Feature Request: Implement Percentile Rank Feedback System

[LuminousYuzu]

Objective:

Develop a system to provide users with their percentile rank based on a summary statistic from an experiment. This involves backend modifications and a new frontend component for visual feedback.

Requirements:

Backend Modifications:

• tabulateAndPostResults Endpoint:
  • Modify to handle and pass the summary statistic.
  • Ensure it writes results to the database.
• getFeedback Endpoint:
  • Consider creating a new endpoint to retrieve feedback without writing to the database.
  • Allow users to access their percentile rank using userID, experiment, and summary statistic.

Frontend Component:

• Create a visual feedback component to display the percentile rank.
• Use react-icons for iconography and ensure the design is engaging.

Integration with pushkin Packages:

• Update pushkin-client, pushkin-api, and pushkin-worker to support the new functionality.
• Implement database queries in pushkin-worker to calculate percentile ranks.

Implementation Steps:

1. Backend Development:

• tabulateAndPostResults:
  • Modify to accept and process the summary statistic.
  • Ensure it updates the database with the necessary data.
• getFeedback Endpoint:
  • Develop a new endpoint or modify existing logic to fetch percentile rank without database writes.
  • Implement logic to calculate percentile rank based on the summary statistic.

2. Frontend Development:

• Create or update the ResultsPage component to display the percentile rank.
• Use FaUser icons to visually represent the rank.
• Implement dynamic highlighting based on the rank.

3. Integration and Testing:

• Update pushkin-client to handle requests and responses for the new endpoints.
• Ensure pushkin-api and pushkin-worker are updated to support the new logic.
• Test the entire flow from data submission to feedback retrieval.

Acceptance Criteria:

• Users can view their percentile rank based on the summary statistic.
• The system accurately calculates and displays the rank.
• The feature works seamlessly across the frontend and backend, including Docker environments.

Additional Considerations:

• Ensure the system is scalable and can handle multiple experiments and users.
• Consider security and privacy implications when handling user data.
• Provide clear documentation for any new endpoints or modifications.

[github-actions]

👋
Thanks for opening this issue! We will investigate the matter and get back to you as soon as possible. In the meantime, double check that your issue:

• Includes any relevant code snippets and CLI output. If possible, copy/paste the error text rather than screenshotting (although screenshots will also be helpful for some front-end issues). Make sure to include the code that produced the issue within blocks set off by triple backticks (```).
• Includes the version number of each of the following: pushkin-cli, experiment/site templates, Docker Engine, and Node.js. In some cases, it may also be helpful to include the version numbers of other components, e.g. pushkin-api, pushkin-client, pushkin-worker, yarn, jsPsych plugins, etc.
• Links to any existing issues/discussions/PRs which might be relevant.
• Pertains to the interaction between jsPsych and Pushkin (if relevant), rather than jsPsych itself. Otherwise, raise it in jsPsych’s discussions page or issues. A good workflow for developing Pushkin experiments is to first create a working example as a plain jsPsych experiment. If your experiment doesn’t do what you expect in plain jsPsych, your issue isn’t with Pushkin.

While waiting for a response, head on over to our discussions page! And for more information, check out:

• Pushkin Docs
• JavaScript Tutorial
• JsPsych Tutorial

[LuminousYuzu]

Frontend Exploration:

The frontend component to display the percentile rank will be implemented in the ResultsPage within this repository: https://github.com/LuminousYuzu/Pushkin-TestSite. This will include visual feedback and dynamic highlighting of the rank, using react-icons for enhanced UI engagement.

[jessestorbeck]

@LuminousYuzu, eventually you'll want to integrate your results component into the exp-feedback branch. The file you'll want to edit is /src/web page/src/results.js for each of the experiment templates. See the current latest commit on the branch: adcad4d.

[jessestorbeck]

@LuminousYuzu, see the latest commit: b873b1a. I think the next step is creating some end-to-end tests for each experiment's results page. I would look first at the experiment tests that are already there, e.g. templates/experiments/basic/src/e2e/experiment.test.js. Then think about what tests we'd want for a results.test.js. Ideally the tests would be substantially similar if not identical for each of the templates, so it doesn't need to be so granular down to the exact text on the page.