Adjustment with three continuous variables

Correlation is not causation. Spurious association between X and Y may be due to a confounding variable W. Statisticians may adjust for W using a variety of techniques. This article presents the results of simulations conducted to assess the performance of these techniques under various, elementary, data-generating processes. The results indicate that no technique is best overall and that specific techniques should be selected based on the particulars of the data-generating process. Here, we show how causal graphs can guide the selection or design of techniques for statistical adjustment. R programs are provided for researchers interested in generalization.     

WEIGHING THE MILITARY OPTION: THE EFFECTS OF WARTIME CONDITIONS ON INVESTMENTS IN HUMAN CAPITAL

Serving in the military is an important vehicle through which young Americans invest in their human capital. As such, changes in the desirability of military service may affect the attainment of enlistment requirements, such as a high school degree or equivalent. Using American Community Survey data, we find that exposure to home‐state combat fatalities during the Iraq and Afghanistan Wars decreased the probability of high school completion, and increased the probability of general equivalency diploma completion. Using military data, we confirm that exposure to home‐state fatalities selectively deterred some individuals from enlisting. The results suggest military service and educational investments are complements. (JEL I20, I26, J24)     

Case-based methods and agent-based modelling: bridging the divide to leverage their combined strengths

Two leading camps for studying social complexity are case-based methods (CBM) and agent-based modelling (ABM). Despite the potential epistemological links between ‘cases’ and ‘agents,’ neither camp has leveraged their combined strengths. A bridge can be built, however, by drawing on Abbott’s insight that ‘agents are cases doing things’, Byrne’s suggestion that ‘cases are complex systems with agency’, and by viewing CBM and ABM within the broader trend towards computational modelling of cases. To demonstrate the utility of this bridge, we describe how CBM can utilise ABM to identify case-based trends; explore the interactions and collective behaviour of cases; and study different scenarios. We also describe how ABM can utilise CBM to identify agent types; construct agent behaviour rules; and link these to outcomes to calibrate and validate model results. To further demonstrate the bridge, we review a public health study that made initial steps in combining CBM and ABM.     

An organizational model to distinguish between and integrate research and evaluation activities in a theory based evaluation

Theory-based evaluation (TBE) is an evaluation method that shows how a program will work under certain conditions and has been supported as a viable, evidence-based option in cases where randomized trials or high-quality quasi-experiments are not feasible. Despite the model's widely accepted theoretical appeal there are few examples of its well-implemented use, probably due to time and money limitations necessary for planning and a confusion over the definitions between research and evaluation functions and roles. In this paper, we describe the development of a theory-based evaluation design in a Math and Science Partnership (MSP) research project funded by the U.S. National Science Foundation (NSF). Through this work we developed an organizational model distinguishing between and integrating evaluation and research functions, explicating personnel roles and responsibilities, and highlighting connections between research and evaluation work. Although the research and evaluation components operated on independent budgeting, staffing, and implementation activities, we were able to combine datasets across activities to allow us to assess the integrity of the program theory, not just the hypothesized connections within it. This model has since been used for proposal development and has been invaluable as it creates a research and evaluation plan that is seamless from the beginning.     

Confidence in preferences

Indeterminate preferences have long been a tricky subject for choice theory. One reason for which preferences may be less than fully determinate is the lack of confidence in one’s preferences. In this paper, a representation of confidence in preferences is proposed. It is used to develop and axiomatise an account of the role of confidence in choice which rests on the following intuition: the more important the decision to be taken, the more confidence is required in the preferences needed to take it. This theory provides a natural account of when an agent should defer a decision; namely, when the importance of the decision exceeds his confidence in the relevant preferences. Possible applications of the notion of confidence in preferences to social choice are briefly explored.     

Emptying the Nest: Older Men in the United States, 1880–2000

Between 1880 and 2000, the percentage of married men 60 and older living only with their wives in empty nest households rose from 19 percent to 78 percent. Data drawn from the US census show that more than half of this transformation occurred in the 30‐year period from 1940 to 1970, bookended by moderate increases between 1880 and 1940 and very modest increases after 1970. Two literatures have presented demographic, cultural, and economic explanations for the decline in elderly co‐residence with their children, but none adequately accounts for a sharp change in the mid‐twentieth century. Both aggregate comparisons and multivariate analysis of factors influencing the living arrangements of elderly men suggest that economic advances for all age groups in the critical 30‐year period, along with trends in fertility and immigration, best explain the three‐stage shift that made the empty nest the dominant household form for older men by the beginning of the twenty‐first century.     

A Longitudinal Study of the Developmental Trajectories of Parental Attachment and Career Maturity of South Korean Adolescents

This study tested the developmental trajectories of career maturity (CM) and parental attachment (PA), the longitudinal influence of both, and gender as a moderator. Findings showed developmental progressions in adolescents’ PA and CM over 4 years. The change in PA was positively related to the developmental change in CM. For gender, there was a statistically significant difference in the initial status of CM and PA but no significant difference in developmental change. This study verified the longitudinal influence of PA on CM and showed that the cross‐sectional effect of predictors on CM needs to be distinguished from the longitudinal effect in future studies.     

Using process evaluation to determine effectiveness of participatory ergonomics training interventions in construction

The construction industry continues to experience high rates of musculoskeletal injuries despite the widespread promotion of ergonomic solutions. Participatory ergonomics (PE) has been suggested as one approach to engage workers and employers for reducing physical exposures from work tasks but a systematic review of participatory ergonomics programs showed inconclusive results.. A process evaluation is used to monitor and document the implementation of a program and can aid in understanding the relationship between the program elements and the program outcomes. The purpose of this project is to describe a proposed process evaluation for use in a participatory ergonomic training program in construction workers and to evaluate its utility in a demonstration project among floor layers.     

Masculine Status,Sexual Performance,and the Sexual Stigmatization of Women

Collegiate hookup culture advances ideas of masculinity but contradicts notions of appropriate feminine sexuality. Drawing on focus group and interview data with college students, I examine how a group of class‐ and race‐privileged fraternity men face dilemmas as they enact a group constructed masculinity focused on sexual performance and the objectification of women. I employ a symbolic interactionist framework to illustrate how men, attentive to peer status yet anxious about the sexual stigmatization of women, draw on cultural ideas about appropriate feminine sexuality as they account for their approaches to sex and women (both with whom they interact sexually and how) along a range of intimacy—from hookups to committed relationships. I demonstrate that heterosexual interaction does not unequivocally link to masculine status and that men sometimes strive to limit the impact of casual sex or avoid it altogether.