Personal data fascinates me. I collect data about myself mostly as a way to journal and document the present so that I can look back on it later – similar to how someone else might flip through an old photo album.
In just about every interview I’ve read with Nicholas Felton, author of several personal annual reports, he’s asked how the data, or rather the information from that data, has changed his behavior. For the most part, it doesn’t. It’s more of an interesting view into the past year for him.
However, there are plenty of others who collect data in an effort to change their behavior in some way. They might be trying to lose weight or stay more disciplined with their exercise regimen.
So if you collect data about yourself, whether it be an automated system or with pen and paper, why do you do it? How long have the you been doing it? What do you track? Have you found anything interesting or surprising from your data?
If you don’t collect data, what’s holding you back?
Leave your thoughts in the comments below.
I am diabetic and therefore collect my sugar levels everyday (like 6 times per day). This helps me to “understand” my metabolism, my diet and my stress levels… It is not linear, and there are a lot of trial & error but I have my whole life to learn…
At one time, over a period of a couple of months, I was using a spreadsheet to track everything I ate, calories burned through exercise and daily weight in an effort to reach a target weight for competition. I was using it to predict how long it would take to reach a goal weight and evaluate whether what I was doing was working or not. I found it surprising that there was a difference in my perception of how much food I required and how much food I actually required. Also surprising was that it helped me develop some habits around food that stuck with me after I stopped tracking.
1. I don’t collect data on myself because data is theory-laden, which means I have to maintain a consistent theory of myself in order for the data to gain meaning and retain meaning.
My self-model is gooey. I like it that way. I don’t want it to harden.
2. If I capture the data consciously, that means I’m thinking about the data while I’m living my life. Data capture would therefore be intruding on my life experience. No thank you. I don’t want “data” to dominate my life. I don’t want to live a second-order life. First-order, please.
Collecting data about one’s self is just the geeky equivalent of writing a diary. Nobody could say it’s a bad thing to sit and write down your thoughts each day. Collecting data is one other way to do it.
My own personal data collection? I keep a record of whether it rains or not when I commute to work:
(note – the viz has not been updated since March, since when it’s been absurdly dry; I will update it soon with the newest data…)
Why? To silence the people who say “I don’t cycle because it rains too much”
@Jame Bach – I appreciate your sentiment, but what would your opinion be of someone writing a diary – is that an example of data-driven reflection on life?
Currently I collect:
– running and biking stats (day, distance, time)
– weigh lifting stats (day, weight, type of exercice, nb)
– books I read, games I play, Movies & Tv episodes (day of start, day of finish)
I plan to collect (next year):
– stats from computer usage
– stats from productivity (todolist, projects, etc)
– stats from my business
– stats from nutrition (for some months)
I think that important things to consider when collecting are:
– it’s better if it’s automatic
– it’s better if it’s positive
– it’s better if it’s objective
– it shouldn’t be a burden to report, we don’t want to be addicted to that
– it’s ok if we miss some reporting
– it has to mean something or help us in some way
@bob I’m working on a piece of software for tracking and analyzing weight-lifting/calisthetics type data. I’d love to hear any features you’d love to have in such a program.
I use CalorieKing for tracking what I eat and a fitbit for daily activity.
oops, my email is [email protected]. I see now Paddy also uses a fitbit below. I’m mostly just using mine for my steps. Oh, and I also use a Polar F11 to track the calorie expenditure and effort during my workouts. I’m planning on starting a blog on tracking health/exercise data, but work, family and software development seem to have filled up my time for now. BTW I love this blog. Cheers.
I started collecting personal data when I embarked on a weight-loss adventure of my own making, and began maintaining a detailed food diary. I ended up learning a TON through that process, really woke up to what I had been eating for years, and what that was getting me. This was definitely an area in which tracking affected the results, in a totally positive way for me. Dropped 80 lbs and developed awareness and habits that I simply did not have before, and likely still wouldn’t, if not for the data closing that huge information gap I had.
I’ve since moved on to collecting data on workouts (via heart-rate monitor and annotations), and I’ve picked up a fitbit (http://www.fitbit.com) to track step counts/distance, sleep data, and to integrate all of that with the food intake data. This is a work in progress, but the fitbit site has some decent pre-configured ways to present data and compare your performance with others. Neat stuff there, and a premium package is now on offer for real data geeks…
I’m definitely interested in as much automated collection as possible, but if I’m motivated (as I was with the food diary) and it’s not grossly inconvenient (I sit at a computer 8+ hrs a day, so logging is not a lot of extra work), then I’ll do some extra work to get results I’m interested in.
I don’t see this behaviour as artificial or intrusive at all, but rather as an often surprising exercise in self-discovery. This one is guided by data as opposed to (strictly) intuition or emotions, but it is about self-discovery nevertheless (for me, anyway).
I collect my blood glucose level every 5 minutes through a continuous glucose monitor stuck in my gut.
I log carbs, protein and fat intake.
I cycle, so I log routes, miles, pace, cadence, grade and heart rate.
I am a type 1 diabetic. All data I can collect, analyze and take action on improves my health and prolongs my life.
And it’s a huge pain in the ass.
I’ve been posting a daily mugshot of myself on a website (dailymugshot.com) for over 400 days now, just for the hell of it really & a bit of fun :)
– Runninng (auto through Nike+)
– Heart Rate while Running (manual in spreadsheet, through Timex Wristwatch and chest band)
– Weight / Fat Mass / Lean Mass / BMI (daily, auto through Withings scale and site)
– Waist/Hip Circumference (daily, manual in spreadsheet)
– Pushups, Pullups, Situps, Squats (daily, manual in spreadsheet)
– Blood Pressure and Pulse (daily, manual in spreadsheet)
I track these things because I can have an overview of how my health is doing and act accordingly if necessary to optimize it. I don’t feel wasting time or dominated by this.
-Income / Expenses (manual through Quicken Essentials)
– Books Read (manual, Google Books)
– News/Blog Posts Read (Google Reader)
– Movies Watched (manual, iMDB)
– Places I’ve Been (FourSquare)
I track finances as a way to improve wealth knowing where I’m spending more, how often, etc.
I think these other trackings are fun to track and enhance my memory.
Pingback: Life-tracking | Katherine
I track date, distance, location, and time in a running log. If there was something notable about the run, I write that down too. I have a Polar watch that stores the last several runs, but then I copy them out longhand into a journal.
The only hard core data collection I’ve done has been thru your.flowingdata.com, where I tracked everything I consumed for three weeks. I kept a google spreadsheet which I updated with nutrition labels for everything I consumed. If I didn’t have a nutrition label (usually I could find one) I didn’t eat it.
That’s as far as I’ve gotten. Next step is to merge the your.flowingdata tsv with the google doc – I used the same id in both the logging and the google doc so I can do this.
Ultimate goal: To create a sort of “game”, where I understand all of the nutrition I consume and all of the energy I output. Essentially, there are certain levels of nutrition that I can expect to require for, say, an evening workout… or even more so for something ridiculous, like a 9 mile run.
If I know how I expect to expend the energy I receive from the food I eat, then I should be able to make very informed decisions about what I need to eat – i.e., know where my nutrition is, use that to decide what to eat for lunch (for example) and be completely prepared for an intense pool workout… or, no workout, which requires a difference balance of nutrition.
Hypothetically, whenever I’m hungry or thirsty, etc., I can reference a sort of “recommended nutrition” for that moment based on how I plan to use the energy, and find foods that fit that profile. The matching foods to nutrition recommendations is right along the lines of my work for Booklamp.org
I’m a ways from making this happen. I hope I don’t let it just fizzle out. I really like the idea of personal data collection… interesting that Feltron mentions that it doesn’t change his personal behavior much.
Additionally, I write my twitter posts (@countrynole) with some audience in mind, as we all do, but I also collect all of my posts via Google Reader with the intention of using it as a perspective on my own history. Booklamp does not really deal with text passages that short (right now), but there’s also some interesting thoughts about analyzing my historical tweets en masse. That’s a large project for a day way down the line.
Thanks for the post. I’m a fan of data collection in general, but if we all collected personal data and had a way to intelligently parse that, we’d all be better off.
I track how I spend most of my time throughout the day in Microsoft Outlook’s calendar and my Google Calendar, including travel time, productive vs. unproductive work, meetings, etc. I’ve been doing it since January 2010, and it helps me perform the following:
– Understand where I spent my time. This has the benefit of being able to report on exactly how long it takes me (on average) to perform a particular work task, and to plan future work tasks accordingly.
– Track how much time is wasted. Shockingly, I’ve never broken the “50% of my time being wasted per day, on average” barrier. I believe this is due to the incredible amount of task switching required of me in cubicle land, and that I tend to spend WAY too much time doing reporting activities rather than actually productive work. Sure, my reporting activities help management make some decisions, but most of my reporting is still not automated and an effort that usually doesn’t change management’s decisions in the end anyway.
– Budget in more time throughout my days to spend on just thinking, creating ideas for new ways of doing things, and to just generally structure more time that will lead to more productivity rather than more spinning my wheels.
I don’t keep data on myself because I don’t want to set it up.
If there was a site that I could customize what I want to track then get an email every day that told me to fill out my personal data survey, I would.
I track my sexlife at bedposted.com (duration, intensity, positions).
Knowing how often and how each activity was, is exciting (pun intended) and makes me feel wanting more, which is good.
You can customize an online form using Google Spreadsheets with any data you want.
About the notifications, just set some calendar to send you an email everyday.
I take snapshots of my tag cloud in Google Reader every 6 months or so. It’s neat to see how my interests shift.
Hello, Nathan. I track “mood”, “costs”, “bloging” (idea – 1 point, draft – 2 point, post – 3 point), listen podcast (counter).
@Andy Cotgreave, thanks for link. Tableau it’s good, but Your.FlowingData amazing!
I currently track steps/calories with fitbit (2 mo) and at various points in the past have tracked spending information with spend lite for iphone (4 mo) and recurring tasks such as watering plants, exercise, etc. with an app i wrote (4 mo).
different types of tracking have different utility for me. spend tracking directly affected my spending behavior — it was pretty easy to rebudget based on how i was actually spending. in contrast, fitbit simply makes me feel better about exercising because i can see a record of it afterward. task tracking helped me turn tasks into habits.
I have kept my personal expences for the last 5.5years and fund it very usefull to know how i spend my money. Info is very detail.
I usually just collect data on my spending to make sure I’m sticking to budgets and meeting financial goals. But I was actually inspired by The Feltron Report recently. I’ve just started a month-long data collection of how I spend my time, how much I read, and what I eat. It would be fun to have the data (although a bitch to collect and log) to play with and use my analytical/data visualization skills, but I’m sure it’ll also lead to some interesting findings about myself and areas of success and improvement.
Oops: I forgot to add I asked this a while back: Related: Attention Data Hounds: What Personal Data Are You Tracking? – http://matthewcornell.org/2009/06/attention-data-hounds-what-personal-data-are-you-tracking.html
[I don’t think my first try went through.] That is the central question and the whole point – to gain insight, change behavior, and ultimately be happier. The “Tool Trap” (focusing on tools, rather than purpose) is tempting, esp. to us early-adopting self-tracking pron-loving geeks (myself included!)
I think what’s missing in the data-tracking movement is a cohesive philosophy of life that puts tracking in a larger perspective. In our Think, Try, Learn work, that philosophy is the scientific method, with the idea being treating everything in our lives as a kind of experiment. So rather than a data-driven life, I think we should be thinking of it as an *experiment*-driven one. That way people have a process and a goal, not just an activity. I tried to express this at http://www.matthewcornell.org/2010/06/the-experiment-driven-life.html. Still working on it…
Re your questions: I’ve been tracking decisions I make and lessons I learn for about four years. I wrote about these: A Key To Continuous Learning: Keep A Decision Log (http://matthewcornell.org/blog/2007/04/key-to-continuous-learning-keep.html) and Some Thoughts From Tracking “lessons Learned” For A Year (http://matthewcornell.org/blog/2006/12/some-thoughts-from-tracking-lessons.html).
I’ve started more vigorous tracking in our TTL tool, Edison, the experimenter’s journal. It’s our first attempt to support people thinking of life-as-experiment. It’s at http://edison.thinktrylearn.com/ if anyone wants to check it out. We don’t have a data layer yet – we have to use one of the many tracking tools available (hey – one user is using yfd – http://edison.thinktrylearn.com/experiments/show/163), but I think there might be room for a tool that puts data tracking in the scientific perspective. The most common topics there are health-related, though we have productivity ones too. My focus is on health, along with developing my TTL work (writing the book and developing Edison), both of which are tasty meta-experiments.
Re: holding me back, I think tool usability is still being fleshed out. I recently reviewed all the iPhone apps I could find, but nothing has blown me away yet. (Unfortunately, the one that supports yfd – Your Flowing Data Uploader- only works on iOS 4.) Capturing data must be extremely fast, or compliance drops way off. Surprises: Well, for sleep there are so many variables that it’s hard to do good science. Another is I found that cutting out one of my mood disorder meds made resulted in no change – a big and welcome surprise.
I search the internet looking for references to myself. It sounds vain, but I have learned a great deal about where my online activity ends up, plus I am learning a lot about how internet applications use personal data.
Pingback: Do you collect data about yourself? « ScienceRoll
Pingback: Do you collect data about yourself? | Science Report | Biology News, Economics News, Computer Science News, Mathematics News, Physics News, Psychology News
This is an awesome idea, I’d be interested in tracking proper psychometrics. Then you could correlate these with life events; a change in diet, excercise, job, relationship – may seem like pretty obvious correlations, i.e. I just got a great job now I’m happier than when I was destitute, but if you had more detailed indicators of outlook, values and mental states you might find some pretty interesting things (might be difficult it is to get objective measures – any suggestions of decent metrics?). Make your own income-contentment curves and sort out once and for all whether money buys your happiness!
Unfortunately, I’ve never attempted to collect data about myself, other than my elementary attempt of keeping a journal once when I was a kid, and it was ultimately derailed once my sister quoted me about it. But that is an interesting proposition, especially if they’re just vague representations just like the image above.
Pingback: Why People Collect Data | Quantified Self