OBJECTIVE: The purpose of this study was to examine whether computer administration of the Symptom Check List (SCL-90-R) is equivalent to paper-and-pencil originals. METHOD:282 psychosomatic outpatients were randomly assigned to computer or paper-and-pencil conditions. Statistical equivalence tests were used to examine psychometric equivalence for the means. Reliabilities and correlations were compared for the two methods of administration. RESULTS: No systematic differences were observed in group means for most of the subscales. Subjects of the computer-administered group scored higher on the SCL-90-R subscale 'Obsessive-Compulsive' and 'Anger-Hostility' than the control subjects. Gender and administration mode interaction was observed for one subscale, while age and administration interaction was observed for another subscale. CONCLUSION: Using computer-administered tests makes administration and scoring of tests more efficient. The differences between the two administration modes were small, although noticeable. Further research is needed to determine whether computer environment, computer experience and age may influence the test results.
RCT Entities:
OBJECTIVE: The purpose of this study was to examine whether computer administration of the Symptom Check List (SCL-90-R) is equivalent to paper-and-pencil originals. METHOD: 282 psychosomatic outpatients were randomly assigned to computer or paper-and-pencil conditions. Statistical equivalence tests were used to examine psychometric equivalence for the means. Reliabilities and correlations were compared for the two methods of administration. RESULTS: No systematic differences were observed in group means for most of the subscales. Subjects of the computer-administered group scored higher on the SCL-90-R subscale 'Obsessive-Compulsive' and 'Anger-Hostility' than the control subjects. Gender and administration mode interaction was observed for one subscale, while age and administration interaction was observed for another subscale. CONCLUSION: Using computer-administered tests makes administration and scoring of tests more efficient. The differences between the two administration modes were small, although noticeable. Further research is needed to determine whether computer environment, computer experience and age may influence the test results.
Authors: Elise Braekman; Finaba Berete; Rana Charafeddine; Stefaan Demarest; Sabine Drieskens; Lydia Gisle; Geert Molenberghs; Jean Tafforeau; Johan Van der Heyden; Guido Van Hal Journal: PLoS One Date: 2018-05-21 Impact factor: 3.240