在高等教育持續擴張與失業率屢創新高的台灣，「教育與職業不相稱」的問題值得重新獲得關注。有鑑於此，本研究在比較不相稱測度方法的同時，檢視台灣勞力市場中，「教育與職業不相稱」對工作者薪資的影響。本研究分析「台灣地區社會變遷基本調查」的資料顯示，過去經常為研究者所使用的「標準差法」容易發生低估「過量」教育的情形，而「工作者的自我評量法」則容易產生高估「過量」教育的偏誤。 本研究主要是對此兩種測度方式進行修正，以「半標準差」模式取代傳統的「標準差法」，而以「自我評量的標準差法」調整修正「自我評量法」。修正後的測量方式對於薪資具有提升解釋力的效益，有利於更清楚地解析「教育與職業不相稱」的現象。此外，在控制個人職業相關因素之後，相對於適量教育者，「過量教育」依然對薪資報酬有顯著正向的效益。這個結果顯示出，「教育」在台灣勞力市場中，除了具有人力資本的實質效果之外，亦可能帶有相當影響作用的象徵意義。

Continued expansion of higher education coupled with all-time high unemployment rates in Taiwan during the recent years have rekindled concerns among scholars about the consequences of mismatch between occupation and schooling. The major purpose of this study is twofold: first, to examine the influence of occupational mismatch on earnings; second, to compare the effects of alternative mismatch measures on earnings. The data analyzed in this study come from the Fourth Wave of Taiwan Social Change Survey, an island-wide survey. The results indicate that the standard deviation approach of mismatch tends to underestimate the extent of over-education, whereas the self-assessment approach tends to over-estimate it. This study adjusts the measurement bias of these two approaches. Specifically, it employs half of one standard deviation, instead of one standard deviation, to correct the measurement bias resulting from the standard deviation approach. In addition, this study proposes a new measure combining self-evaluation with the standard deviation method to adjust the bias created by the self-assessment approach. Both remedy measures enhance the statistical power in explaining the variance of earnings; they also improve our understanding of occupational mismatch. Furthermore, this study finds that over-education, compared to adequate-education, still has a significant positive effect on earnings return. The results indicate that in Taiwan’s labor market, education not only serves as one form of human capital, but also has a significant symbolic meaning.