Abstract:

The purpose of this research is to develop a valid and reliable measurement tool to determine teacher behaviors within classroom. In addition, the relationship between teachers' behavior and a) gender, and b) profession (classroom teachers or subject matter teachers) variables are examined. 301 teachers participated in this research. Voluntary participants were 101 primary school teachers and 190 secondary school teachers working in different cities. 10 teachers didn't state his/her profession.To provide content validity, the draft form of the scale with 46 items was examined by seven experts on the 5 point Likert Scale. According to experts evaluations, 9 items that were not suitable for the purpose of scale were extracted besides pilot survey with 37 items were formed and called as TBCS (Teacher Behaviors Within Classroom Scale). The construct validity of TBCS was examined with EFA (exploratory factor analysis). Factor loadings of 22 items of the scale were greater than .45 and factor loadings of remaining 5 items were greater than .35. This result supported that the scale has one factor.To determine the degree of factor structure fiting with data, DFA (confirmatory factor analysis) was done by Lisrel 8.30 statistic software. The result was .028, so one factorial scale structure fits with the data. For reliability of the scale, Cronbach Alpha coefficient and Spearman Brown splithalf correlation were computed. Cronbach Alpha coefficient was .904 (.893 for the first half, and .911 for the second half). Spearman Brown correlation was computed as .836 (.805 for the first half, and .856 for the second half). To estimate the stability of measure over time, testretest correlation was examined and computed as .84. The results of the analysis showed that the scale which has 27 items was reliable and valid with one factor. Also there was a statistically significant difference between teachers' behaviors and gender, and profession.

Keywords: