Construct Validity and Factor Structure of the Posttraumatic Growth Inventory (PTGI) among Korean Americans

Purpose: To examine the psychometric properties of the Korean version of Posttraumatic Growth Inventory (PTGI-K) including factor structure of the PTGI-K and potential construct bias in the PTGI-K across gender. PTGI is designed to measure positive changes emerging from traumatic experiences have been described as an enhancement of an individual’s ability to cope with adversity.

Design: This study employed a purposive and non-probability sampling strategy, and a total of 310 Korean Americans from the Korean community in New York and New Jersey completed surveys.  Data were collected between July 2014 and November 2014.  

Method: Participants completed paper questionnaires consisting of socio-demographic inventory, PTGI-K, Trauma history questions. Exploratory, confirmatory, and multiple-group confirmatory factor analyses were used for data analyses. To ensure that the exploratory and confirmatory factor analyses were conducted with distinct sets of participants, the total sample of participants (N=310) was randomly divided into two equal subsamples (n1=155, n2=155).  T-test and Chi-square analyses were used to determine no differences existed between the subsamples. 

Findings: Results suggested that a three-factor structure of the PTGI-K was the best fit for the data and the new structure of the PTGI-K was variant across gender.  The most frequent traumatic event was reported to be the death of a loved one (26.3%).  Other frequent lifetime traumatic events were acute/chronic diseases (14.9%), financial problems (12.5%), or sudden accidents (10.7%). The sample consisted of 60.3% of women and 39.7 % men and the mean age was 54.8 years (SD = 16.24), ranging between 19 and 93.  KA men and women showed very similar socio-demographic profiles, except PTGI-K scores and marital status.  As compared with their KA men counterparts, KA women were more likely to have higher PTGI-K scores (t = -3.55, p < .001) and less likely to be married or cohabited (χ² = 9.16, p < .01).  

Conclusion: This study supports the use of PTGI-K in future research for measuring positive growth as a result of traumatic or highly stressful events PTGI among KAs.  Study findings also imply the ways in which KAs experience trauma and growth is different from their American counterparts. The new factor structure of the PTGI-K can help mental health professionals and researchers better understand how KAs express their psychological changes after traumatic events.  Further research is needed to investigate the psychometric structure of the PTGI-K with a large sample of data both cross-sectionally and longitudinally.