Add weekly recruiting metrics to dashboard

[mbcann01]

• [x] A text box that gives the number of calls made two weeks ago (Mon-Fri)
• [x] A text box that gives the number of calls made last week (Mon-Fri)
• [x] A text box that gives the percent change between two weeks ago and last week.
• [ ] A text box with the highest number of calls made in any week.
• [ ] A text box that gives the number of follow-up interviews scheduled two weeks ago (Mon-Fri)
• [ ] A text box that gives the number of follow-up interviews scheduled last week (Mon-Fri)
• [ ] A text box with the highest number of follow-up interviews scheduled in any week.

[mbcann01]

Hi @nussbaum596, Great job on getting the new text boxes added to the dashboard. Your updates are now added to the main branch (gh-pages). Before we close this issue out, let's do the following:

• Remove the "Ratio of number of scheduled F/U interviews to number of calls by day" graph from the dashboard. Just delete that code.
• Change the Overview tab of the dashboard from a column layout to a row layout.
  • The first row should contain the "Recruiting calls made and follow-up interviews scheduled by date" graph.
  • The second row should contain a grid of all the text boxes that are currently in the first column (i.e., "Total follow-up interviews completed" through "Percent change between two weeks ago and last week").
• Let's add few more new text boxes too (also in description above):
  • A text box with the highest number of calls made in any week.
  • A text box that gives the number of follow-up interviews scheduled two weeks ago (Mon-Fri)
  • A text box that gives the number of follow-up interviews scheduled last week (Mon-Fri)
  • A text box with the highest number of follow-up interviews scheduled in any week.

I've created a new branch for the tasks above. It is issue-29-weekly-metrics. Please use that branch for all tasks above.

After that all looks good, we will close this issue and work on removing the data cleaning code from index.Rmd and use a single analysis data frame for all calculations. single data frame.

Please just let me know if you have any questions.

Thanks!

[mbcann01]

Hi @nussbaum596 ,

Sorry this has taken me so long! I have a couple of requests for you.

1. I went ahead and merged your last pull request (by accident 🤦‍♂️), but I wasn't actually done making changes yet. However, I'm swamped right now and if you could finish making them, that would be super helpful!

2. When I tried to run the code in the last pull request you submitted, it wouldn't work. Specifically, there was a comma missing after line 19 (i.e., uid = keyring::key_list("detect_fm_db_readonly")[1,2]) in data_import.R. This makes me think that you probably weren't able to test the code before you submitted it. If that's true, please make sure you always test the code before submitting it. I think this is a good thing to do no matter who you are working with. I apologize in advance if you did text the code and I'm missing something.

3. For the boxes that give calls per day two weeks ago, calls per day last week, and percent change in calls per day, I made some changes to the code we were using. Specifically, I separated much of the data management code from the code that actually displays the text boxes. Lines 80-107 in index.Rmd are where I moved the data management code to. That makes the code to create the text boxes -- lines 281-307 -- easier to read. It also makes the code faster because we aren't repeating the same calculations multiple times. Please do the same thing for the F/U interviews scheduled two weeks ago, F/U interviews scheduled last week, and percent change in F/U interviews text boxes.

4. I didn't get around to checking the maximum number of calls made in any week box. Please double-check it and make sure it works.

5. In general, let's both try to be really diligent about commenting our code. What it's doing and why.

Please just let me know if you have any questions, and thank you again for all your help!