Date of Award
2001
Degree Type
Dissertation
First Advisor
Lisa Harlow
Abstract
The social sciences are seeking expanded methodological models that go beyond significance testing for the assessment of theoretical efficacy (Cohen, 1984; Harlow, Muliak & Steiger, 1997; Meehl, 1997). Faust and Meehl (1992) contend that the social sciences are lacking in the integration of data, while Miller and Burbank, (1994) call for models that offer congruency in theory and practice. In response, this dissertation presents an ontological model that embraces the epistemological views of both quantitative and qualitative researchers. It builds on a previous longitudinal study (Webster, 2000) involving outcomes for 32 juvenile delinquents in Alaska. Results from a qualitative thematic content analysis and quantitative data (Achenbach Youth Self-Report (1991); Rosenberg's Self-Esteem Test (1965)) were organized in conceptual matrix displays (Miles & Huberman, 1994). Findings were woven interactively (Newman & Benz, 1998), allowing for confirmation and theoretical synthesis. Discriminate Function Analysis (DFA) and MANOVA were performed to assess the relationships between combined qualitative and quantitative results as they related to sub-samples of youths. Meehl's (1997) corroboration index was applied to assess theoretical efficacy of self-in-relation (Surrey, 1991) and sexual abuse models on later existential well-being (Mennen & Meadow, 1993). Qualitative results indicated that Native Alaskan females, compared with other youths, reported no relational concerns, expressed feelings of powerlessness and social withdrawal, and were less satisfied with treatment. Contrasting collectivist cultural models, Alaskan Native females valued autonomy and self-empowerment. Sexual abuse was reported by 79% of female respondents. Based on lower levels of self-esteem and higher existential distress, DFA classified group membership for abused versus non-abused youths substantially higher than a 50% chance rate of 72%. MANOVA results indicated that Native Alaskan females reported significantly higher levels of anxiety F(3, 26) = 4.59, p = .001; withdrawal, F(3, 26) = 3.52, p = .029; and lower levels of self-esteem F(3, 26) = 5.57, p = .004 than Whites, regardless of abuse status. Though no established norms yet exist, Meehl's corroboration index indicated moderate levels of corroboration for sexual abuse models Ci = .60 and seemingly low levels for self-in-relation theory Ci = .31. Implications are addressed in the context of broadening methodological practices.
Recommended Citation
Cady-Webster, Kathleen S., "Dialogue in the development of a consilience index: A quantum approach to integrating qualitative and quantitative practices in the social sciences" (2001). Open Access Dissertations. Paper 1738.
https://digitalcommons.uri.edu/oa_diss/1738
Terms of Use
All rights reserved under copyright.