1. Linking Life Histories and Mental Health: A Person-Centered Strategy

Abstract

1We present a strategy for using longitudinal survey data to identify life history pathways linked with mental health outcomes. The central aim is to begin with richly detailed descriptions of individual lives and, from them, to discern generalizable features of aggregates of multiple lives. Conceptual principles guiding the organization and interpretation of life history information are summarized. Data from the Wisconsin Longitudinal Study (WLS) are used to illustrate the specific steps for analyzing life histories of "resilient" women (those with a history of depression who report high levels of current wellbeing). The steps begin with writing narratives of individual life histories, which are then reviewed for commonalities, and subsequently thinned to more generic descriptions. The process culminates with tests of distinguishability, contrasting the "resilient" with three other mental health groups. Illustrating the constructive tensions between idiographic and nomothetic analyses, our approach documents multiple life pathways to resilience. The methodology also underscores the delicate interplay between activities of the mind and machine in facilitating scientific discovery.

Diverse pathways to positive and negative affect in adulthood and later life: An integrative approach using recursive partitioning.

“…As such, the methodology facilitates working in the middle territory between strictly nomothetic and idiographic approaches (see Singer et al, 1998), a challenge that has frequently eluded much social scientific and health inquiry. Such investigations thus go beyond the study of averages, both in reported levels of well-being and in what predicts them, and as such, also call for refinement in guiding theoretical frameworks.…”

Zip code-level risk factors for tuberculosis: neighborhood environment and residential segregation in New Jersey, 1985-1992

“…Boolean methods have been used in political science and sociology 29,30 and more recently in health research to study mental health outcomes (i.e., depression) [31][32][33] and child development outcomes. 34 Applications to health services research also have been suggested (C. C. Ragin, PhD, unpublished data, 1999).…”

“…34 Applications to health services research also have been suggested (C. C. Ragin, PhD, unpublished data, 1999). The methodology used in this article follows that proposed by Singer et al, 33 who used it to examine health survey data from the Wisconsin Longitudinal Study.…”

“…The key distinction between regression analyses and Boolean analyses is that the former are appropriate for answering variableoriented research questions, whereas the latter are appropriate for answering case-oriented research questions 33 (also C. C. Ragin, PhD, unpublished data, 1999). Consider a data matrix in which columns correspond to variables and rows correspond to cases.…”

“…Regression analyses focus on the columns-that is, on the average effect of a given variable on the outcome variable across all cases (holding the rest of the analytic variables constant). Boolean analyses focus on the rows-that is, on the combinations of analytic variables that characterize cases that share a common outcome 33 (also C. C. Ragin, PhD, unpublished data, 1999).…”