Student Grade Data: Grit

Estimating Student Fixed Effects in Explaining Grade Outcomes: The Importance of Grit in Determining Student Performance Grit is a term that has emerged in the literature as a way of describing a person’s persistence over time to overcome challenges and accomplish goals. Past measures of grit have been based on self-reported data. We develop a model of student grit that does not rely on self-reported data by employing a rich data set and fixed effects regression techniques to model student letter grade in a course, a proxy for student learning and success, while controlling for the student’s measured cognitive performance, sociodemographic variables, academic rank, relative high school performance, subject matter of the course, instructor, advisor and an individual student effect, interpreted as grit. We find the individual student effect explains significant variation in student academic performance in the classroom. We also demonstrate a method for predicting student grit using secondary education data. Data represents a cohort of economics majors at a public four-year, full-time selective university located in the midwest. Rows represent a single university course taken by the student. Columns: Grade Grade earned by the student Parent_Work_Income Parent's income, normalized SCHOOL_GPA Student's high school grade point average, normalized ACT_COMPOSITE Student's composite ACT score, normalized STUDENT_RANK Student's college rank when the course was taken (Freshman, Sophomore, etc.) Age Student's age when the course was taken, normalized GENDER_DESC Student's self-identified gender, URM Under-represented minority status Student.22 Student indicators Student.01 Student.36 Student.04 Student.05 Student.23 Student.15 Student.28 Student.27 Student.10 Student.11 Student.12 Student.13 Student.14 Student.30 Student.16 Student.18 Student.31 Student.19 Student.20 Student.03 Student.09 Student.26 Student.21 Student.02 Student.06 Student.32 Student.08 Student.29 Student.07 Student.17 Student.25 Student.33 Student.34 Student.24 Student.35 SubjECON Subject indicators SubjMGT SubjFIN SubjENGL SubjACCT SubjCIS SubjMKT SubjCOMM SubjMATH SubjBIOL SubjBLAW SubjHIST SubjAE SubjPSY SubjMUS SubjELCT SubjPE SubjART SubjPOLS SubjSPAN Instructor.04 Instructor indicators Instructor.08 Instructor.07 Instructor.03 Instructor.16 Instructor.05 Instructor.12 Instructor.02 Instructor.06 Instructor.11 Instructor.13 Instructor.10 Instructor.01 Instructor.14 Instructor.09 Instructor.15 Advisor.02 Advisor indicators Advisor.01 Advisor.04 Advisor.11 Advisor.05 Advisor.09 Advisor.07 Advisor.03 Advisor.06 Advisor.10 Advisor.08 Advisor.12估算学生固定效应以解释学业成绩：坚毅（Grit）对学生学业表现的决定性作用 坚毅（Grit）是学界新近提出的概念，用于描述个体长期克服挑战、达成目标的持续性特质。既往的坚毅测评工具均基于自我报告数据构建。本研究构建了无需依赖自我报告数据的学生坚毅模型：通过利用丰富的数据集与固定效应回归技术，以学生课程字母成绩（作为学生学习成果与学业成就的代理变量）为建模对象，同时控制学生已测得的认知表现、社会人口学变量、学业排名、相对高中表现、课程学科、授课教师、指导教师，以及被诠释为坚毅特质的个体学生固定效应项。研究表明，个体学生固定效应项能够解释课堂中学生学业表现的显著差异。此外，本文还提出了一种利用中等教育数据预测学生坚毅特质的方法。 本数据集的样本为美国中西部一所公立四年制全日制选拔性大学的经济学专业本科生队列。数据集的每一行代表一名学生修读的单门大学课程，各字段说明如下： 1. Grade：学生获得的课程成绩 2. Parent_Work_Income：家长收入（已标准化处理） 3. SCHOOL_GPA：学生高中平均绩点（已标准化处理） 4. ACT_COMPOSITE：学生ACT综合得分（已标准化处理） 5. STUDENT_RANK：学生修读该课程时的大学学业层级（如大一、大二等） 6. Age：学生修读该课程时的年龄（已标准化处理） 7. GENDER_DESC：学生自我报告的性别 8. URM：学生是否属于代表性不足少数族裔（Under-represented minority） 9. Student.22、Student.01、Student.36、Student.04、Student.05、Student.23、Student.15、Student.28、Student.27、Student.10、Student.11、Student.12、Student.13、Student.14、Student.30、Student.16、Student.18、Student.31、Student.19、Student.20、Student.03、Student.09、Student.26、Student.21、Student.02、Student.06、Student.32、Student.08、Student.29、Student.07、Student.17、Student.25、Student.33、Student.34、Student.24、Student.35：学生个体指示变量 10. SubjECON、SubjMGT、SubjFIN、SubjENGL、SubjACCT、SubjCIS、SubjMKT、SubjCOMM、SubjMATH、SubjBIOL、SubjBLAW、SubjHIST、SubjAE、SubjPSY、SubjMUS、SubjELCT、SubjPE、SubjART、SubjPOLS、SubjSPAN：学科指示变量 11. Instructor.04、Instructor.08、Instructor.07、Instructor.03、Instructor.16、Instructor.05、Instructor.12、Instructor.02、Instructor.06、Instructor.11、Instructor.13、Instructor.10、Instructor.01、Instructor.14、Instructor.09、Instructor.15：授课教师指示变量 12. Advisor.02、Advisor.01、Advisor.04、Advisor.11、Advisor.05、Advisor.09、Advisor.07、Advisor.03、Advisor.06、Advisor.10、Advisor.08、Advisor.12：指导教师指示变量