Amplification of Sensitivity Analysis in Matched Observational Studies

A sensitivity analysis displays the increase in uncertainty that attends an inference when a key assumption is relaxed. In matched observational studies of treatment effects, a key assumption in some analyses is that subjects matched for observed covariates are comparable, and this assumption is relaxed by positing a relevant covariate that was not observed and not controlled by matching. What properties would such an unobserved covariate need to have to materially alter the inference about treatment effects? For ease of calculation and reporting, it is convenient that the sensitivity analysis be of low dimension, perhaps indexed by a scalar sensitivity parameter, but for interpretation in specific contexts, a higher dimensional analysis may be of greater relevance. An amplification of a sensitivity analysis is defined as a map from each point in a low dimensional sensitivity analysis to a set of points, perhaps a 'curve,' in a higher dimensional sensitivity analysis such that the possible inferences are the same for all points in the set. Possessing an amplification, an investigator may calculate and report the low dimensional analysis, yet have available the interpretations of the higher dimensional analysis.     

Designing a Programme to Support Smallholder Agriculture in Mexico: Lessons from PROCAMPO and Oportunidades

This article argues that there is a gap in current Mexican rural development policy that could be filled by a smallholder‐targeted programme. The gap exists because agricultural programmes tend to benefit better‐off farmers, while social programmes tend to focus on long‐term investment in human capital, specifically for the children of today's poor. Drawing on lessons from PROCAMPO and Oportunidades, it offers a specific design for such a programme at a time when the government is considering how to reform PROCAMPO, given that, after 15 years, it is due to come to an end.     

Immediate and Longer-Term Connections Between Support and Stress in Pregnant/Parenting and Non-Pregnant/Non-Parenting Adolescents

Despite the substantial amount of literature focusing on social support for pregnant and parenting adolescents, few studies have directly examined the relationships among stress and social support across their transition to parenting. The present study investigates the nature of the relationship between stress and support both before and after the birth of the baby. Two groups of adolescent females (one group facing parenthood and one not) completed measures of support and stress across a 7-month period (N = 231) to coincide with the transition to parenthood. In general, support and stress were negatively related when measured concurrently and positively related over time for parenting adolescents but not the non-parenting group. Understanding how long social support impacts stress will enable practitioners to identify when best to provide interventions for pregnant and parenting adolescents.     

On evaluating interactive statistical program packages

Most statistical computing for data analysis has come to depend upon statistical program packages. In recent years, interactive computing has become widespread both on large time-sharing systems and on mini-computers. This paper identifies ways in which interactive statistical software packages differ from batchoriented software and discusses evaluation considerations pertaining specifically to interactive packages.     

Confounding in mixed factorial experiments through isomorphisms

The derivation of a simpie mexhoa for confounding in mixed factorial experiments from an isomorphism of finite abelian groups is presented. The theoretical bases of confounding procedures that use modular arithmetic for such experiments are compared.     

A Critical Look at Some Analyses of Major League Baseball Salaries

At a data analysis exposition sponsored by the Section on Statistical Graphics of the ASA in 1988, 15 groups of statisticians analyzed the same data about salaries of major league baseball players. By examining what they did, what worked, and what failed, we can begin to learn about the relative strengths and weaknesses of different approaches to analyzing data. The data are rich in difficulties. They require reexpression, contain errors and outliers, and exhibit nonlinear relationships. They thus pose a realistic challenge to the variety of data analysis techniques used. The analysis groups chose a wide range of model-fitting methods, including regression, principal components, factor analysis, time series, and CART. We thus have an effective framework for comparing these approaches so that we can learn more about them. Our examination shows that approaches commonly identified with Exploratory Data Analysis are substantially more effective at revealing the underlying patterns in the data and at building parsimonious, understandable models that fit the data well. We also find that common data displays, when applied carefully, are often sufficient for even complex analyses such as this.     

RELIGIOUS REFERENCE GROUPS AND THE PERSISTENCE OF NORMATIVE BEHAVIOR: AN EMPIRICAL TEST

Reference groups and significant others are vitally important in both the formation and the persistence or change of normative as well as deviant behavior patterns. Thus one's initial religious beliefs and behavior (or lack thereof) reflect the socializing influence of the family. However, the situation may change when young people leave home for education or work, as demonstrated by research that shows decreases in religious beliefs or church attendance when young people leave home to attend college. In contrast to the pattern whereby religiosity declines in a college or university environment, we maintain that students who develop close ties with others who are religious, especially in a highly religious community, will maintain the same patterns of high commitment developed in their families. Specifically, we hypothesize that religious beliefs and participation will be positively related to (1) parents' religious beliefs and practices and (2) current friends' religious beliefs and participation. These hypotheses were tested with a sample of college students living on campus (n = 339). The data support the argument that students' current religious beliefs and behavior are related to both their parents' religiosity and the reinforcing effects of the religiosity of their current friends.