Remote-Instruction-Analysis

Comparing pre (2019 Spring, 2018 Spring, 2017 Spring) and post (2020 Spring) pandemic student and instructor engagement analytics from university-wide LMS logs.

• Set up hyperparameters in file utils.py, which includes name, start date, end date of each semester, when remote instruction took place, and directories of LMS logs.

• Before the start of analysis, run the following code to preprocess users.

  • python preprocess_user.py

  • Two files will be generated:

    • user_id_groups.pkl: a dictionary mapping each role into user_IDs of that role. (roles including designer, observer, student, studentview, ta, teacher)
    • user_racial_groups.pkl: a dictionary mapping each racial group into student_IDs of that race.

Instructor Engagement:

• Assignments released per day over the semester:

  • Run instructor_engagement/assignments_rate.ipynb

• Number of late excused submissions:

  • Run instructor_engagement/excused_submission.ipynb

• Instructor changes to due dates after release (i.e., reactive extensions)

  • Run instructor_engagement/instructor_override_grade_asn.ipynb

• Instructor changes to due dates before release (i.e., proactive extensions)

  • Run instructor_engagement/instructor_override_grade_asn.ipynb

• Instructional staff announcements

  • Run instructor_engagement/Rate_instructional_announcements.ipynb

• Assignment submission types (2020, 2019, 2018, 2017)

  • Run instructor_engagement/submission_type_changes.ipynb

• Instructor time to grade assignments after submission

  • Run instructor_engagement/instructor_grade_time.ipynb

• Instructor comments on submissions

  • Run instructor_engagement/instructor_comments.ipynb

• Length of instructor comments (in bytes)

  • Run instructor_engagement/instructor_comments.ipynb

Student Engagement:

• Number of submissions per day

  • Run student_engagement/Rate_submission_completed.ipynb

• Number of student late submissions

  • Run student_engagement/late_submission.ipynb

• Total number of daily student events

  • Run student_engagement/student_actions.ipynb

• Number of course drops per day

  • Run student_engagement/Drop_out_rate.ipynb

• **Student discussion activity over the semester **

  • Run student_engagement/Student_discussion_trend.ipynb

• Student discussion activity over the semester on graded/ungraded topics

  • Run student_engagement/Student_discussion_trend (graded or ungraded).ipynb

• Average grade on assignments turned in per day

  • Run student_engagement/assignment_GPA.ipynb

• Assignment category proportions (Spring 2020)

  • Run instructional_quality/assignment_group_Spring20.ipynb

Student Engagement by Race:

• Percentage of active students per week

  • Run student_engagement_race/active_student_race_proportion.ipynb

• Weekly student events

  • Run student_engagement_race/student_actions_race.ipynb

• Weekly proportion of submission

  • Run student_engagement_race/Rate_submission_completed_race.ipynb

• Average grade on assignments turned in per day

  • Run student_engagement_race/assignment_GPA_race_weekly.ipynb

Enrollment Analysis:

• Did student grades in Fall 2020 classes differ if the prerequisite for the class was satisfied in the emergency remote instruction semester?
  • Run enrollment_analysis/preprocess_edw.py , the results will be saved in a file named results.pkl.