Graduation Year


Document Type




Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department


Major Professor

Michael T. Brannick, Ph.D.

Co-Major Professor

Walter C. Borman, Ph.D.

Committee Member

Stephen Stark, Ph.D.

Committee Member

Vicky Phares, Ph.D.

Committee Member

Yi-hsin Chen, Ph.D.


head-to-head comparison, latent profile analysis, person-centered approach, resilience, variable-centered approach


Resilience research has become increasingly popular in organizational sciences in recent years. Different factor structures of resilience have been proposed and yet no consensus has been reached regarding its underlying dimensions. Such a variable-centered perspective of studying resilience may be well complemented by a typological approach, which may shed fresh light on the nature of resilience. The current study took a person-centered approach with the advantage of using latent profile analysis to explore a set of latent profiles of resilience. Two studies were conducted. In Study 1 (N = 479), archival undergraduate data were used to explore resilience profiles and their relationships with personality variables (i.e., the Big Five and trait affect) and related outcomes (i.e., interpersonal counterproductive work behavior, job satisfaction, and life satisfaction). In Study 2 (N = 483), an employee sample collected on Amazon’s Mechanical Turk (MTurk) were used to confirm the set of identified latent profiles and extend Study 1 by including two additional outcome variables (job burnout and stress). Finally, this study compared the person- and variable-centered approaches to examine whether latent profiles of resilience provided incremental validity over dimensions of resilience in predicting outcomes in both studies. Four latent profiles of resilience emerged in the undergraduate sample and showed differential relationships with personality variables and outcomes; three out of four profiles were replicated in the MTurk sample. Profile membership provided incremental validity beyond dimensions of resilience, the Big Five, and trait affect in the MTurk sample but not the undergraduate sample. These mixed findings, strengths and limitations of the study, and implications for theory and practice were discussed.