NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED161958
Record Type: RIE
Publication Date: 1978-Jun
Pages: 50
Abstractor: N/A
Reference Count: 0
ISBN: N/A
ISSN: N/A
Psychometrics, Mathematical Psychology, and Cognition: Confessions of a Closet Psychometrician. Technical Report No. 10.
Sternberg, Robert J.
The importance of validity and reliability in the componential analysis of human intelligence is discussed. The roles of construct, internal, external, convergent, discriminant, and ecological validity in componential analysis are outlined. Within-replication (internal-consistency) and between-replication (alternate-forms or test-retest) reliability are also described and illustrated with examples from the author's research on analogical reasoning. The psychometric and multivariate techniques used in componential analysis of intelligence are described by the author, including linear multiple regression, nonlinear multiple regression, canonical regression, principle-axis analysis, confirmatory maximum-likelihood factor analysis, nonmetric multidimensional scaling, and additive clustering. Following a discussion of these procedures, the author describes the evolution of psychometrics and cognitive psychology as separate disciplines. The author contends that cognitive psychologists and psychometricians would benefit from an active interchange, and that psychometricians have technical skills to teach the cognitive psychologist. The first step in an interchange would be to convince the psychologist that these skills are worth learning. Psychometrists can accomplish this by showing cognitive psychologists that psychometrics is responsive to the needs of cognitive researchers. (Author/JAC)
Publication Type: Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: Office of Naval Research, Washington, DC. Psychological Sciences Div.
Authoring Institution: Yale Univ., New Haven, CT. Dept. of Psychology.
Identifiers: N/A
Note: Paper presented at the Annual Meeting of The Psychometric Society (Hamilton, Ontario, Canada, August, 1978)