Daily tracking of 40 things about yourself
Over at Kevin Kelly and Gary Wolf's Quantified Self blog ("Tools for knowing your own mind and body") guest blogger Alexandra Carmichael explains how she keeps a record of 40 different things in her life every day, and what she's learned about herself from studying the data.
I track these things about my health and personal patterns every day:She's come to the realization that her mood is much better on days she exercises, and on days when her mood rating is low, she overeats.- sleep (bed time, wake time, sleep quality, naps)
- morning weight
- daily caloric intake (each meal, total calculated at end of day)
- mealtimes- mood (average of 3 positive and 3 negative factors on 0-5 scale)
- day of menstrual cycle
- sex (quantity, quality)
- exercise (duration, type)- supplements I take (time, dosage)
- treatments for vulvodynia (a chronic pain condition)
- pain of administering the vulvodynia treatment I take (0-5)
- vulvodynia-related pain (0-5)
- headache,nausea (0-5)- time spent working, time with kids
- number of nursings and night wakings (I'm a mom)
- weather
- unusual events (text)The mood factors I measure every day are:
1. Happiness
2. Irritability
3. Calmness
4. Sadness
5. Feeling beautiful / self-love
6. Feeling fat / ate too much
I hadn't expected my tracking to unearth such deep, emotionally charged issues. I did expect the optimization which often accompanies tracking, but when striving for an optimized ideal, the question becomes how to decide what "ideal" means. I just don't have an intuitive sense of what the data "should" look like. Are such wild swings in caloric intake normal? What do other people's patterns of mood, sleep, and exercise look like? I'd love to see some kind of comparable, to get some sense of where my patterns fit on the distribution curve. Part of my motivation in sharing my data is to encourage others to do the same. Let's learn from each other!It's fascinating stuff, and it will be even more fascinating when people start sharing this data and analyzing it in various ways. Quantifying Myself
the latest
latest episodes
I made a Facebook app that lets you track happiness (with the key benefit of being polled via Instant Messenger, allowing data to be collected without the user having to be... well... insanely devoted to capturing their data!).
I decided 'happiness' was the important factor and deliberately didn't include other axes. The Facebook integration means I have a hope of being able to cross reference with weather / location data, etc..., in time. Here it is: Mood Tracker.
1. Pass
Not to be a Negative Nancy, but that people feel better when they exercise and overeat when they're depressed is hardly something which we needed hardcore data to discover.
Bahh. . . she didn't monitor the consistency and quality of her bowel movements, that could very well be a key indicator of . . . something.
I'm just starting a similar project at http://mwinkelmann.com/2008/12/the-neurogenesis-experiment/
It's a trial to basically implement anything that has been shown to be beneficial for neurogenesis (neuron growth).
Won't publish my sex life, though.
Are there good tools for logging this kind of information and creating those pretty graphs? I didn't see how she was doing it on the site.
Daytum.com (in private bayta? takes about a week to get an account) is a tool that makes micro-tracking easy.
#3:
There's a difference psychologically between knowing that "people" feel better when they exercise and worse when they over-eat and knowing that *SHE* feels that way by looking at the data.
"- pain of administering the vulvodynia treatment I take"
Eeeh. I'm not squeamish but I could probably do without knowing this about Ms. Carmichael. I understand how useful it could be to her, however; keeping a record of such values (well, mood and other habits, at least) has long been recommended to people with bipolar and other depressive disorders.
You can do similar tracking with DAYTUM; it doesn't necessarily have the health slant but it has a lot of cool features for tracking and displaying "daily data."
iamcantaloupe @ #3 —
From the linked post:
And to expand on what Guysmiley says, there's a difference between intellectually knowing "exercise is good for people" and truly understanding (on a bone-deep level) "when I exercise I feel noticeably better, and when I don't exercise I feel noticeably worse."As a scientist, I'd love to track this kind of data about myself, make plots, draw conclusions.
As someone in recovery from an eating disorder, I've found that I absolutely cannot track calories or weight (or any related quantities) without it setting off really bad stuff for me. I've tried a few times. It never goes well.
I kind of envy people who can. It seems like a good direct method to effect change.
Can we say 'OCD'? I knew we could!
- sex (quantity, quality)
I am sure that metric will be intriguing to her husband and/or boyfriend
And/or girlfriend.
Hey! I was trying to do something like this as well, but I never got around to setting it up. I thought I'd track BMs, as mentioned in post #4, mood, how my arthritis is feeling, breakouts, sleep, weight, as well as some daily weather info.
And yeah, I agree, the revelation about the vulvoick is just a tad TMI.
there's an interesting presumption of causation, and its direction:
"my mood went up significantly on days that I did more exercise"
or do you exercise more when you're in a good mood? i know that when i wake up feeling crappy, i have a hard time motivating to work out...
"on days when my mood was down, I ate a lot more"
or do you feel bad after you've eaten a big lunch? (or is there some external factor that affects both in tandem?)
the next step is to move from quantifying yourself to experimenting on yourself.
naturally, it will have to be a double-blind experiment, so we'll need to come up with a substance for you to eat that makes you feel the same (same size, taste, color, density, etc.) whether it has many calories or few calories. then two groups of scientists can conspire to feed you a low calorie or high calorie substance without introducing any bias...
I suspect I'd need a German girlfriend to do this kind of longitudinal study any justice.
Does she say how much time she spends logging all this data, and then blogging about, er... logging?
Hi, this is Alexandra,
Thanks for all of the comments about my data! I'll try to answer some of your questions:
- I made the charts using Google spreadsheets. CureTogether does have basic charts (we just launched a new personalized home page last night for members where it shows you your latest tracking charts and what's new for your condition(s)), and more fancy analytics will be released over the coming weeks.
- My husband knows I track our sex life (using a simple 0-5 scale), but I'd want to talk it over with him before sharing that data publicly. :)
- Good point, there is inherent bias in measuring yourself, and causality can be fuzzy in such a poorly-defined experiment.
- I spend about 10 minutes a day logging. It took me about 3 hours to prepare the blog post.
This sounds remarkably similar to the "personal informatics" story that ran in the WSJ:
http://online.wsj.com/article/SB122852285532784401.html
I actually did this on a smaller scale for a couple of months and the things it told me about myself were really useful.
[Broadly speaking: (1) I get depressed when I don't get enough sleep; (2) I need a lot of sleep.]