Computerized testing in a hospital setting: Psychometric and psychological effects

摘要

The aim of the studies reported here was to evaluate a touch-screen computer-based (CB) test administration system in a ‘real-life’, hospital setting, by comparing it with an ordinary paper-and-pencil (PP) administration routine. A heterogeneoys set of tests were selected, including a mood adjective checklist, the Beck Depression Inventory (BDI), and a performance intelligence test. A total of 57 patients were randomly allocated to two groups. CB and PP versions of the tests were administered to the groups at different occasions, and in different orders for the two groups. Mean test results were comparable for the two versions, with a tendency for the BDI to yield higher scores with the CB version. A post hoc analysis was made using the BDI data: two subsets of items were compared, one with neutral and the other with ‘sensitive’ content. The ‘sensitive’ subset showed significantly higher mean scores and fewer ‘no problem’ answers with the CB than with the PP version.