Mastery Grading in a Software Engineering Course

Carlos Rojas, Ph.D.
Gina Quan, Ph.D.

San José State University

Motivation

Our current system has turned higher education into a game that students win when they obtain the most coveted rewards—that is, high grades—for putting in the least amount of time and effort.
-Linda B. Nilson, Ph.D.

L. B. Nilson. Specifications Grading: Restoring Rigor, Motivating Students, and Saving Faculty Time. Stylus Publishing, LLC, 2015

What is Mastery Grading?

Students are assessed on whether they meet a threshold, i.e. mastery

Each assignment has rubrics that establishes the threshold

There is no partial credit

Students are allowed retakes for assessments

Course Background

  • First upper division course for Computer Engineering and Software Engineering majors
  • Spring 2022 taught two sections with 34 and 49 respectively (N=83)
  • 10%-30% non-majors

How I applied Mastery Grading?

  • I created several assessment categories
  • Each category specifies how many assignments need to pass to earn letter grade
  • Created a currency called Tokens to limit revisions and add flexibility to their learning

Quizzes

Homeworks

Project

Exams

Letter Grade Mapping

Assignments were graded on a tertiary system: needs revision, low pass, and high pass

A student needs to meet the grade for every category to earn that grade

Results

Method & Goal

We survey students at regular intervals during the semester

We wanted to know how students attitude toward mastery grading evolved over time

Student Survey Distributions

Student Survey Distributions

  • First and last survey
  • Students felt they had the background knowledge to succeed
  • Understood how to to improve their grade
  • Median did not prefer mastery over other assessments

Results of all Surveys

Results of survey questions taken at five time points

We converted the likert scale to range from -2 to +2 and averaged

What did we learn?

  • Need to invest in student buy-in and possibly departmental buy-in
  • Course preparation is time consuming, but it can be done in stages
  • Need to give consideration of how well your assessments map to the course learning outcomes

Acknowledgements

Teaching Experiment Academy (TEA)

Spring 2022 CMPE 131: Software Engineering class

Thanks!

emailcarlos.rojas@sjsu.edu

Websitehttps://carlosrojas.xyz

Mastodon@cr

GitHub@carlosrojas