Date of Award

Spring 2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Engineering and Technology

Committee Director

Dr. Pilar Pazos-Lago

Committee Member

Dr. Charles Daniels

Committee Member

Dr. David Walters

Abstract

The Job Demands–Resources (JD-R) model is a widely used framework for understanding how work conditions influence employee attitudes and retention-related outcomes. However, empirical evidence testing JD-R dynamics in classified technical work environments remains limited, particularly using longitudinal designs that account for structural constraints, nonrandom participation, and outcome-specific measurement challenges. Classified work is characterized by persistent demands, restricted flexibility, and security-driven limitations on available resources, raising questions about whether and how JD-R processes generalize to this context.

This study examines whether job demands and job resources predict subsequent job satisfaction and retention intent in a classified technical workforce using multi-wave longitudinal survey data. To address the unbalanced panel structure and mixed outcome types, the study employs a triangulated analytic strategy that includes balanced-panel cross-lagged structural equation models, unbalanced-panel within-between decomposition models, fixed-effects sensitivity analyses, and outcome-appropriate binary logistic models for retention intent. The study further evaluates panel continuation and selection through baseline diagnostics, continuation modeling, and inverse probability weighting sensitivity analyses.

Results provide convergent evidence that JD-R processes operate meaningfully for job satisfaction in this setting. Higher job demands predict lower subsequent job satisfaction, and higher job resources predict higher subsequent job satisfaction, with the strongest support observed in models that isolate within-person change and leverage the broader unbalanced panel. Joint models demonstrate that resources provide incremental predictive contribution beyond demands, supporting the conceptual distinction between the two constructs even under security-constrained conditions. In contrast, retention intent exhibits a more constrained pattern. Demand exposure is negatively associated with retention intent, most clearly at the between-person level, while job resources do not demonstrate robust predictive contribution for short-run changes in a binary retention intention measure over annual intervals. Limited within-person switching and outcome scale are identified as key constraints on detectability.

Continuation analyses indicate that panel participation is demand-linked and predicted by retention intent, confirming that missingness is not missing completely at random in this dataset. Sensitivity analyses suggest that selection does not overturn the substantive conclusions for job satisfaction when triangulated models are considered. Overall, the findings support the applicability of the JD-R framework in classified technical work while highlighting outcome specific dynamics, the dominant role of demands as a strain-related signal, and the importance of measurement sensitivity and triangulation for credible longitudinal inference in security constrained organizational contexts.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

https://doi.org/10.25777/9t1q-3380

ISBN

9798197809322

ORCID

0009-0004-1247-020X

Download

Share

COinS