rikkibarton
commented
11 months ago I received this from a PRC staff today, having the same issue. She found a new workaround (other than the one I listed above) and then that stopped working, too. Here is what she said:
I was working on the TCMHS FY24 T/TA Logic Model August 2023 report this morning and needed to make a change to my MDS numbers. Last month I had discovered a different work around; instead of deleting previously entered numbers, you can work backwards and just change the numbers from the last demographic to the first. I was doing that this morning on “Provide support to school-based youth prevention clubs and groups” and when I got back to the first demographic, “age”, I made the sum equal 56, just like the others. However, when I try to click ‘next’ to move on through the form, I get the error message, “Stats sum must equal 56” and cannot go any further.
I have had two PRCs bring this problem to my attention. They are working on their July reports. If they save and exit, then come back to enter more information (or if another user opens the report to add their month's activities), the sums matching statistic questions for the bottom section won't allow them to change their number. It pops up with a box on the Age question that says "sum must equal ___", which interestingly is showing the NEW number sum they entered, but I'm assuming it does that because it's not equaling the sum for Race, Ethnicity, and Gender questions (which are still the old number until they advance to update it).
For a practical example, in my FY2021 PRC test model, I had entered presentations as the last piece of the logic model report. I said 13 people were trained. I just went back in and tried to add 12 more people to Age, but it popped up that the sum needed to equal 25 (not the original 13). Of course, Race, Ethnicity and Gender are still at 13 because I can't advance and update those numbers.
This will be a recurring problem because as you know, many staff from the PRCs will be entering activities and therefore number for Age, Race, Ethnicity, and Gender.
I found a workaround for now, and that is that I'll tell them to write down the numbers they had previously entered and then delete them backwards (starting with Gender, then Ethnicity, then Race), then add the new numbers to Age and continue forward. This worked for my test logic model, but I did have to save it with blank numbers first and then go back into it. So that's not an acceptable fix, but it'll work for them to get their numbers in this month.