Perspectives, Performance Measurement, and Multivariate Analysis of Grades in Teaching Statistics

• Activity Number: 30 - SPEED: Statistics and Econometrics
• Type: Contributed
• Date/Time: Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Education
• Author(s): William Seaver* and Missy Morris and Thomas Edmiston
• Companies: Univ. of Tennessee at Knoxville and Univ. of Tennessee at Knoxville and Univ. of Tennessee at Knoxville
• Keywords: grading; sampling; multivariate analysis; concept scores

Abstract

This research deals with some suggestions on measuring performance on exams, homework, and projects as they relate to the final course grade. Typically, we compute grades as to the contract from our syllabus. On close calls, every instructor has his/her ways to make close calls, such as attendance, final exam grade, homework, trends, etc. In addition, it is not unusual for today's students to nit-pick about grades, a point here and there. Large classes create a lot of work addressing this issue. In addition, how does one deal with the variation in grading that could come from the teaching assistant? Or what about the instructor second guessing the weighting scheme for the class? This instructor has pulled together a former student and teaching assistant for such additional insights. Finally, as statisticians should we not use statistics to check our individual student grades, grading schemes and close calls? We make some suggestions with sampling, multivariate statistics, and conceptual scores. Several different classes are used to display the options for deciding grades.