On certain kinds of pressure-heavy days, I’ve tended to ignore TagTime entirely. This makes my data imprecise and very lopsided. On the most crazy days, my data isn’t recorded, which means there’s a pretty big bias in the data, and I can’t really even track the impact of these pressure-heavy days. I have a plan for moving forward, but I’d like to backfill some of this, without it being a huge project since, as always, it’s happened around a very busy time.
Any suggestions for how people have caught up on TagTime backlogs/backfilling in the past? Have you just chosen to abandon that data, or do you try to piece things back together? And how?
i think you just have to guess. write down a plausible schedule without looking at the ping times, like “11pm-7am: sleep; 9am-5pm: work, but with 3 hours of distraction so call it 9am-2pm work, etc etc” once you’ve written that all down, go back and back-fill all the pings as if the schedule you guessed was literally and exactly true
I’ve been wondering about this also… Maybe another way is to tag the missed data as eg “busyday” or “vacation” and then you’d get estimates of how much of your time is spent on those types of days
On a related note how to tag when I missed a few pings just because I was away from my phone or didn’t hear it? I’ve been trying to recall what I was doing and if I’m pretty certain then I tag what I recall and if I can’t remember then I tag “unk”…
But I was thinking what if I tagged all the missed pings that I don’t recall (the “unk”) with what I’m doing when I’m answering them, would this also eventually average out to the right values, and I wouldn’t need to try to remember?
Not unless the time you spend answering them has the same distribution as the time you missed. Like anything you do more often when away from your phone will be underrepresented.
I think I’m with @zedmango here. You have to go by when you heard the ping ping and if you didn’t hear it ping then going by when you noticed it later would introduce bias.