Unlike most university science, DIY Science can actually make you happier, right here and now.
Really? Even if I’m already in the libertarian-atheist-countercultural-hacker-intellectual cluster? I checked two statistics in high school...car crashes were high enough that I chose to drive conservatively, and STI rates were high enough that I chose to make a big deal about condoms, although not so high that I took arguments from STDs for abstinence or monogamy seriously. Since then, I just haven’t felt the need to go look at the data as I make personal life choices. I feel like I’m surrounded, socially, by the kind of weirdos who also looked up a couple of statistics in high school, and together we all bring each other’s factual assumptions about lifestyle risks and rewards into something approaching the reality zone.
There are, of course, some questions to which I don’t know the answer and that would interest me because the answer could help me make choices. But these questions tend to be narrow enough that there’s no “Forbes 100” list to use as an objective reference—I have to go and compile the data myself, which is both more annoying and less reliable. E.g., I don’t really care how one becomes a top-earning CEO, because it involves, among other things, decades of sacrifices that I’m not willing to make. I do care about how one becomes a successful solo practitioner in the field of California consumer law, but they don’t exactly have databases about that...just a couple of pages scattered here and there, which of course I’m reading, for all the good it will do me. Likewise, I don’t really care what the odds are that I die of some horrible virus—I agree with you that at my age (10 − 30) they’re quite low unless somebody bio-engineers a plague. I am interested in the success rate for patellar smoothing surgery, but it’s a technique, not a drug, so it doesn’t have publicly scrutinized, large-n trials, and PubMed rarely goes into enough detail to let me assess whether, e.g., post-surgery functionality was self-reported (in which case it was probably driven by the placebo effect and cognitive dissonance) or was actually measured with objective performance tests.
Probably if I did enough research I could get a better estimate than I have now on some of these narrower questions. Maybe I will someday if one of the questions becomes important or if I find myself with spare time. I think the real science deficit, though, is not in our personal lives, but in the calculations of policymakers and professionals. Atul Gawande’s article on Dr. Jeffrey Brenner in the 1/24/2011 New Yorker is a fascinating case study in how applying science to your day job can help you do it more effectively and at a fraction of the cost.
I do care about how one becomes a successful solo practitioner in the field of California consumer law, but they don’t exactly have databases about that.
Here’s what I do:
Look at the job type you want. Look at professional websites of the people who have the jobs you want. Look at their CV’s and see how they got there. If any of that info is expressible in a quantitative form (e.g. percent who went to top ten law schools) tabulate that.
You might notice “Oh, wait, most people who have the job I want have background X that I don’t have!” (Different college major or whatever.) That might be evidence that you can’t get that job; but before you start worrying, send someone an email and ask “How likely is it for me to get a job like yours with background Y instead of X?” It may be that your background is unusual but not a handicap.
Is it less rigorous than a scientific study? You bet. Is it better than nothing? Much.
If you have access to the attention of lots of professionals, a homemade poll can be very illuminating even if it’s informal. For example, this survey about how novelists get published is more informative than most “how to be a writer” advice out there.
It would also be worthwhile to look at people who did those things and see how they ended up i.e., look from the other side.
For example if you look at rock musicians you are likely to find they neglected their studies and focused entirely on their music. But this seems to be a strategy with a pretty low expectation and very high variance in outcomes.
Really? Even if I’m already in the libertarian-atheist-countercultural-hacker-intellectual cluster? I checked two statistics in high school...car crashes were high enough that I chose to drive conservatively, and STI rates were high enough that I chose to make a big deal about condoms, although not so high that I took arguments from STDs for abstinence or monogamy seriously. Since then, I just haven’t felt the need to go look at the data as I make personal life choices. I feel like I’m surrounded, socially, by the kind of weirdos who also looked up a couple of statistics in high school, and together we all bring each other’s factual assumptions about lifestyle risks and rewards into something approaching the reality zone.
There are, of course, some questions to which I don’t know the answer and that would interest me because the answer could help me make choices. But these questions tend to be narrow enough that there’s no “Forbes 100” list to use as an objective reference—I have to go and compile the data myself, which is both more annoying and less reliable. E.g., I don’t really care how one becomes a top-earning CEO, because it involves, among other things, decades of sacrifices that I’m not willing to make. I do care about how one becomes a successful solo practitioner in the field of California consumer law, but they don’t exactly have databases about that...just a couple of pages scattered here and there, which of course I’m reading, for all the good it will do me. Likewise, I don’t really care what the odds are that I die of some horrible virus—I agree with you that at my age (10 − 30) they’re quite low unless somebody bio-engineers a plague. I am interested in the success rate for patellar smoothing surgery, but it’s a technique, not a drug, so it doesn’t have publicly scrutinized, large-n trials, and PubMed rarely goes into enough detail to let me assess whether, e.g., post-surgery functionality was self-reported (in which case it was probably driven by the placebo effect and cognitive dissonance) or was actually measured with objective performance tests.
Probably if I did enough research I could get a better estimate than I have now on some of these narrower questions. Maybe I will someday if one of the questions becomes important or if I find myself with spare time. I think the real science deficit, though, is not in our personal lives, but in the calculations of policymakers and professionals. Atul Gawande’s article on Dr. Jeffrey Brenner in the 1/24/2011 New Yorker is a fascinating case study in how applying science to your day job can help you do it more effectively and at a fraction of the cost.
Here’s what I do:
Look at the job type you want. Look at professional websites of the people who have the jobs you want. Look at their CV’s and see how they got there. If any of that info is expressible in a quantitative form (e.g. percent who went to top ten law schools) tabulate that.
You might notice “Oh, wait, most people who have the job I want have background X that I don’t have!” (Different college major or whatever.) That might be evidence that you can’t get that job; but before you start worrying, send someone an email and ask “How likely is it for me to get a job like yours with background Y instead of X?” It may be that your background is unusual but not a handicap.
Is it less rigorous than a scientific study? You bet. Is it better than nothing? Much.
If you have access to the attention of lots of professionals, a homemade poll can be very illuminating even if it’s informal. For example, this survey about how novelists get published is more informative than most “how to be a writer” advice out there.
It would also be worthwhile to look at people who did those things and see how they ended up i.e., look from the other side.
For example if you look at rock musicians you are likely to find they neglected their studies and focused entirely on their music. But this seems to be a strategy with a pretty low expectation and very high variance in outcomes.
How do you practically do that?