Lies, Damn Lies & Damn Internets
I've been buried under numbers and Power Point all day and right now I'd really like to write about a Jim Lindgren post on a study published by David Blanchower and Andrew Oswald. The study deals with the relationships between marriage and money and sex and happiness. Not exactly thin stuff. Unfortunately, I can't seem to pull up the post anymore (it was on www.volokh.com, which seems to be down). But here's Ann Althouse linking to it. And I guess I'll just go from memory.
Won't be that difficult to go from memory, actually, because I really had just one thought as I read the post: this kind of research positively screams for an ethnographic element. As Althouse says, the relationships between entities and emotions as enormous as the ones in question are so complicated that it's nearly impossible to untangle them. To state definitely that, say, happiness is positively correlated to marriage when money's in the picture, or that study participants with more money had more sex might be true, but it misses the point.
Add a layer here: a little explanation of why. Take these numbers and hypothesize something. Then test it by going deep with a smallish sample. Interview them, have them keep diaries, take photo essays. Get into that data. Then, these numbers will be really valuable.
Now, I'll admit that I haven't read the original paper that Lindgren takes to task. But I've read enough market research studies to know that a little extra depth goes a long way. That a miniscule collection of pictures really does do more than a thousand numbers could, even if those numbers are attached to attribute statements and even if they have the ultimate in statistical significance.
And I don't mean to discount the value of quantitative data. It's great. But when you're dealing with happiness, shouldn't you go a little deeper?
In my former job, I was a part of an ethnographic project of sorts that dealt with the concept of happiness. Throughout 2003, people living all over the world sent us photos with short explanations that defined "happiness" for them. During the year, we selected the best photos and analyzed them, to understand what happiness meant across and within cultures. At the end of the year, we looked at our previous analysis and the entire body of photos from a global point of view. The resulting analysis session was heavy, to say the least, but it was really interesting. And we did come up with some interesting conclusions (the final report is here).
I can only imagine what conclusions and great ideas (that might not be "conclusions" per se) could result from heavy analysis of Blanchower and Oswald's numbers PLUS a solid ethnography exploring the relationships between marriage, sex, money and happiness. Not necessarily the meaning of life, but certainly something a lot closer than the numbers alone can provide.
P.S. I am a little grumpy about numbers today. I've been staring at spreadsheets all day and my eyes hurt.
Won't be that difficult to go from memory, actually, because I really had just one thought as I read the post: this kind of research positively screams for an ethnographic element. As Althouse says, the relationships between entities and emotions as enormous as the ones in question are so complicated that it's nearly impossible to untangle them. To state definitely that, say, happiness is positively correlated to marriage when money's in the picture, or that study participants with more money had more sex might be true, but it misses the point.
Add a layer here: a little explanation of why. Take these numbers and hypothesize something. Then test it by going deep with a smallish sample. Interview them, have them keep diaries, take photo essays. Get into that data. Then, these numbers will be really valuable.
Now, I'll admit that I haven't read the original paper that Lindgren takes to task. But I've read enough market research studies to know that a little extra depth goes a long way. That a miniscule collection of pictures really does do more than a thousand numbers could, even if those numbers are attached to attribute statements and even if they have the ultimate in statistical significance.
And I don't mean to discount the value of quantitative data. It's great. But when you're dealing with happiness, shouldn't you go a little deeper?
In my former job, I was a part of an ethnographic project of sorts that dealt with the concept of happiness. Throughout 2003, people living all over the world sent us photos with short explanations that defined "happiness" for them. During the year, we selected the best photos and analyzed them, to understand what happiness meant across and within cultures. At the end of the year, we looked at our previous analysis and the entire body of photos from a global point of view. The resulting analysis session was heavy, to say the least, but it was really interesting. And we did come up with some interesting conclusions (the final report is here).
I can only imagine what conclusions and great ideas (that might not be "conclusions" per se) could result from heavy analysis of Blanchower and Oswald's numbers PLUS a solid ethnography exploring the relationships between marriage, sex, money and happiness. Not necessarily the meaning of life, but certainly something a lot closer than the numbers alone can provide.
P.S. I am a little grumpy about numbers today. I've been staring at spreadsheets all day and my eyes hurt.
posted by Kit Pollard at
6:23 PM
0 Comments:
Post a Comment
<< Home