Providing Project Feedback

Every student should have a chance to understand how they're doing and to get valuable feedback on the work they've done from trusted mentors like us.

Our goal is to provide meaningful suggestions and personal guidance to all of our students. While there will undoubtedly be nuances in how each of us evaluates student portfolio projects, using these guidelines will ensure that students get the personalized feedback they deserve.

Guiding Principles

• Prompts and rubrics are provided to structure feedback for student projects.

• Rubrics are based on a simple 3-point scale, tailored for each project.

• Metrics for "meets expectations" are clearly defined to help you provide targeted feedback.

Feedback Rules

• Students should be shown a copy of the blank rubric when given their project prompt.

• Rubric categories and evaluation metrics should always be tied to project specifications.

• Project categories are based off of the course standards, found in the Data Science Workflow.

• Instructors should work with their TA's to provide targeted feedback for every student.

• Students should receive feedback on their projects within 2 - 4 days after submission!

Rubric Assessment Process

1. Make a copy of the relevant project rubric.

2. Score project requirements using rubric criteria.

• Note: Half points are ok (1.5) but beyond that, detailed percentages are unnecessary.

3. Enter any notes for specific project deliverables.

4. Enter any notes for overall student progress report, if applicable.

Feedback Tips

• Provide precise praise

• Identify action step for improvement

• Give context as to why what action step is important

• End with a targeted question related to the action step, that tests whether the student has understood the feedback.

  • Good Example: 'Great work! Your code is well organized. Good use of breaking tasks into separate functions. I see that in your app.js file you used the variable name x on line 84. Consider using more descriptive variable names. As a code reviewer, this would jump out at me. I want to know what every variable contains just by looking at the variable name itself. I do see some other descriptive variables! If we were to refactor this code, what would you want to rename x on line 84?'

  • Poor Example: 'Great Work! You’re doing a good job with all this. Do you have any questions about anything?'

In-Person Assessment Process

Instructors and TA's are encouraged to meet with students during office hours or one-on-one to review project feedback and discuss growth opportunities. Students should ideally have up to 30 minutes to review their project evaluation or do a code review with instructors and TA's during the week after project submission.

Model / Code Review Tips

Code review is not troubleshooting time. Instead, the goal is to review code and talk about actionable feedback for enhancing the student's project in the future.

• Student (10 mins): Have the student walk through and defend their project / model / code.

• Student (10 mins): Have the student answer the following questions.

  • What do you like most about your project?

  • What would you change if you started from scratch?

  • What would you like to add when you have more time?

• Instructor/TA (10 mins): Talk through the evaluation, and share a “glow and grow”.

  • Make it personal – what about the project was exciting or intriguing?

  • Share any highlights and positive areas (“glow”) where the student went “above and beyond” the requirements

  • Share growth areas around time management, workflow, or project approach

  • Share growth areas around technical skills and ways to remediate any difficulties the student encountered

  • Help the student prioritize and identify which growth area is most important to focus on before the next project.

  • Set 1 or 2 specific goals for the next project.

  • If you’re reviewing the second or third project with the student, ask them how they did against the goals they set previously?

• Finally, be sure to report a summary of your evaluations back to your team, so we can evaluate how the class is doing as a whole!