ERIC Number: ED052138
Record Type: RIE
Publication Date: 1971
Reference Count: 0
Empirical Validation of a New Category System: One Example.
Rosenshine, Barak; And Others
This study found that data from previous research can be used to validate a new observational category system and that subscripting of the original ten categories of the Flanders Interaction Analysis System is useful in identifying more specific behaviors which correlate with student achievement. The new category system was the Expanded Interaction Analysis System (EIAS), in which each of the 10 Flanders categories is subscripted into two to four smaller categories. The material used for validation was the audiotapes, typescripts, and residual student achievement scores used in the original study by Wright and Nuthall (1970). In this study, 17 teachers presented three 10-minute lessons on the black-billed gull to intact classes of third grade students. Results from the EIAS coding and analysis showed the major advantage of subscripting to be in categories 3 and 4. In category 3 (teacher accepts students' ideas), the entire category had a correlation of .15 with achievement, whereas the subscript 3a (acknowledges ideas by simple reflection) yielded a correlation of .30. In category 4, although the total frequency of questions yielded a correlation of .18, the asking of factual questions had a correlation of .55 with student achievement. This was considered to be due to the factual nature of the material. This technique of using the same data with different systems is recommended for future use. (RT)
Publication Type: N/A
Education Level: N/A
Authoring Institution: Temple Univ., Philadelphia, PA.
Identifiers: Flanders System of Interaction Analysis