RichardKennaway comments on Quantified Self recommendations - Less Wrong

9 Post author: ciphergoth 18 May 2012 10:16AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (13)

You are viewing a single comment's thread.

Comment author: RichardKennaway 18 May 2012 12:32:05PM 8 points [-]

None of this rises to the level of Quantified Self Geek, but for the last few years I've been measuring a few things, using an ordinary spreadsheet:

  • Weight, every day, same time every day, for about 7 years now. Leaving aside occasional excursions (in either direction) due to ill-health, the long-term trend is flat, but I have noticed a strong relationship between day-to-day weight and how well I feel. Above 125 pounds or below 120 I can tell there's something wrong.

  • My daily exercises, "every" day. I actually average about 5 days out of 7 -- which I only know because I have the record. This I've been doing for three years, ever since I started doing a fixed set of exercises every day. I do other exercise, but this is the only part I record, just to keep me honest. Everything else is as and when.

  • Time I went to bed, time I got up, and how I woke (e.g. alarm clock).

  • Any transient indispositions.

I once tried recording "happiness", but gave up on it, for reasons described in my comments on that thread.

Not quite the same thing, but in another spreadsheet I record every financial transaction on every store of financial value (current account, credit cards, etc.), organised month by month, and project estimates of everything out to at least a year ahead.

I find that just making records of this sort changes what I do, just from paying attention as a side-effect of recording them, and the accessibility of the data. I'm more likely to do my daily exercises if I see that I skipped them the last few days. I will notice if I'm on track to spend £500 on books this month, and ask myself, "Do I really want to direct this much money in that way?" And it's always worth knowing (years of FY money) = (total financial worth)/(total annual expenditure).

Comment author: Rhwawn 19 May 2012 12:04:20AM 0 points [-]

Have you considered randomizing your exercises so you could begin drawing causal inferences?

Comment author: RichardKennaway 19 May 2012 06:33:53AM 1 point [-]

Yes, but I haven't, because what would I measure? I seem to have "more energy" from doing these exercises, but I expect that rating "energy" on a 5-point scale would be as impossible as rating "happiness". I could instead keep a log of things done (which is, after all, the point of having "energy"). Come to think of it, that might be useful anyway.

Comment author: Rhwawn 01 July 2012 03:44:38PM 0 points [-]

I think you underestimate simple self-ratings. You could just do those, and yes, there are automated ways. For example, you could turn on a Web browser plugin like RescueTime but disable any blocking functionality - so it's just tracking time spent. Randomize intervention X for a few months, pull the RescueTime logs, and voila! A (non-blind) randomized experiment.