No Activity

[Neil Davilo]

Hello All, 

Recently, we've noticed a spike in counts labelled "No Activity." We aren't sure how they're tabulated, why they show up, or what causes them to spike. On some days we'll get 0 No Activity counts, and then the next day a single location will show 200+ instances of No Activity. 

Has anyone experienced this? It's adding a lot of extraneous data for us to wade through, so we'd really like to get rid of the No Activity if at all possible. Attached is a screenshot of the Spreadsheet data for our Family Study Room location, a generally low-traffic area.

Thanks,

Neil Davilo

neild...@suu.edu

435-865-8240

[Fred Seaton]

Have you looked at the weblogs for the server to see if there is any unusual activity corresponding with those dates? 

 

--
You received this message because you are subscribed to the Google Groups "Suma: mobile library assessment toolkit" group.
To unsubscribe from this group and stop receiving emails from it, send an email to suma-mobile-library-asses...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/suma-mobile-library-assessment-toolkit/30649f92-d74c-4c0d-9e0c-9af90d64f0e8n%40googlegroups.com.

[Jason Casden]

Could you also check the "Sessions Summary" report for any unusual sessions? Do you know if your Suma client is publicly accessible?

Thanks,

Jason

To view this discussion visit https://groups.google.com/d/msgid/suma-mobile-library-assessment-toolkit/PH8PR11MB6831A7811A53A76C049A30C5D9AE2%40PH8PR11MB6831.namprd11.prod.outlook.com.

[Neil Davilo]

I attached a snippet of the Sessions Summary that coincides with the image I attached earlier. The problem with the Sessions Summary is that it doesn't subdivide the counts into the sub-sections, so I'm not sure if anything is unusual. 

I can say that our client is not publicly available. The server it's on is locally hosted and it has an active firewall. 

On Mon, Apr 7, 2025 at 2:15 PM Jason Casden <cas...@gmail.com> wrote:

Could you also check the "Sessions Summary" report for any unusual sessions? Do you know if your Suma client is publicly accessible? Thanks, Jason On Thu, Apr 3, 2025 at 7: 11 PM Fred Seaton <fseaton@ uchicago. edu> wrote: Have

To view this discussion visit https://groups.google.com/d/msgid/suma-mobile-library-assessment-toolkit/CADwVd-OE%2B%3DQkipC%3DNH3xCPV9S9s-zp6PjUyMNwR0ox7QYpJWzQ%40mail.gmail.com.

[Jason Casden]

You should be able to compare what you see with what you know about data collection. For example, are hourly counts around lunchtime of between 70 and 142 roughly consistent with what people are seeing in real life? Are headcounts being collected that often? etc. 

Jason

To view this discussion visit https://groups.google.com/d/msgid/suma-mobile-library-assessment-toolkit/CALJ211VbJQFVTE%2B%2BbaUrE86GHeZCa0woWnELnEk8JhjHBwgiUg%40mail.gmail.com.

[Jason Casden]

If these totals seem roughly accurate, I think you may actually have normal counts here. If you use the Time Series report, there are some nice interactive tools for browsing the locations and activities. Generally, a count in Suma has both a location and zero or more optional "activities" (just labels, really). In a headcount initiative, generally, no activities will be configured in the admin tool, so every single count is a "no activity" count. These counts also have locations, though, so you wouldn't combine the location and "no activity" numbers. For example:

"""

I have a headcount initiative with no activities configured in the admin tool. I walk into room 1, I tap on "room 1" in the location tree, and a "zero" count is automatically created with no activities. I then hit the count button five times. That "zero" count has been replaced with 5 individual "no activity" counts in room 1.

I travel to room 2, and the process repeats, but for that location.

I submit the session. A raw report for that session should show:

5 counts in room 1

5 counts in room 2

10 "No activity counts" (these are the same of the 5 and 5 above).

So, 10 people were counted. None of them were tagged with any activities.

"""

It gets a little weird when you tap on a location but don't count anyone, since that creates a "zero count" for that location as evidence that the location was visited but no one was counted (useful for calculating headcount averages). You can include or exclude those in the Time Series report. Still, the maximum number of zero counts you could have in one session is equal to the number of countable locations (so probably not hundreds).

It's a bit awkward because Suma data doesn't neatly map to a tabular presentation. These activity columns are there to help you answer questions like "how many of the 100 counts in this session involved eating? and how many of them were in groups?" and be able to answer 58 and 65, even though that's more than 100 people, since people can eat food in groups. That's why we've tried to make the interactive reporting tools as useful as possible.

I hope this helps.

Jason

[Neil Davilo]

I think I see what you mean. So "Activity" is a tag that helps us describe behavior, but since we are opting not to use those behavior tags every count is attributed a "No Activity" designation?

That is helpful, thank you for giving the example and additional information.

To view this discussion visit https://groups.google.com/d/msgid/suma-mobile-library-assessment-toolkit/CADwVd-PLP6VGNaTiBzQR3TLvA0H2_1A241SibByGFbpsOvhyNQ%40mail.gmail.com.