Examining Middle School Students' Speaking Self-Efficacy in Terms of Various Variables

This study aimed to determine middle school students’ speaking self-efficacy levels and examine whether these levels differed according to individual characteristics, digital media use, reading habits, and parental education. The study employed a relational survey model within a quantitative research approach. The sample consisted of 546 middle school students attending schools in a province in western Türkiye. Data were collected using the Speaking Self-Efficacy Scale for Middle School Students along with a personal information form prepared by the researchers. Independent-samples t-tests, one-way analysis of variance, Welch tests, and appropriate post hoc comparison tests were used to analyze the data. The findings showed that speaking self-efficacy scores differed significantly according to some of the variables examined. Regarding gender, a significant difference in favor of girls was identified only in the susceptibility to influence subscale. As age and grade level increased, scores tended to decrease, particularly in the affective, content, and paralinguistic subscales. Longer daily durations of digital gaming, social media use, and internet use were associated with lower scores in several speaking self- efficacy subscales. In contrast, speaking self-efficacy scores increased as weekly reading time rose. The number of books read per month produced a significant difference only in the affective subscale. A significant difference was identified in the paralinguistic subscale according to maternal education and in the content subscale according to paternal education. Overall, the findings indicate that middle school students’ speaking self-efficacy is associated not only with individual characteristics but also with digital media use, reading habits, and parental education. The results are expected to contribute to instructional practices aimed at improving speaking skills and to future research in the field.