Ah, my theme. Dating. Yes, there has been a tiny bit of real life dating in the last month, but I couldn't write about it. Why? Because he reads the blog. In fact, I met him through the blog. (No, he is not my boyfriend. I would tell you about that. And there were only two dates.)
I realized that I cannot write about dates that occur as a result of this blog. Very annoying! And, no, I'm not going to tell you who he is, but he's a good guy and actually said, "I don't mind if you write about me, that would be great."
I said, "Really?"
"Sure, I could use you as a reference."
"A reference?" I thought about that for a second and said, "Oh, you think I would write something good."
I told my mother about this conversation and her reaction was, "You amaze me."
"I do?" I worried a little about what was coming next.
"You're just so…sharp. I don't know how you come up with those things."
Gee, Mom, I don't know either. I say stupid things all the time. If I'm lucky, people find them humorous.
She was also impressed that I've met people through the blog. That it has opened up my social world, not narrowed it. I know a lot of us have had this experience, and I've mentioned it before. It is one of the best things about blogging.
Here is another unexpected result of blogging: a whole bunch of us "dating bloggers" have been offered a review copy of "Unhooked Generation" a new book by Jillian Straus.
Truth be told, I was flattered to be asked--even though I wasn't the only one.
But when I got the request (or solicitation? what's the right word?), I had some pause. I've heard of the book—I believe it got a big write up in the WA Post (which I didn't read). I skimmed the book's website—and it confirmed my suspicions that it is based on a bunch of anecdotes. From these, overarching conclusions are drawn. It's the sloppiest kind of case study work imaginable. And I do a lot of lousy case studies.
After yesterday's post, you know something about my job. You may also recall that I have a Ph.D. in Sociology. My graduate program was methodologically intense. I know more about statistics than the rest of my division, and no less than anyone else in the entire office. I don't think of myself as an expert, but I know others do. Oh well.
But, expert or not, I cannot help but look askance at a book purporting to solve the problems of a generation when it's based on an unsystematic, purposive sample.
Before I accepted my copy of the book, I consulted my mother, "Should I tell them my bias?"
Mom said, "No. They are asking you to read it because of the blog, not because of your Ph.D."
"That's what I thought. I don't see any problem accepting a free copy of the book, though." I said.
"No, that's a non-issue. I've gotten review copies many times. That's standard practice."
"Right. The NY Times doesn't pay for books. I mean, I'm not the NY Times, but it's to their benefit for me to write about this book and I gain nothing. Except a free book."
"That's right. There's no reason not to do it. And you don't have to write about it." Mom said.
"But they might not want me to review it if they knew my background."
"Too bad." Mom said, "But is it on your site that you have a Ph.D.?"
"Well, it's not in my profile, but I have mentioned it. It's not a secret, but it might not be obvious."
"I still think it doesn't matter."
"I agree. I'll tell them they can send me a copy."
My negative bias towards this book is because of it's methodology—or lack thereof. The 100 people who are interviewed were not selected at random AND there is no theoretical underpinning for the conclusions. (There is nothing magic about a 100 person sample, either. It just sounds good.) Perhaps you can get away with this if you have a theory that is constructed based on experience, possibly therapeutic experience, and then you use case studies to emphasize your points. It is still questionable because you may intentionally or unintentionally exclude cases that do not support your theory.
Or you could use case studies to illustrate something, the way we do at work—we don't say "this is how the program works", rather, we use case studies as examples of how the program works in a particular places (though people often can't help but generalize from them). Sometimes we only include "good" examples—then we call the findings "best practices." It's lame, but at least it's honest.
In this case, the "findings" cannot be generalized, even to the target group (middle- and upper-middle class, heterosexual, (white?), people ages 25-39), with any confidence. I'm not saying the author is making any false claims about her sampling (she isn't) but she formulates some powerful conclusions based on her interviews. For example, "The cultural influences specific to this generation together have created obstacles in our search for true love." (p. 15). (Aside: what is that "together" doing in the middle of the sentence? I have no idea.)
She can't know about our obstacles to true love. You would have to define "true love" first, which is impossible. You would also have to isolate "cultural influences" from all other factors, such as economics, health, and population (demographic) trends.
But even if you could define those things, probability theory still says she can't make that statement based on this sample. Because the sample is not random, you have no way to predict how likely it is that your hypothesis (if you had one, which she doesn't) is supported for real or you just got lucky and found the few cases that confirm it. Every survey has a margin of error. That is the power of probability sampling—it lets you put boundaries on error. It gives you a sense of how believable your findings are.
In this case, the margin of error around the author's claims is unknown and unknowable. If you are not going to draw a probability sample, but plan to do ethnographic work, in-depth interviewing or case studies, it makes sense to start out with some ideas and hypotheses so we, as readers, understand where you're coming from. We need to know the author's bias in order to evaluate her conclusions.
Still, the book could be useful. Anything that makes you take on a different perspective can be useful. It just can't be the answer to our generation's woes. Later age at first marriage and a higher rate of "never married" people could be due to a whole host of things that have nothing to do with our emotional state. Here are just a few alternative explanations (pitched at the population level, not the individual level) for the patterns the author observes:
- Almost all women participate in the labor force now, reducing the economic reasons for marriage.
- Everyone stays in school longer, delaying time to first marriage until after joining the labor force.
- Sex outside of marriage is more acceptable so there is less social pressure to marry.
- Having children outside of marriage is more socially acceptable and decreases the urgency for marriage.
- There may be as many long-term "unions" as in the past, but they may not be marriages. Fewer people getting married just means there are fewer marriages, not necessarily fewer long-term relationships. (No one can say if long term unions are the same as marriages because we don't collect much information about them.)
Did I mention that I studied demography in grad school?
After I read the rest of the book, I can let you what I think in even more detail. Interested?
I'm pretty sure this is not what the PR firm had in mind.
Grateful for: opportunities.
Drop me a line.