Canvas Get More Quiz Details

⚠️ WIP! questions? Ask Alison!

• name: canvas-better-quiz-stats
• runs-with: terminal
• python >= 3.8
• canvasapi>=2.0.0 (likely built with 3.0)

⚠️ Things I don't know yet

• if it works properly with multiple attempts
• what randomization in questions looks like (variables)
• how to get the points of grouped questions

Summary

Canvas quiz statistics are unavailable for large courses, and do not display useful information when randomization in questions (either through pick X, or variables) are used. This script creates a number of csv files for a single quiz - including the questions as seen (and responded to) by individual students, as well as the original question information.

Input

• Canvas Instance (instance of Canvas being used - ex. https://ubc.instructure.com)
• Canvas API Token (generate through Account => Settings)
• Course ID (last digits of URL when visiting course page)
• Quiz ID

Output

Creates outputs in folder: data/course_id/quiz_id

• Note: question points possible not included - for grouped questions (i.e. "pick X of Y") the points applied may overwrite original question points

student_view_questions_answers.csv

This file is a combination of the three files listed below.

• quiz_id: the Canvas quiz id
• attempt: the quiz attempt of the individual student
• user_id: the Canvas user id of the submission
• quiz_submission_id: a Canvas submission id (unique to student and attempt)
• quiz_submission_question_id: the question id of the submission (should match the question_id, included as sanity check)
• question_id: the Canvas question id
• quiz_submission_question_question_name: the submission question name (as seen by the student). Important for "pick X" type questions and/or randomization of order. This will be the question order as seen by the student - quiz_submission_question_question_text: the submission question text as seen by the student
• quiz_submission_question_variables: the submission question variables (for question specific to student in case of randomization)
• submission_history_id: (submission_id) a unique id for the student's submission submission_data_correct: for the student whether the submission is correct (TRUE, FALSE, undefined, and partial)
• submission_data_points: the points given to the student's question submission - submission_data_text: the text of the student question submission - Canvas also uses text for m/c and other variable inputs, can be joined with "answers"
• question_name_tagged: the name of the question as created by the instructor. This is not visible to students
• question_type: the question type
• question_text_original: the original question text (if variables or randomization included)
• variables: the original question variables
• answers: answers to the original question

question_info.csv

• the original question information (what the instructor would see when building the assessment

question_submission_info.csv

• the question information (what the student sees)

student_submission_response_info.csv

• the submission answers for the student

Getting Started

First Time (do once)

1. Ensure you have conda installed (Python 3.9 version)
2. Clone Canvas Better Quiz Stats repository
3. Import environment (once): $ conda env create -f environment.yml
4. Create .env file and include:

API_KEY = ''
API_URL = 'https://ubc.instructure.com'

Every Time

1. Run:
   1. navigate to your directory $ cd YOUR_PATH/canvas-better-quiz-stats
   2. activate the environment (see step 3 on first run) $ conda activate canvas_better_quiz_stats
   3. update your environment (.env) file with new token as needed
   4. run the script and follow prompts in terminal $ python src/canvas_better_quiz_stats.py