Observation-Team Reliability Determined by Analysis of Variance
- 1 September 1969
- journal article
- research article
- Published by Taylor & Francis in The Journal of Experimental Education
- Vol. 38 (1) , 1-4
- https://doi.org/10.1080/00220973.1969.11011158
Abstract
This report examines the reliability and the sources of variation of a part of an observation schedule using the model first proposed by Medley and Mitzel. Reliability coefficients for the three categories of behavior, mobility, involvement, and irrelevant acts, were found to be . 72, . 67, and . 69 for the entire team. When the data were considered only in terms of the four observers present during both observations the reliability coefficients were . 84, . 85, and . 90. That the second analysis yielded higher r’s was to be expected, since there was less variation in observers. The question of why the third category was lowest in the first instance and highest in the second is still to be investigated.Keywords
This publication has 1 reference indexed in Scilit:
- Application of Analysis of Variance to the Estimation of the Reliability of Observations of Teachers’ Classroom BehaviorThe Journal of Experimental Education, 1958