Re: Extra Bad

1

The wording on that kind of icks me out. I know they're talking about parental consent for abortion, but the words "parental involvement...sexual activity teenaged girls" just sound kinda creepy together.

horizontal rule
2

I'm immediately wary that it's by Klick. Couldn't remember where I'd seen that name before, but a little Googling brought my memory back. He's written for AEI.

This, however, is some funny shit.

horizontal rule
3

Harford is fudging the difference between incentives and information. In his introduction he asks "What sort of incentives have changed?" that might lead to substituting oral for penetrative sex, but his answer has to do with students being bombarded with information about the risks of penetrative sex.

The negative outcomes of penetrative sex wouldn't seem to have changed for about 15-20 years (whenever AIDS was at its peak); what's changed is people's awareness of them. Which would have to do with public advertising campaigns, probably prevailing mores, etc. All of which is inimical to the rational horny teenager hypothesis.

So, until I'm convinced otherwise, I'm going to figure this as economists overreaching. Maybe dsquared can weigh in.

horizontal rule
4

I should probably IMO the first paragraph or two a bit more, since I don't know for sure that economists don't fold information into what they call 'incentives'.

horizontal rule
5

Assuming choices to engage in risky sex are made rationally

Hmmm...

horizontal rule
6

5: "Rationally" here is used in the technical economic sense of changing behaviour (at the margins) in response to incentives.

horizontal rule
7

I know that. I'm still skeptical, though.

horizontal rule
8

From the abstract:

"While we find a relatively small reduction in rates for black girls, it is not statistically significant. We speculate that the racial heterogeneity has to do with differences in family structure across races.

Heh. Cause black girls ain't got no daddies.

horizontal rule
9

Yeah, the abstract does seem to be studded with right-wing code words.

horizontal rule
10

Oh, and 4: He's certainly playing fast and loose with the difference between information and incentives. What matters is that the incentives have changed because of the content of that information - oral sex is safer because of x, y, and z.

At an econ 101 level, it's assumed that everybody has all the relevant information. Clearly that's not really true, but assuming otherwise makes the math hard.

This is far enough from the stuff I'm into that I can't really comment on the methodology, but nothing screams out at me.

horizontal rule
11

Going for some sort of records for consecutive comments, I'm skeptical that the assumption of rationality is less valid for teens than adults. It's just that their utility functions are so different - dealing with parents, other authority figures, hormones, and such.

Then again, the assumption of rational behaviour among adults is pretty dubious.

horizontal rule
12

Then again, the assumption of rational behaviour among adults is pretty dubious.

This is what I'm skeptical about, it's just that the article is about teens.

horizontal rule
13

One of the flaws in this is that it assumes that the social acceptability of oral sex has remained constant. For teenagers, there isn't just the "cost" of sexual activity in terms of pregnancy/disease/having to tell your parents, there's also a reputational cost. I think I was in the last wave of high school students for whom sex was considered more socially acceptable than oral sex. If you had sex with your boyfriend, that was OK. Oral sex? You were labeled a total slut. And that was with your boyfriend. We aren't even talking about the casual oral sex that people claim is going on now.

horizontal rule
14

Becks's public shaming: revealed!

Otherwise, good point.

horizontal rule
15

I think I was in the last wave of high school students for whom sex was considered more socially acceptable than oral sex.

I doubt this. My memory of high school attitudes is that oral sex outside of a relationship was definitely seen as "slutty" in a way that sex in a similar context wasn't. Within a relationship I guess both were considered acceptable (which supports your larger point), but I don't remember much talk about it. All this only applies to fellatio, of course.

horizontal rule
16

12: As useful as it is, the strong version of rationality (people as utility-maximizing automatons) clearly doesn't really describe our psychology. However, a weaker version is just "peopel respond to incentives", which generally works.

horizontal rule
17

It's stuff like that that made me realize I could never be an economist.

horizontal rule
18

That is, I'm not opposed to economics, I just can't for the life of me make myself think the way economists do.

horizontal rule
19

This strikes me as a terribly unlikely result -- I just can't imagine that all that large a percentage of girls know what the parental notification laws in their states are. Thirty percent of adults don't know what year 9/11 happened. For the effect to be significant through the background level of ignorance, it would have to be improbably huge.

Obviously, I could be wrong, but this sets off giant blinking "does not make sense" lights in my head.

horizontal rule
20

18: yeah, it's pretty pathological. Even worse, it seems like the good ones think that way in their personal lives.

One credit - they aren't as nuts as physicists. Man, are those guys weird.

horizontal rule
21

Or epistemologists. What's up with them?

[suspicious glance in the direction of Lubbock]

horizontal rule
22

19 - that's an excellent point.

horizontal rule
23

I always read "epistemologists" as "episiotomists", which is why I stay as far away from Weiner as I can.

horizontal rule
24

That joke is even better when you say it out loud.

horizontal rule
25

If you had sex with your boyfriend, that was OK. Oral sex? You were labeled a total slut.

God, this stuff is all so complicated. As a horny teenager - even a nominally conservative horny teenager - I was too concerned with various schemes to procure sex myself to quite figure out what was and wasn't supposed to be "slutty." I could never understand the desire to stigmatize sex because that would clearly only make my goal that much harder to accomplish. So for me sex just went from being something gross and icky that gave you cooties to something totally hot that everybody needed to have all the time.

horizontal rule
26

20:

What's the problem with physicists, huh?

horizontal rule
27

I always read "epistemologists" as "episiotomists", which is why I stay as far away from Weiner as I can.

Cue Henley.

horizontal rule
28

Wow, 23 really is better out loud.

horizontal rule
29

which is why I stay as far away from weiner as I can.

Sure it isn't the notification laws?

horizontal rule
30

which is why I stay as far away from Weiner as I can.

Sure it isn't the notification laws?

Also, you're a lesbian.

horizontal rule
31

Whole lotta emendation goin' on.

horizontal rule
32

So, to recap, in the short space of 30 comments, I have been revealed to be both a slut and a lesbian. A good night's work, I'd say.

horizontal rule
33

OK, I skimmed the paper. Something is setting off alarm bells. I can't put my finger on it, but they seem to be routinely conflating "teenagers" and "teenage women," and "risky sex" with "risk of pregnancy."

Specifically, the study says: "we use the incidence of gonorrhea among women under the age of 20 as our proxy for risky sex." If they only looked at disease rates for women, how does this tell us anything real about "risky teenage behavior"? Doesn't it, at best, tell us about teenage women?

But wait -- turns out they did look at male rates. (Only as "robustness check" for the quality of their other data.) What did they find? There are no statistically significant effects for white or Hispanic male teenagers (pp. 19-20 of the pdf).

So I dunno. The abstract does say "teenage girls," but it seems sneaky. Are they really studying how notification laws affect teenage sexual behavior, or are they checking to see whether female teens are being "scared straight"? (The study notes that male teens can just decide to keep having risky sex with older partners).

Another alarm bell: Reference to HPV as "the one STD for which condoms do not appear to be an effective protectant." Less than 100% effective, sure, but pretty good, according to this report.

I'm not impressed.

horizontal rule
34

the methodology appears to be superficially rather similar to the methodology used to try to explain variations in crime rates by the passage of shall-issue concealed weapons laws. That's not to say it's hackwork, but my personal opinion was always that even if Lott hadn't been a hack, he was still engaged in something profoundly not worth doing.

AFAICS, it is one of those horrifically disheartening papers that Stephen Levitt co-writes so many of, where Jim-Bob goes "hey, I found a dataset", Johnny-Boy goes "I just bought a copy of STATA" then they look at each other and say "LET'S DO THE SHOW RIGHT HERE!!!".

I mean, fuck, it doesn't make a glaring statistical howler in the first three pages, but there is no way that something like this is ever going to give an answer to the question asked; the statistical tools are massively underpowered compared to the problem. The real danger here is that someone will mistake this for social research - beyond that, it's just a bit of a waste of money and denial of the human soul.

horizontal rule