Abstract

The methodological issue of the effect of modalities on clinical ratings was examined using audio recordings, verbatim transcripts, and both modalities combined. The California Q-Sort , the Experiencing Scale , and a Global Affective Rating scale , were used to examine the interviews. ;The data consisted of audio recordings from initial interviews with thirteen parents whose children had been diagnosed with cancer. The audio recordings were transcribed into verbatim transcripts and divided into halves at a clinically determined point. Ratings were then performed by a total of eighteen raters over the three modalities. Two raters were assigned to each half of an interview in each modality with a total of six raters being assigned for each half of an interview, resulting in a total of 156 possible ratings. ;Three questions were addressed: Are there differences in clinical ratings, when raters use audio recordings versus verbatim transcripts versus both modalities combined? Are there differences in the ratings between the first and second half of an interview? Is there a modality by half of an interview interaction effect? ;A review of the literature revealed that the majority of studies that had been performed attempting to determine which modality produces more accurate results had resulted in no definitive conclusion. Many reseachers had encountered difficulties as a result of technical problems with the data and inadequate planning for the data collection at the beginning of the research. ;An analysis of variance produced results indicating that there were neither differences among the three modalities nor between the first and second halves of the interviews, neither was there any interaction effect between modalities and halves. The three scales produced very similar ratings between the modalities and the halves of interviews. Even the Q-Sort, with its many items, produced similar ratings. ;While these results indicate that it is appropriate to use either of these three modalities in studying clinical data, further study should be undertaken in order to obtain conclusive results