NotesFAQContact Us
Search Tips
ERIC Number: ED552085
Record Type: Non-Journal
Publication Date: 2012
Pages: 249
Abstractor: As Provided
Reference Count: N/A
ISBN: 978-1-2678-8211-0
Validity of the Consensual Assessment Technique--Evidence with Three Groups of Judges and an Elementary School Student Sample
Long, Haiying
ProQuest LLC, Ph.D. Dissertation, Indiana University
As one of the most widely used creativity assessment tools, the Consensual Assessment Technique (CAT) has been praised as a valid tool to assess creativity. In Amabile's (1982) seminal work, the inter-rater reliability was defined as construct validity of the CAT. During the past three decades, researchers followed this definition and regarded inter-rater reliability that is mainly estimated by Cronbach's alpha as validity evidence of the CAT. However, Cronbach's alpha is not an appropriate approach to estimate inter-rater reliability. Furthermore, reliability evidence can hardly support validity. Therefore, even though there has been abundant evidence of reliability for the CAT, little validity evidence has been provided in the field. To further complicate the matter, the studies that examined validity of the CAT only employed the old notion of validity. The current research used Messick's (1989a, 1995) unified view of validity as conceptual framework and investigated four aspects of validity of the CAT, content, substantive, external, and generalizability, with three groups of judges and an elementary school student sample. Responses to two science tasks from 24 students were collected from two sixth grade classes and were later evaluated on a 1-5 scale by three different groups of judges (i.e., educational researcher, elementary school teachers, and undergraduate students) using their own criteria of creativity. Judges were required to explain their ratings and to answer a few questions regarding their rating criteria and procedures at the end of the survey. Those who evaluated students' responses differently were asked follow-up questions and conducted semi-structured interviews. Mixed methods were employed in the study. Quantitative data were analyzed by repeated measures analysis of variance, correlation, multiple linear regression, and generalizability and dependability studies. Qualitative data were analyzed by framing analysis. The study found very little evidence to support the validity of the inferences made based on the CAT scores. This result was further discussed with the proposed purpose of the CAT and the assumptions of measurement. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page:]
ProQuest LLC. 789 East Eisenhower Parkway, P.O. Box 1346, Ann Arbor, MI 48106. Tel: 800-521-0600; Web site:
Publication Type: Dissertations/Theses - Doctoral Dissertations
Education Level: Elementary Education; Higher Education; Postsecondary Education
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A