The success of educational programs hinges on students' attitudes towards learning and their school engagement. This study investigates the factors influencing school engagement, focusing on attitudes towards learning and the opposite sex, and comparing engagement levels in single-gender versus mixed-gender schools. School engagement is defined as the extent to which students identify with the school, embrace its objectives, and participate in school activities. It encompasses affective, cognitive, and behavioral dimensions. High school engagement is linked to higher attendance, lower dropout rates, and better academic performance. Prior research has explored various factors influencing school engagement, including friendship relations, teacher attitudes, peer support, and parental socioeconomic status, with varying results. This study examines the role of attitude towards learning, acknowledging the diverse definitions of learning (behaviorist, cognitive), and the influence of positive learning attitudes on academic success. The study also explores the relationship between attitudes towards the opposite sex and school engagement, considering the developmental stages of these attitudes and their potential impact on identity formation. Finally, the study investigates the impact of school type (single-gender vs. mixed-gender) on school engagement, acknowledging the ongoing debate about the relative merits of each type of school environment and contrasting previous research findings.

Existing literature presents diverse perspectives on school engagement, its definition, and its predictors. While some studies highlight the role of friendship relations and teacher attitudes (Erdoğdu, 2016), others emphasize peer support (Shin et al., 2007) and teacher support (Thaliah and Hashim, 2008). Socioeconomic status also appears to play a role, with some studies showing higher engagement among students from lower socioeconomic backgrounds (Arastaman, 2009; Conchas, 2001). The concept of attitude towards learning is also multifaceted, with behaviorist and cognitive perspectives offering different interpretations. Positive attitudes towards learning are consistently linked to improved academic performance (Bråten and Strømsø, 2006; Duarte, 2007; Aktürk, 2012; Erdoğdu, 2017). The relationship between attitudes towards the opposite sex and school engagement is less explored, though some research suggests a link between opposite-sex friendships and social self-efficacy (Başaranoğlu, 2011; Türkoğlu et al., 2015). The literature concerning coeducation versus single-sex education also presents conflicting findings regarding academic achievement, with some studies favoring single-sex education for girls (Kohlhaas et al., 2010; McFarland et al., 2011; Mulholland et al., 2004; Thom, 2006) and others finding no significant difference (Marsh and Rowe, 1996; Elam, 2009). This study aims to contribute to this complex body of research by analyzing multiple predictors of school engagement.

This relational survey study utilized stratified purposive sampling to select students from girls', boys', and coeducational high schools in Istanbul. The sample comprised 843 students (525 girls, 318 boys). Data collection involved three instruments: the School Engagement Questionnaire (SEQ) (Arastaman, 2006), measuring overall school engagement; the Scale of Attitudes Towards Learning (SATL) (Kara, 2010), assessing attitudes towards learning across four factors (Nature of Learning, Expectation, Openness, Anxiety); and the Opposite-Sex Attitude Scale (OSAS) (Erdoğdu, 2018), measuring attitudes towards the opposite sex. An information form gathered demographic data, including parental attitudes and education levels. Cronbach's alpha was calculated to assess the reliability of each scale, with values of 0.95 (SEQ), 0.85 (SATL), and 0.96 (OSAS) obtained. Pearson correlation coefficients examined relationships between variables. Multiple regression analysis determined the extent to which categorical variables (school type, achievement level, parental attitudes, parental education) predicted school engagement. Stepwise regression identified the most significant predictors and their contribution to the variance in school engagement scores.

Descriptive statistics revealed mean scores of 100.87 (SEQ), 93.35 (OSAS), and 146.40 (SATL). Correlations showed a moderate positive relationship between SEQ and SATL (r = 0.486, p < 0.01), but no significant correlation between SEQ and OSAS. A positive correlation existed between SEQ and achievement level (r = 0.236, p < 0.01), and negative correlations with mother's (r = -0.154, p < 0.01) and father's education levels (r = -0.185, p < 0.01). Multiple regression analyses revealed that school type significantly predicted school engagement (R = 0.342, R² = 0.117, p < 0.001), with girls' high schools showing the highest engagement. Achievement levels also significantly predicted engagement (R = 0.261, R² = 0.068, p < 0.001), with unsuccessful students showing lower engagement. Parental attitudes (R = 0.145, R² = 0.021, p < 0.001) and father's education level (R = 0.228, R² = 0.052, p < 0.001) also showed significant, though smaller, effects. Stepwise regression analysis indicated that attitude towards learning was the most significant predictor (24% variance explained), followed by school type (7%), achievement level (2%), and parental factors (less than 5% combined).

The findings confirm that attitudes towards learning are the strongest predictor of school engagement, supporting prior research on the importance of positive learning attitudes for academic success. The higher engagement levels among girls in single-gender schools suggest that the school environment plays a significant role in influencing engagement, particularly for girls. This could be attributed to factors like increased interaction among girls and the potential pressure on girls in conservative family structures to attend school regularly. The negative correlation between OSAS scores and school engagement in adolescents suggests that their increasing interest in the opposite sex might divert their attention from school-related activities during this developmental stage. The influence of parental education levels, particularly fathers', on school engagement is complex and warrants further investigation. The contrasting findings regarding parental education levels across studies highlight the need for more nuanced research in this area.

This study contributes to a better understanding of the factors that influence school engagement. The findings highlight the crucial role of attitudes towards learning, school type (especially for girls), achievement levels, and parental factors. The complexity of these relationships suggests the need for a multi-pronged approach to enhance school engagement, addressing students' learning attitudes, creating supportive school environments, improving academic performance, and engaging parents. Future research could focus on longitudinal studies examining the development of these attitudes over time, exploring the specific mechanisms by which school type affects engagement differently for boys and girls, and investigating the interplay between parental factors and other contextual influences.

The study's limitations include its cross-sectional design, limiting causal inferences. The sample was drawn from Istanbul schools, limiting the generalizability of the findings to other contexts. The reliance on self-report measures may be subject to response biases. Further research using longitudinal designs and diverse samples is needed to strengthen the findings and increase their generalizability.