学业自我概念的大鱼小池效应是指进入水平相对较高学校的学生较同等学业水平但进入水平相对较低学校的学生，学业自我概念偏低。该效应是当前学校影响学生发展研究的重要问题之一。本研究利用中国儿童青少年心理发育特征调查项目所建立的来自全国31个省市区607所中小学校25172名3至9年级学生的数据库（男生约占54.98%），第一次探讨了我国中、小学数学自我概念发展的大鱼小池效应，并从学生年级、性别和个人能力等个体特征和学校心理环境这一学校特征探讨了数学学科上学业自我概念发展的大鱼小池效应，及其在个体、学校间的差异。研究结果表明：1. 控制了学生个体学业成就后，学校平均数学学业成就对学生的数学自我概念存在显著负向效应，即我国中、小学生的学业自我概念发展中存在大鱼小池效应。控制了学生所属城乡和学生家庭社会经济地位后，数学学业自我概念上的大鱼小池依然存在。2. 数学学业自我概念上的大鱼小池效应在学生五年级时出现，并表现出年级越高，负效应越大的变化特点。3. 总体上，学生数学学业自我概念的大鱼小池效应在男、女生间没有显著的差异。仅在9年级，学生数学学业自我概念的大鱼小池效应在女生中更强。4. 总体上，学生个体学业成就对数学学业自我概念的大鱼小池效应具有U型的调节作用。4-9年级学生中，数学学业自我概念上的大鱼小池效应在数学成就较好和较差的个体中较强，在数学成就中等的个体中较弱。3年级学生中，数学学业自我概念上的大鱼小池效应在数学成就较好和较差的个体中较弱，在数学成就中等的个体中较强。5. 学校心理环境特点对小学、中学数学学业自我概念具有保护作用。相比在女生中，这个作用在男生中更强，在不同年级和不同学业成就的个体中没有差异。

Big-Fish-Little-Pond Effect (BFLPE) refers to the negative effect on students’ academic self-concepts from school-average ability level. As having been well documented widely in western and eastern societies, it has significant implications for improving school effectiveness. Utilizing data from National Study of Child, data of 25172 students from 3rd to 9th grade in 607 schools, was used to examine the existence of BFLPE within Chinese representative sample and scrutinized possible moderators within individual and from schools on math subject. Suggestions for future research directions also were discussed. Research question 1 examined the existence of BFLPE within Chinese representative sample by building two-level (student-school) hierarchical linear model (HLM) on students’ academic self-concept, academic attitude and external motivation. Based on research question 1, research question 2 discussed the possible moderators within individual students, like grade, gender and individual ability. Extending the search of moderators to school level, research question 3 examined the moderating effect of school psychological environment on BFLPE. Main findings were:1. Controlled for student’s individual academic ability, school-average ability had negative effect on student’s academic self-concept in math, which confirmed the existence of BFLPE in Chinese sample. To ensure the robustness of results, we controlled region and family SES, the BFLPE were still significant.2. The negative effects of school-average math ability on math self-concept emerged on 5th grade and kept going stronger to 9th grade. 3. BFLPEs were consistent among boys and girls, with an exception of 9th grades. BFLPE was stronger in girls among 9th grades. 4. In general, individual math ability moderated BFLPE in U shape. Among 4th to 9th graders, the BFLPEs were most severe among most able and least able students in math. Among 3rd graders, the BFLPE was most severe among average able students in math.5. School psychological environment protected students’ math self-concept. The moderating effect was stronger in boys and consistent among individuals of different grade or math ability.