Predictive power of different psychometric variables to predict grammar success of secondary school students

The aim of this study is to determine to what extent different psychometric variables predict the grammar success of secondary school students. For this reason, the relationships between the grammar success of students and their attitude towards Turkish lesson, anxiety towards grammar, metacognitive awareness, and self-efficacy beliefs were analyzed. In accordance with this aim, as one of the quantitative research methods, the correlational survey model was used in the collection of research data, and the analysis and interpretation of the obtained data. The research was carried out with a total of 663 students as being 363 girls and 300 boys. In the research, the data was collected through the "Grammar Success Test" and "Grammar Anxiety Scale" developed by the researcher, "Attitude Scale Towards Turkish Lessons" developed by Topçuoğlu Ünal and Köse (2014), Karakelle ve Saraç's (2007) "Metacognitive Awareness Scale for Children Form B," "Self-Efficacy Scale for Children" developed by Telef (2011), and "Marmara Learning Styles Scale" prepared on the basis of Dunn and Dunn Learning Style Inventory and developed by Şimşek (2007). One-Way ANOVA was conducted to determine whether the grammar success, grammar anxiety, attitude towards Turkish lesson, self-efficacy and metacognition scores of students differ by the status of father's and mother's education, income, reading books and the course status they are successful in; and Independent Samples t-Test analysis was conducted to determine whether there is a statistically significant difference in score distribution according to gender. Correlation analysis was carried out to determine the relationships between grammar success, grammar anxiety, attitude towards Turkish lesson, self-efficacy and metacognition variables; and multiple regression analysis was carried out to identify predictive power of grammar anxiety, self-efficacy, metacognition and attitude towards Turkish lesson to predict grammar success. Descriptive statistics (mean, standard deviation, standard error, median value, minimum, maximum, number and percentile) were used for other data in the study. According to the findings obtained as a result of the research, it was concluded that attitude towards Turkish lesson, anxiety towards grammar, metacognitive awareness and self-efficacy beliefs were statistically significant predictors of grammar success both individually and together. In addition to the mentioned variables, grammar success was also analyzed in terms of learning styles. It was determined that learning styles should also be taken into consideration in grammar studies. Grammar success also varies by gender, education level of parents, income level, the number of books read daily and the lesson in which students are successful.