural theories using urban areas as the unit of analysis. This “urban bias” has resulted in a limited understanding of the social forces driving violence in nonmetropolitan areas. To partially address this problem, we link the literature on the spatial and social organization of nonmetropolitan communities with the social isolation perspective from the urban poverty literature. We hypothesize that the spatial concentration of poverty drives up rates of homicide in both metropolitan and nonmetropolitan areas regardless of levels of socioeconomic disadvantage. Negative binomial regression for 1,746 nonmetropolitan and 778 metropolitan counties suggest that both socioeconomic disadvantage and poverty concentration elevate homicide in metropolitan areas. However, in nonmetropolitan counties only socioeconomic disadvantage has a significant impact. We conclude by discussing the implications of these differential findings for the social isolation perspective.