Final
The mart_student_activity_score.final mart presents the final student activity score for a student in an academic term for a given week in the term. The mart_student_activity_score_rolling.final mart presents the final student activity score for a student in an academic term for a rolling 7-day window.
BQ Prod Dataset Locations
mart_student_activity_score
mart_student_activity_score_rolling
Student Activity Score vs. Student Activity Score Rolling
Both the student_activity_score and student_activity_score_rolling marts aggregate a student's performance and activities in a course in a 7-day time window. The primary difference is whether the 7-day time window is static or moving.
The student_activity_score mart defines static weeks in a term, every week starting on Sunday and ending on Saturday. The weeks are calculated based on the start date of the academic session, term, or course. We rely on BigQuery's DATE_DIFF function to determine a week number, and this BigQuery function has the start of the weeks on Sundays. We perform the date diff anchored against the start date of the academic session, the academic term, or the course in scope. Our logic prefers the academic session or academic term start dates when possible; this creates a more even baseline for week numbers across courses at scale. Each week in the term will have a fixed start and end date based on the session/term/course start date.
The student_activity_score_rolling mart, on the other hand, defines rolling weeks by defining a 7-day time window for each day in a term. The day in a term is calculated using the same logic described above for calculating the week in a term. The time window includes the current day and the previous 6 days. There are some instances where the time window is shorter: if the day in the term is still within 6 days of the first day in the term, since the earliest possible start date is the first day in the term. The rolling 7-day window can be defined on and after the 7th day in the term, until the final day in the term.
In this documentation, any reference to week in the student_activity_score mart is a static week, while week in the student_activity_score_rolling mart is a rolling 7-day window.
Interactive Mart Dependency Diagram
The following visualization shows the construction of this data mart defined in the UDP marts repository. More information on the repository and diagram can be found on this marts page.
Schema
mart/student_activity_score/final
university_id
STRING
The SIS ID of the student.
global_user_id
INTEGER
The UDP ID of the student.
canvas_user_id
STRING
The Canvas ID of the student.
term_name
STRING
The academic term the student is enrolled in, i.e. Fall 2020.
campus_name
STRING
The campus of the student for the given academic term.
academic_program
STRING
The academic program the student is enrolled in for the given academic term.
week_number
INTEGER
The week of the academic term.
week_start_date
DATE
The start date of the week of the academic term.
week_end_date
DATE
The end date of the week of the academic term.
avg_student_activity_score
FLOAT
The average student activity z-score for the given week in the academic term.
avg_student_activity_score_cumulative
FLOAT
The average student activity z-score for the given week in the academic term using cumulative assignment and submission metrics.
student_activity_score_octile
FLOAT
The octile of the student's average student activity z-score in comparison to the average student activity score for a cohort of students based on their academic program.
student_activity_score_octile_cumulative
FLOAT
The octile of the student's average student activity z-score in comparison to the average student activity score for a cohort of students based on their academic program. Based on cumulative assignment and submission metrics
mart/student_activity_score_rolling/final
The mart_student_activity_score_rolling.final schema is the same as mart_student_activity_score.final, except that the static week definitions are replaced with rolling week definitions.
week_number
INTEGER
The week of the academic term.
week_start_date
DATE
The start date of the rolling 7-day window, inclusive.
week_end_date
DATE
The end date of the rolling 7-day window, inclusive.
Student Activity Score
The student activity score defines a "score" for a student's activity in a course's learning environment. The score is calculated using the formula:
5*[submissions/assignments_due]+2*[navigation_time]+[num_sessions]
The metrics in brackets are z-scores. The z-score for a student is computed for a week in the course, whether the week is static or rolling. In student_activity_score_rolling, the navigation time and number of sessions are based on a one-week period instead of two.
Average Student Activity Score
The average_student_activity_score field is the average student activity z-score across all courses a student is enrolled in for a given week (static or rolling) in an academic term.
Student Activity Score Octile
The student_activity_score_octile is the percentile of a student's activity z-score compared to other students. Students are cohorted based on their academic program for the given academic term. The octile of the student's activity score is computed using the PERCENTILE_CONT function in BigQuery.
10 or More Students Criteria
Originally, the Student Activity Score from IU was only computed on courses with 10 or more students enrolled. This filter is no longer enforced. Instead, the course_final and course_section_final tables have a boolean field called course_has 10_or_more_students as a flag that allows users to include or exclude course(s) to align with IU's original definitions. Please see the documentation pages for course_final and course_section_final for more information on this flag.
Last updated