A study of the relationship between teachers' predicted and observed verbal behavior in the classroom
Date
1980
Authors
Sitwell, John Alexander
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The purpose of this study was to determine the degree of congruence existing between teachers' predicted and observed styles of verbal interaction. Four hypotheses were posited: 1) There is no significant relationship between teachers' predicted and observed time spent using the categories of the Flanders System of Interaction Analysis. 2) There is no significant difference between the amount of time teachers predicted they would spend on each of the categories of the Flanders system and the observed time spent. 3) There is no significant difference between the amount of time teachers' spent using categories 1, 2, 3 and 4 and their predicted time. 4) There is no significant difference between the amount of time teachers' spent using categories 5, 6 and 7 and their predicted time.
The sample for this study was randomly selected from English and Social Studies teachers employed in junior secondary, junior-senior secondary and middle schools of the Greater Victoria School District. These teachers were required, on three separate occasions, to predict the amount of time they would spend using the categories of a revised Flanders System of Interaction Analysis. These lessons were recorded and analyzed by an independent coder.
Analysis was performed using Spearman's rho to determine the relationship between teachers' predicted and observed use of all eight categories. The Wilcoxon matched pairs signed-ranks test was used to test, first, whether a significant difference existed between the teachers' predictions and the time they spent using each of the categories and,second, to determine whether a significant difference existed between teachers' predicted use of indirect and direct verbal teaching style and their observed verbal behavior.
The findings constituted a rejection for hypothesis one. Teachers could, in fact, make an overall prediction regarding the relative proportion of time they would spend in the categories of the Flanders System of Interaction Analysis. Hypothesis two was rejected when data from six of the eight categories indicated a significant difference between the predicted and observed time spent in each category. Hypotheses three and four were also rejected for the same reason.