Developing play-based observation protocol for assessing preschool children

The aim of this study was to develop an observation protocol that can be utilized to observe children's behaviours during play and monitor their development. The study groups of the research consist of 595 preschool students. The data analysis was performed using SPSS 23, JASP, and MPlus software programs. The study was conducted in accordance with the theoretical structure of the Ministry of National Education (MEB) Preschool Education Program (2013) achievement and indicators. In the first part of the study, an item pool was created, and expert opinions were obtained for these items, which were then evaluated based on the experts' feedback. After the expert opinions, a candidate observation protocol form was prepared and named "Play-Based Observation Protocol" (PBOP). Exploratory factor analysis (EFA), Confirmatory Factor Analysis (CFA), discriminant validity, and convergent validity analyses were conducted to establish the construct validity of the observation protocol. To provide evidence of the study's reliability, intra-class correlation reliability was used for inter-rater reliability after EFA and Cronbach's α, and McDonald's ω coefficients were examined for internal consistency reliability after EFA and CFA. Examinations were carried out regarding the reliability of the 27% upper-lower group after CFA. EFA, conducted using the principal component analysis and varimax rotation as factor extraction methods, revealed that the structure consisting of 25 items with four factors explained 56.1% of the variance. CFA conducted using the MLM method resulted in the following fit indices: RMSEA = .06, SRMR = .06, CFI = .91, and TLI = .90, indicating an acceptable model fit. Convergent and discriminant validity assessments were conducted after CFA. The examination of the 27% upper-lower group reliability after CFA found significant t-values for the majority of the items and the total observation protocol score (p <.05).