ERIC Number: ED206712
Record Type: Non-Journal
Publication Date: 1979-Sep
Reference Count: N/A
The Effect of Using More than One Type of Teacher Observation System in the Same Study.
Marston, Paul T.; Zimmerer, Linda K.
The multiple use of several classroom observation instruments was investigated to determine its influence upon the way an observer records behavior, since such an influence may cause discrepancies in findings between studies. Twenty-four volunteer students were divided into three groups and trained to use one or more of three classroom observation instruments: the first group was trained to use the low-inference (requires little observer judgment) Flanders System of Interaction Analysis; group 2 was trained to use the high-inference (requiring high observer judgment) Classroom Observation Record and Classroom Observation Scales; group 3 was trained in all three. Observers then matched and coded videotapes of classroom action in a fifth grade social studies session. Experiments were then conducted to test the influences of high and low-inference coding techniques upon one another. Analysis showed an influence both ways, with coders bringing their high and low-inference codings into consistency with one another; the effect of the low-inference on the high-inference being strongest. Examination of the specific items changed indicated that overlap in content between the systems probably accounts for most changes. The results pointed to a possible source of bias in process-product studies that use multiple observation systems to maximize data collection. (Author/AEF)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: National Inst. of Education (DHEW), Washington, DC.
Authoring Institution: Texas Univ., Austin. Research and Development Center for Teacher Education.
Identifiers - Assessments and Surveys: Flanders System of Interaction Analysis