Open AccessConsistency and factorial invariance of the Davidson trauma scale in heterogeneous populations: results from the 2010 Chilean earthquake
Author(s) -
Kevan Bryan
Publication year - 2017
Publication title -
international journal of methods in psychiatric research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.275
H-Index - 73
eISSN - 1557-0657
pISSN - 1049-8931
DOI - 10.1002/mpr.1516
Subject(s) - scale (ratio) , psychology , confirmatory factor analysis , sample (material) , reliability (semiconductor) , vulnerability (computing) , measurement invariance , clinical psychology , concurrent validity , population , external validity , consistency (knowledge bases) , statistics , sample size determination , econometrics , construct validity , psychometrics , internal consistency , social psychology , structural equation modeling , mathematics , computer science , geography , medicine , cartography , environmental health , artificial intelligence , power (physics) , computer security , chemistry , chromatography , quantum mechanics , physics
This investigation seeks to validate an application of a standardized post‐traumatic stress symptom self‐report survey, the Davidson Trauma Scale (DTS), with a large, heterogeneous population of earthquake victims. While previous studies have focused primarily on small samples, this investigation uses a unique dataset to assess the validity of this application of the DTS while accounting for heterogeneity and sample size. We use concurrent validity and reliability analysis tests to confirm the validity of the scale. Further, confirmatory factor analysis is used to test the fit of the data's factor structure against previously established trauma models. Finally, these fit tests are repeated across different mutually exclusive vulnerability subsets of the data in order to investigate how the invariance of the scale is affected by sample heterogeneity. We find that this particular application of the scale is, on the whole, reliable and valid, showing good concurrent validity. However, evidence of variability is found across specific vulnerability subsets, indicating that a heterogeneous sample can have a measurable impact on model fit. © 2016 The Authors International Journal of Methods in Psychiatric Research Published by John Wiley & Sons Ltd