Measurement Invariance Across Administration Mode

Examining the Posttraumatic Stress Disorder (PTSD) Checklist

Published in: Psychological Assessment, 2016

Posted on RAND.org on April 20, 2016

by Ashley Boal, Christine Anne Vaughan, Carra S. Sims, Jeremy N. V. Miles

Read More

Access further information on this document at Psychological Assessment

This article was published outside of RAND. The full text of the article can be found at the link above.

The Posttraumatic Stress Disorder (PTSD) Checklist (PCL) is commonly used to screen for PTSD in clinical and research contexts. While the PCL is utilized within numerous settings and populations, research has not yet established the extent to which individuals respond similarly across different modes of administration. The use of both telephone and web survey administration modes has numerous potential benefits, including data quality improvement, but may introduce an additional source of measurement error. The current study examined the psychometric properties, including factor structure and measurement invariance, of the PCL across telephone and web administration modes among 455 wounded, ill, or injured airmen who were medically retired or undergoing evaluation for disability caused by injuries and illnesses of a physical or psychological nature. Findings suggest the properties of the PCL were invariant with regard to the mode of administration, such that the overall scale structure and size of the loadings were similar across groups. Corrections were applied to the computation of probable PTSD diagnosis to account for partial scalar invariance. The lack of complete invariance did not affect probable PTSD diagnosis. Finally, differences in latent means across the telephone and web group were nonsignificant and modest in magnitude. These results indicate that although the PCL only achieved partial scalar invariance across administration modes, the practical impact of this difference on rates of probable PTSD is negligible. The practical benefits of administering the PCL over the telephone and on the web do not appear to be outweighed by the potential cost of additional measurement error. (PsycINFO Database Record (c) 2016 APA, all rights reserved)

This report is part of the RAND Corporation external publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.