以縱貫學業表現檢驗大魚小池效應與見賢思齊效應

Examining the Big-Fish-Little-Pond and Upward Comparison Effects Using Longitudinal Learning Achievement

在社會比較效應中，大魚小池效應和見賢思齊效應是最常被探討的，不但在學業自我概念的變項上累積許多實徵證據，後續研究更指出這些效應可延伸至學業成就。然而，無論是大魚小池效應或見賢思齊效應，在與長期學業成就相關的研究上皆缺乏合理的驗證標的。因此，本研究在兩方面延伸並擴充以往的研究：首先，在變項檢驗方面，同時以兩種縱貫性的學業成就──在校成績和標準化大學入學考試成績──來檢驗大魚小池效應和見賢思齊效應。其次，在檢驗方法上，除以整體學校分析法來重複驗證過往的研究結果外，另創相鄰志願學校比較法，來比較學業能力相近的高軌學校底端學生和低軌學校頂端學生的學業成就差異。本研究發現，當以在校成績為依變項時，整體學校分析法和相鄰志願學校比較法都支持大魚小池效應而未支持見賢思齊效應，與先前的研究結果一致。然而以標準化測驗為依變項時，整體學校分析法和相鄰志願學校比較法都未支持大魚小池效應，也都未支持見賢思齊效應。

Since tracking became a global trend, numerous countries have implemented ability grouping in elementary and middle schools. Under these circumstances, whether students should attend a standard school or a highly selective school has become a crucial issue. Previous studies investigating the big-fish-little-pond effect (BFLPE) inferred that studying at an elite school is not ideal because students with similar abilities from higher-track schools tend to have lower academic self-concept and achievements compared to their counterparts from lower-track schools. However, studies examining upward comparison effects found that students who compared themselves with academically advanced peers performed better in various academic domains, which suggests that students should attend prestigious schools. These conflicting results indicate that the impact of tracking on learning achievement remains unclear. Additionally, previous studies on both effects did not use appropriate variables to measure long-term academic achievement. This study employs two techniques based on the results of previous studies to increase current understanding of both effects: (1) Two indices for assessing longitudinal academic achievements, that is, students' grade point average (GPA) and standardized college entrance examination, are used to investigate the BFLPE and the upward comparison effect. (2) Two methods of data analysis, that is, an overall-school analysis and an adjacent-school comparison, are used to identify differences in academic achievement among students with a similar ability comprising the lower end of a higher-track school and their counterparts from the higher end of a lower-track school. When GPA was the dependent variable, the results of the overall-school analysis and the adjacent-school comparison supported the BFLPE, but not the upward comparison effect. By contrast, when the standardized test score was the dependent variable, the results of the overall-school analysis and the adjacent-school comparison revealed an absence of both the BFLPE and the upward comparison effect. The results of this study suggest that studying at an elite school induces neither the BFLPE nor the upward comparison effect. Students of similar ability, who attended schools with different rankings, displayed no differences in their longitudinal academic achievement. Therefore, fighting to qualify for a higher-ranked school regardless of all other considerations is not required. The results of this study can be further explored in future Western studies.