ERIC Number: ED041856
Record Type: RIE
Publication Date: 1970-Jan
Reference Count: 0
Development of Improved Techniques for Establishing the Reliability of Observation Ratings. Final Report.
This investigation sought to develop and apply analysis of variance techniques to the estimation of the reliability of observation schedules. It placed special emphasis on the different possible designs and the various administrative situations in which they might be applied. The application of the general model to a specific instance was then carried out. The study was conducted with ten recorders who observed five teachers through a one-way mirror and rated them on an observational schedule. The SUTEC (School University Teacher Education Center) Observational Schedule was used. (A copy is attached to the document.) Items observed included teacher mobility, the involvement of the children, materials present, materials in use, irrelevant behavior by the children, directed behavior, and spontaneous behavior. Analysis of variance was applied to the data obtained. The major conclusions were that different variance component models could be applied in different situations to estimate the reliability of either the entire observation schedule or parts of it, and that the items comprising the SUTEC schedule did differentiate fairly well between teachers. (Author/MBM)
Publication Type: N/A
Education Level: N/A
Sponsor: Office of Education (DHEW), Washington, DC. Bureau of Research.
Authoring Institution: City Univ. of New York, NY.