臺灣隨著大專院校的擴增，升學率普遍提高，唸大學已不再是重點，能不能唸好的大學才是關鍵。本文分析明星高中是否存在進入頂尖大學的學校特定效果，利用「臺灣青少年成長歷程研究」問卷調查資料，採Heckman兩階段估計法先控制樣本就讀大學的選擇性偏誤，再以羅吉斯迴歸模型估計，涵蓋的控制變數除了家庭背景、地區變數外，再加入個人能力的不同代理變數，並進行不同組合樣本的敏感性分析。實證結果發現，在控制個人能力變數與明星高中為吸引優秀學生的效果後，明星高中對進入頂尖大學仍具顯著學校特定效果優勢，相對勝算比約2.5倍至2.9倍。至於明星高中的學校特定效果來源將因不同的內涵而有不同的政策意涵，故尚待後續研究釐清。

Along with the rapid expansion of higher education in Taiwan since the mid-1990s, being admitted by a university is no longer a key issue, whereas entering an elite university remains as a keen space for competition. This paper uses data from Taiwan Youth Project to test the school-specific effect of elite high schools on entering elite universities. We deal with problems of sample selection and the appealing effect for students with good academic performance in elite high schools. After controlling for regions, family socioeconomic status, personal characteristics, and unobserved ability, our estimation results show that there exists a positive and significant school-specific effect with an odd-ratio of 2.5 to 2.9, for students of elite high schools to enter elite universities. However, the connotations of the school-specific effect may lead to varied policy implications and thus require further investigations.