Also tracked weight this year
Guess when I stopped playing squash with @ashwinvish_pic.twitter.com/mh8qAzu1wO
Show this thread
I've been doing some pretty consistent mood tracking (about >4x/day) with @hellodaylio this year
Did a quick regression and found these correlations with mood. Interestingly very little time-dependence (eg, travel >1hr prior has no effect).
Suggestions for other analysis?pic.twitter.com/bxv6KAkpIz
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Time of day? Day of the week?
-
yeah I should try that... I wish I knew what time I woke up + went to bed
-
turns out I really hate thursdays
-
and strong positive correlation with time and mood
End of conversation
New conversation -
-
-
this is so cool! i want to try this.
-
it's super easy actually! daylio is a pretty intuitive app, easily the best tracker I've found. Can export to CSV but maybe they'll add a regression feature in their 'reports' one of these days...
-
Surprised there is no history dependence of your mood
-
Yeah... how does that fit with
@yaelniv work? What dimensions of mood are you reporting? -
I'm only reporting valence (awful, bad, meh, good, rad) and minimum time between logs is ~1 hour. Lack of history dependence could be: - timescale - strong activity-dependence - self-regulation (I'm hunger and sad; therefore I eat + okay next check-in) - ?
-
I hadn't seen Yael's work on this before... I'm not sure how to integrate her theory about learning into my data (I'll have to read more, maybe) maybe I can take some emotions for tagging from
@DianaTamir@Mark_A_Thornton? but I'm worried about making this overwhelmingpic.twitter.com/rb3oDulUKO
End of conversation
New conversation -
-
-
What is your problem with music?!And who's in a better mood cleaning than playing the piano?? I find these results questionable plz explain thanks bye ;)
-
I mostly forgot to log piano and music... I thought about removing them. So piano I only have 3-4x where I was trying (and failing) to destress
-
I don't even know what I meant by music! Maybe going to gigs? I've only been to...one this year?
End of conversation
New conversation -
-
-
How rapid are mood changes (spectral analysis of mood timeseries)? Is there bias in the features, i.e do you only mark "friends" when you've been around friends and enjoyed it? If you're around friends and it was neutral, maybe you didn't think to mark it as a salient feature
-
I thought mood would be less responsive to stimuli but, apparently, I was wrong. changes very fast... here timescale is ~1 hour between logs and I suspect autocorrelation would die off very quickly if controlled correctly cc
@Mark_A_Thornton -
but I think mood is mostly stable during the day because not much changes during the day... slight upward trend from morning to evening though
-
Yeah, we've found pretty short half-lives (nearly all < 1hr) for most emotions when analyzing large experience-sampling data sets (though temporal resolution of the sampling means it's an extrapolation).
-
That said, although the amount of autocorrelation drops off pretty quickly, the *structure* in the autocorrelation (i.e., which transitions are likely) remains highly consistent over much longer time periods (e.g., days).
End of conversation
New conversation -
-
-
Nice! How long have you been doing this? Also could you give more details details of how you track?
-
since new year's last year. The Daylio app pops up with a "how are you feeling" and "what have you done" (little images) a few times per day
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.