Self-report, peer-report, and teacher-report measures of childhood depression : and analysis by item /
| Main Author: | |
|---|---|
| Other Authors: | , , |
| Format: | Thesis Book |
| Language: | English |
| Published: |
1991.
|
| Subjects: | |
| Online Access: | Link to OAKTrust copy |
| Abstract: | Childhood depression is a phenomenon that is of concern because of its debilitating effects. Multiple instruments have been developed with which to measure this construct. These instruments can be grouped into three categories, self-report, peer-report, and other-report, based on who responds. Studies investigating the degree of correspondence between the different types of measures have found the correlations to be low to moderate. The reason for the low correlations among instruments allegedly measuring the same construct has remained unclear and was the focus of this study. This research sought to investigate the relationship among childhood depression measures at the item level. The Children's Depression Inventory (CDI) was selected as a self-report measure, the Peer Nomination Inventory of Depression (PNID) was the peer-report measure used, and the Child Behavior Checklist-Teacher Report Form (CBCL) was the other-report measure employed. Using these representative instruments, several research questions were posed. First, the relationship among the summary scores of these instruments, the factor structure underlying all the items from the three instruments, and the magnitude and direction of the relationship between items was investigated. Secondly, item pairs from the instruments which were judged to ask "essentially the same question" were identified. The relationship between these items was then explored. Finally, items were identified that could best predict the summary scores of the three instruments when taken as a group. In a single testing session, 275 5th-9th grade children were administered the CDI and PNID. The CBCL was completed by the science teacher. The major conclusion that can be drawn from this study is that the three measures of depression are not assessing the same thing at a primary level. Yet, the results also suggest that an omnibus construct, depression, is being measured at least to some degree by all three measures used. Another important finding is that the lack of item similarity among the different measures cannot account for the low correlations among the summary scores. The similar item correlations were also in the low to moderate range. |
|---|---|
| Item Description: | Typescript (photocopy). Vita. "Major subject: Counseling Psychology." |
| Physical Description: | x, 179 leaves ; 29 cm |
| Bibliography: | Includes bibliographical references. |