Family involvement at secondary schools: Scale adaptation

The aim of this study is to adapt the survey of ?High School and Family Partnerships Teacher Parent Students Survey?, developed by Epstein, Connors and Salinas (1993), into Turkish.The sample group of the research consists of 221 teachers working at secondary education institutions, 285 students and 252 parents in the city center of Kastamonu. After the necessary corrections, the adapted scales have been determined as; 107 questions for teachers, 118 questions for students and 148 questions for parents.The Cronbach Alpha Coefficient used to determine the reliability of the scale was found as 0.84 for teachers, 0.84 for students and 0.89 for parents. To determine the reliability of the scale, Linear Factor Analysis was implemented over 8 dimensions for teacher and student forms and 7 dimensions for parent forms. As a result of this analyze, the rate of ?2/sd for teacher scale was 1.71, RMSEA 0.057, SRMR 0.077, CFI 0.85, GFI 0.61; the rate of ?2/sd for student scale was 1.61, RMSEA 0.047, SRMR 0.072, CFI 0.80, GFI 0.62; the rate of ?2/sd for parent scale was 1.73, RMSEA 0.054, SRMR 0.079, CFI 0.87, GFI 0.64. When the fit indexes of model-data were examined entirely, it was observed that the rates of ?2/sd, RMSEA, and SRMR were at the boundaries of acceptable, whereas the rates of CFI and GFI were below the boundaries of acceptable. Although the scores of CFI and GFI are low, the other indexes?, especially the scores of ?2/sd, RMSEA, and SRMR, being at the boundaries of acceptable and showing the goodness of fit gives information about the goodness of the scale validity. It is stated that the scales which are adapted based on the obtained results of the research can be used in Turkey by researchers and schools.