Psychology Ph.D. Dissertations

Improving Construct Validity and Measurement of Post-Traumatic Growth

Date of Award


Document Type


Degree Name

Doctor of Philosophy (Ph.D.)



First Advisor

Eric Dubow (Advisor)

Second Advisor

William Morrison (Other)

Third Advisor

Carolyn Tompsett (Committee Member)

Fourth Advisor

Dara Musher-Eizenman (Committee Member)


Across studies, Post-Traumatic Growth is often positively correlated with unhelpful psychological processes (e.g., rumination) and post-traumatic stress (PTS). I hypothesized that these counterintuitive findings are due to inadequate measurement models of growth. Research has suggested there could be two types of growth: “illusory” (typically considered “cognitive growth”) and “real” (i.e., actualized growth, which could be considered “behavioral” growth). I had two aims in this dissertation: First, to create and validate a new measure of behavioral post-traumatic growth parallel to an existing measure of cognitive growth (Post-Traumatic Growth Inventory; PTGI). Second, I assessed relations between each type of growth and common trauma and coping-related precursors (i.e., time since trauma, severity of the trauma, adaptive coping, and rumination) and also adjustment variables (i.e., post-traumatic stress, anxiety/depression, and satisfaction with life). I hypothesized that if behavioral growth is the more “actualized” type of growth, that it would have stronger positive correlations with trauma/coping variables compared to cognitive growth, and that behavioral growth would moderate, or weaken, the expected negative relation between cognitive growth and adjustment variables. Method. 11 clinicians were surveyed: qualitative theme analysis informed item creation for the new behavioral growth measure in the same domains as the cognitive PTGI: relating to others, new possibilities, personal strength, spiritual growth, and appreciation of life. 199 US adults were surveyed from Amazon Mechanical Turk to test hypotheses. Results. The behavioral growth measure was adequately validated with Confirmatory Factor Analysis. The hypothesis that trauma and coping-related variables would have stronger positive correlations with behavioral growth than cognitive growth was mostly unsupported. Additionally, behavioral growth did not moderate the relation between cognitive growth and adjustment. However, several quadratic curvilinear relations were identified between trauma/coping variables and adjustment variables with each type of growth. For example: low and high rumination related to lower cognitive growth, while moderate rumination related to higher growth; low and high cognitive growth related to lower PTS, while moderate growth related to higher PTS. Discussion. Researchers should continue exploring alternative measurement models of post-traumatic growth and non-linear relations between variables to address counterintuitive findings and further improve construct validity of post-traumatic growth.