影響科技大學學生英語學習效能之因果模式探討

A Study on the Casual Model of Technology University Students’ English Learning Effectiveness

本研究使用線性分析模式與多群組模式檢定，調查南部科技大學學生501人在英語學習信念、焦慮與效能之間的差異與關係，進而建立英語學習效能之因果模式。線性分析模式研究結果顯示：女生在學習信念上，比男生高，尤其在英語價值層面，性別在學習焦慮沒有顯著差異，女生在學習效能的完成目標層面較高，其餘都低於男生；外語系學生在學習信念與學習效能上，普遍高於非外語系學生，但師生互動焦慮與害怕評價，呈現比非外語系學生高。多群組分析結果顯示：，外語系學生學習信念與焦慮呈現正相關，而非外語系學生則呈現負相關；而學習信念正向影響學習效能，學習焦慮負向影響學習效能，外語系與非外語系學生英語學習效能模式中學習信念與焦慮的直接效果分別為.56與.51，-.28與-.27，整體模式解釋量.38與.36。

 

This study conducted both the general linear model and multiple group analysis to analyze the differences and relationships among English learning beliefs, English learning anxiety, and English learning effectiveness among 501 technology university students in southern Taiwan. Furthermore, this study attempted to develop a causal model of English learning effectiveness. The outcome of the general linear model showed that female students had higher English learning beliefs than male students, especially in the dimension of English value, but there was no significant difference in English learning anxiety based on gender. Female students showed higher English learning effectiveness only in the dimension of goal attainment compared with male students. English majors had higher English learning beliefs and English learning effectiveness than non-English majors, but English majors had more English learning anxiety, especially in the dimensions of teacher-student interaction and negative evaluations. According to the multiple group analysis, English learning beliefs was positively correlated to English learning anxiety for English majors, but there was a negative correlation between them for non-English majors. English learning beliefs had a positive direct effect on English learning effectiveness, with respective effect sizes of .56 and .51 for English majors and non-English majors. English learning anxiety had a negative direct effect on English learning effectiveness, with respective effect sizes of -.28 and -.27 for English majors and non-English majors. The two types of English learning effectiveness models could explain up to .38 and .36 of the variance of English learning effectiveness.