|
Abstract
|
Individuals displaced due to humanitarian disasters—currently 70.8 million [1]—suffer from high rates of mental health problems. Although rates of depression and posttraumatic stress disorder (PTSD) in these populations range widely [2, 3], it is estimated that at least a quarter of all survivors suffer from either depression or PTSD [4]. In order to correctly identify those displaced individuals who are experiencing clinically significant mental distress, it is imperative that psychosocial professionals have measurement tools that are valid and reliable for the populations with which they are working.
In the early years of humanitarian psychosocial interventions (i.e., the 1980s and 1990s), it was common practice when assessing mental health in disaster- and conflict-affected populations to limit psychometric assessment of measurement tools to simply reporting Cronbach’s alpha—a necessary but hardly sufficient practice when working with novel populations. If an adequate alpha were reported, the validity and reliability of the measure were typically assumed across populations. A few measurement tools were developed and validated for specific forcedly displaced populations (e.g., the Harvard Trauma Questionnaire for Cambodian, Laotian, and Vietnamese refugees [5]); however, this led to other researchers using these measures and claiming that they had been validated for refugee populations in general. In a critical review of 183 studies almost 20 years ago, Hollifield and colleagues [6] found that the majority of research articles published on the mental health of refugee and related populations to that point included data from instruments whose scores had limited or untested validity and reliability in refugee populations. Furthermore, half reviewed did not report any statistical properties of the instruments used, and only 12 articles used instruments that had been developed specifically for displaced populations. In a commentary 5 years later Bass, Bolton, and Murray [7] noted that there had been few attempts at that time to develop instruments whose scores were valid and reliable for use in communities affected by humanitarian disaster. Thankfully, in the past decade, there has been a growing realization among those working in psychosocial intervention that the reliability and validity of measurement tools used in populations other than those the measures were originally developed for must be demonstrated, not simply assumed ...
Language: en |