Adventures in Data Visualization (or: Why Brad is Like That Sometimes)

[usernamenumber]

For the last year I've been keeping a log of when I sleep. It hasn't been perfect, but I've managed to note roughly when I went to bed (by which I mean was about to lose conciousness) and when I got up, including any fits of insomnia in between for 302 of the last 365 days, and have cobbled together some scripting that allows me to graph the results.



One need only look at the records of when I go to sleep to infer that I have problems winding down at the end of the day. I figure it would be healthier by most estimates if the majority of the dots were at the top, in the 10p-12a range, but alas, my tendency to go to bed after midnight is even more pronounced than I thought it was.



This might not be a big problem if my waking times were as erratic as my bedtimes, but as you can see I sleep past 10am just a few times a year, and since I started teaching regularly, the majority of my wakup times are clustered around 7:30, and rarely later than 9 even on weekends, regardless of when I went to bed. It's just how my body works: past a certain point something just starts whispering in my ear "Hey! Get up! There's stuff you could be doing!! Sleep is not stuff, so you should stop it!"



But the real kicker is when I graph the number of hours actually slept. It's more erratic than a comparison of the bed/wake graphs might suggest because sometimes I'll wake up and have stretches of insomnia, which aren't reflected there. In any case, I can't help but laugh to look at this graph. Man, it's like a beautiful, starry sky it's so all over the place. Also, I have to say I suspect there are some errors in the interpretation of the data, because while it has happened, I don't think there have been that many days with 2 or fewer hours of sleep.



Turns out I average 6.7 hours of sleep per night. Awesome.



The thing is, I don't see this changing my lifestyle much. The dilemma has always been that in order to get more sleep, I'd have to either drop something else from my life, which I'm not really willing to do, or have greater control than I've historically had over things like like insomnia/hyperactivity.

That said, I've noticed that with the decrease in stress at my job, I've needed less cope to get by. Whaddayaknow. Will try to do a more comprehensive State of The Me soon. For now: Life is pretty good on the whole, though I'm sick and tired (literally, not figuratively) at the moment. If me from just five years ago could have seen me now... wow, it would be a trip, I think. Things are different. Sometimes scary, but mostly I'm enjoying the ride.

Comments

[londo.livejournal.com]

Y'know, I've considered doing this before, and I'm pretty sure your data here is a very very close match to what my sleep patterns are like when I'm 9-to-5ing it. (Or, as some people like to call it, "2010".)

[heiligekuh.livejournal.com]

In the first chart, how do the midnight times relate to each other? If I see a dot slightly before midnight on Column N (near the top), then dots around 1:00 and 4:00 in Column N+1 (near the bottom), does that represent a bed/wake/bed cycle?

I think confusion around 24 hour days and "sleep days" are probably the source of your sub 3 hour sleeps. Those pngs are really small, but it looks like early Nov-09 might have had some days where you slept early-ish, woke up shortly after midnight and then didn't get to bed until after midnight the next night.

I'd extend that question to your daily average as well. With the iregularity of your insomnia, adding up sleep in a midnight-midnight 24 hour period doesn't seem like a good fit. I'd love to see if the data changes if you just switch the day limits to noon-noon.

FOIA! Release your sleep to us!

[heiligekuh.livejournal.com]

Curse you auditory processing. You made me misspell an acronym!

[faerieboots.livejournal.com]

Can you change the first graph so that the hours start around 8PM and end around 6PM? That would make your graph less visually confusing, and possibly give you a better idea of how wide the band actually is.

Also, what's up with the 8AM bedtimes? It looks like there are a number of them.

(OK presentation prep now I swear)

[usernamenumber.livejournal.com]

Haven't figured out how to change the graph, unfortunately. But investigating the 8am bedtimes led to me finding a bug (which I unfortunately haven't yet fixed). Oops! It looks like if I, say, slept from 00:00-05:00, did some stuff, then went back to bed from 07:00-09:00, it only registers the last, two-hour chunk. Will have to fix tomorrow.

[medeaschild.livejournal.com]

You may be familiar with everything in here, or you may not, but it is really worth a read.

http://lifehacker.com/5548150/how-to-reboot-your-sleep-cycle-and-get-the-rest-you-deserve

Also, I don't know what you are doing for meds, but my psychopharm had me look into the research about omega 3s and ADHD/depression, and so now I'm taking a crap ton of them (3 grams. GRAMS) a day, but they have really, really helped smooth out my energy levels. I have very little insomnia anymore and I rarely drop to useless in the mid afternoon.

Which means I can go to bed at a decent hour and be a productiove member of society when I'm awake. Its a world of awesome.