大學科系興趣導向分類系統和本土興趣模式之初探研究

Emic Interest-Based Classification and Interest Model of College Majors in Taiwan

大學科系分類系統是整理高等教育資料的依據，大學科系興趣模式協助學生大學科系興趣的探索。國內使用的中華民國學科標準分類和十八學群，屬於專家導向分類系統。Holland（1959, 1997）興趣模式則常被用來建製大學科系的興趣導向分類系統，以協助學生探索大學科系興趣。然而，Holland（1959）興趣類型模式源自美國人對職業的興趣，目前研究雖證實Holland（1997）類型在國內的適用性，但尚未有研究去探索國內大學科系興趣架構的實際面貌。此研究依據國內高中生對於大學科系興趣，建立興趣導向的大學科系分類系統和大學科系興趣模式。1306位高三學生填寫他們對於166個大學科系的喜好程度，此166個大學科系取自中華民國學科標準分類系統。完整連結式群集分析（complete-linkage cluster analysis）結果得到三階層式的大學科系分類系統。此三階層分別為：166個科系、26個學群和7個領域。多向度標定分析（multidimensional scaling）把26個學群標定在三個向度，形成大學科系興趣模式。此三向度分別是：學習內容（物體－人心）、科系特質（動態－結構），以及功能目的（溝通－發現）。研究結果對照到中華民國學科標準分類、十八學群和Holland（1997）興趣類型，並討論相關應用。

Indigenous interest-based classifications and interest models of college majors are essential for both the career exploration of students and higher education policymaking. In Taiwan, Holland's Codes are frequently used to determine college majors based on interest. However, Holland's Codes were developed in the United States and only later adopted in Taiwan. Therefore, this study developed an emic interest-based classification and an interest model of college majors by analyzing the preference ratings of 1,306 high school students for 166 college majors in Taiwan. A complete-linkage cluster analysis was used to construct a three-level hierarchical classification with 166 college majors, 26 major groups, and 7 major fields. Multidimensional scaling of the 26 major groups was used to create a spatial interest model consisting of three dimensions: content (object-mentality), characteristics (dynamics-structure), and function (communication-discovery). Convergence and divergence were observed between the results of the Holland Codes and the two commonly used content-based classifications of Taiwanese college majors.