Here's a quick guide on using this script/Jupyter Notebook.

1. Download the response sheet from Google Sheets as a .csv file. Note that the columns will need to be ordered correctly, i.e.

   A1 - Member 1	Q1 for A1M1	Q2 for A1M1	Q3 for A1M1	Q4 for A1M1	A1 - Member 2	...

2. Download the student email sheet as a .csv file. Make sure that the student names in this sheet matches the names used in the peer appraisal form.

   Name	Email

3. Copy both .csv files into the same directory as this script.

4. Install the dependencies required by running pip install -r requirements.txt.

5. Update the constants in the script as necessary, e.g. TUTOR_NAME, etc.

6. Keep SHOULD_SEND_EMAIL as False and run the entire notebook once. You should see a peer-appraisals folder being created, with the feedback stored as .txt files within. For example, student-a.txt would contain the feedback to be sent to Student A.

   • If errors occur, e.g. due to assertions failing, check that the files provided are correct + constants are defined correctly.

7. Once you're confident about the feedback generated, update SHOULD_SEND_EMAIL to True, and update all email-related constants.

   • You may need to enable less secure apps for your Gmail account: https://support.google.com/accounts/answer/6010255?hl=en

8. Re-run the entire notebook, which should now blast out the emails as well.

The reason for choosing .csv over .xlsx is that .xlsx files generated by Google Sheets sometimes have strange reference errors (despite the original sheet not having any references), and they have new lines removed from the responses, which is not desirable.