1. Tonight (11/19) I’ll be speaking at Yale on the subject of “How To Ruin A Perfectly Good Randomized Controlled Trial” and other aspects of interpreting scientific evidence. Probably. I’m currently stuck in New Jersey due to airplane mishaps, but 90% confidence I’m going to be able to make it on time. Event is at LC101, Linsly-Chittenden Hall, 63 High St, New Haven, at 7 PM. You’re invited.
2. Monday (11/23) I’ll be speaking at Harvard at Sever Hall 203 at 5 PM on the same subject. You’re invited to this one too.
3. Sunday (11/22) we’ll be having a Boston meetup (thanks to the Boston rationalist and EA communities for setting this up) at MIT, room 5-134 at 6 PM. There is a Facebook event page here. You are invited to all of these things but you are extra definitely invited to this one. Every time I try to have a meetup I specify that everyone is invited even if you are only a lurker and even if you don’t understand everything you see on SSC and even if you are not very interesting and so on and so forth, and every time people still say “I wanted to go but I didn’t because I’m afraid I wouldn’t have been welcome there because I’m not enough like the typical SSC reader,” and so now I have to repeat in bold that you are definitely welcome.
I won’t be around much for the next week or two while I do this stuff and see my family for Thanksgiving, so apologies for not responding to emails/comments, etc.
I seem to recall you had family in Utah. Do you think you’ll ever host a meetup in SLC?
Is there an SSC meetup calendar (maybe a fan made one) I can subscribe to? If not, could someone make one? (Again not necessarily Scott)
I don’t usually read posts as they come out and I miss out on timely announcements like I did here.
I read this blog routinely but do not read LW. Will I be able to follow a conversation at this meetup?
dammit you got me excited with ‘new england’. thought it said a new meetup in england.
a new england
I’m seriously thinking about coming to the meetup on Sunday. I’ve been thinking about finding the rationalist people in Boston for a while now, and this is a perfect occasion.
Anyone know where in MIT this room is? How long y’all expect this to go for? Should I bring anything in the way of food or drink?
You can use the MIT map here to find the location. It’s in building 5, floor 1, room 134. If you walk down Mass ave there’ll be a big important entrance. That goes into lobby 7 (in building 7) so just enter there and take a right.
An even better link.
Probably not the place for this, but speaking of misusing science – this could well be called “When columnists who had a few classes in Gender Studies get their hands on popular science reporting”.
This makes me want to cry, and I’m not a STEM type at all – on top of everything else, my English language sensibilities are offended that she misspelled “wintry”*.
Well now, aren’t you all ashamed of yourselves, you bad, wicked scientists? All the sexism, homophobia and transphobia in human history is all your fault! Naughty, naughty science!
But honestly, if we’re all born with nice unisex brain blank slates, why the hell are we arguing over foetal testosterone exposure and all the rest of it?
I think there may be the kernel of some actual science in amongst all the grooviness, but I’m too depressed by this to try digging it out.
Naturally I came across this on Tumblr where it’s been lauded as “science proves we’re all trans!” or something.
*When speaking or writing of weather, it is “spring-like” or “vernal”; “summery”; “autumnal”; “wintry”.
It’s pretty terrible, but I’m more struck at how unusual it is; for all the complaints about how scientists are terrified about political correctness, the usual pattern seems to be that studies which find differences between the sexes that can be linked to the traditional stereotypes get lots of mostly favorable media attention, and studies which fail to find differences or find unexpected differences are almost completely ignored (except apparently by non-mainstream outlets like the one you link to).
My own impression is that this is a fairly recent trend. But that could be a product sampling bias.
I was attached to an expeditionary squadron back when the women in combat controversy was in full swing (circa 2004) and at the time it seemed like any attempt to argue that there was sort of fundamental difference between the sexes was an invitation for social and professional castigation.
the usual pattern seems to be that studies which find differences between the sexes that can be linked to the traditional stereotypes get lots of mostly favorable media attention
I do think it’s swinging away from that; whatever the reason, the blank slate/we can be programmed to be anything has always seemed to be popular, even if in the background. Probably because of precisely that: traditional stereotypes. So when it was beneficial to say women were more nurturing and compassionate and peaceful and consensus-seeking therefore women should be in charge because then there would be no more wars or inequality, studies showing larger brain areas associated with such were all the rage (like I said, on the level of phrenology).
Now it’s “there are no differences so women are the complete equals of men in every way and you are terrible for saying women can’t be Strong Independent Females”.
It was the mixture of bad science reporting (those hippocampus measurements – absolute or relative?) and Mandatory PC Studies For Everyone that got me.
There’s a certain black humour in the attack on science, but I can’t say I’m surprised. They don’t want facts, they want fodder for their argument of the moment. So if it suits The Cause (whatever the cause of the day may be) to say “Science says men and women have different brains”, that is what is used to back up certain demands. And if it suits to say “Men and women have identical brains, gender is only imaginary” then science – if it knows what is good for it – had better get with the programme.
I don’t know what particular axe this person has to grind, but it will be instructive to see them getting into a fight with trans activists about unisex brains, and see them tripping over themselves trying to be more correct than thou on what are “the civil rights movement(s) of our times” while simultaneously not admitting they were wrong about anything.
So men and women are born with the same brain, gender is a social construct, but homosexuals and transexuals are Born This Way™, it’s in their genes!
What? Logic you say? Don’t you know that logic is a tool of the Patriarchy? /s
So, I’ll probably leave this hear and then not continue the discussion, since we’re incredibly off topic for this post and probably running afoul of the general “race&gender” thread rules, but most people I know who are actually LGBT really, really dislike the “Born This Way” thing, or at least view it as a pedagogical technique that massively simplifies things to help sway moderates who are trying to understand what it’s like to be gay, trans, etc.
Huh, if science ruined everything, then I guess gender roles didn’t exist before the 17th century. Learn something new every day.
This is my very first comment on your blog and it’s to say I’m going to try to make it to the meetup on Sunday. Happy?
Yes.
Well, don’t be that happy because it turns out I won’t be there. 🙁
I promise the worry that I would be unwelcome had nothing to do with this, though! It was just a number of minor barriers and opportunity costs that combined were enough to make it not a good idea to come anymore. Thank you for the invitation!
That MIT room seats 40 people. Isn’t that a little small?
They’ve now added a second room, 5-234.
Wtf? Those are on different floors. How’s that going to work? Who’s going to go to the room that Scott is not in?
Well, I’m mildly intimidated but I think I might actually make it to the Harvard talk and the meetup. I actually moved to the area a couple weeks ago and I am exactly the kind of lurker you are specifically inviting, and I appreciate the bold notice. This, incidentally, also marks my first comment, though I’ve been reading for…I think a year and a half, I’m not sure. So…I look forward to this. Also seconding the question about how late the MIT meetup will go for planning purposes.
How late is the MIT meetup likely to go? Would it be worth showing up an hour and a half late?
Is this presentation based on textbook stuff? Or do you have some experience in designing clinical trials?
The presentation is on some of the same things I talk about regarding analyzing research on this blog and comes from book stuff and my own experiences trying to figure things out. I’ve designed a couple of research projects, but no really big ones of the sort that I talked about tonight.
Since random people I know on Tumblr are being bothered about this, let me give an example. The first thing I talked about in my presentation today was a pharmaceutical trial which “proved” Luvox had fewer side effects than Prozac. The trial came under scrutiny after it was found that they used an unusually low dose of Luvox to minimize side effects. This is a common issue in pharmaceutical trials, most people don’t know to look for it, but I’ve read enough pharmaceutical trials and commentary on pharmaceutical trials by now that I know about this and several other similar tricks.
I think this is useful information, not especially controversial, and not the sort of thing that requires me to be the Director Of Running Randomized Controlled Trials somewhere in order to talk about.
Do they use one trial to show Luvox is effective and a different trial to show it has fewer side effects?
Yes.
Jesus, that’s frickin’ fraudulent. The study on clinical effectiveness needs to be the same one that the side effects info is drawn from.
Not sure if this is relevant or not, but with all the complaining about how the FDA is unreasonably over-cautious and stringent in its testing requirements, the other day I saw an article hauling them over the coals for accepting the manufacturer’s data and assuming that if some antidepressant (I can’t remember which) worked at one dose, it would work the same at a higher dose.
Now it’s been revoked, and the article was very indignant that the American public had been used as guinea pigs for five years (or however long the higher dose was manufactured and prescribed).
So really it’s damned if they do insist on testing every last combination, and damned if they don’t.
I don’t know exactly what “accepting the manufacturer’s data” means in this context, but if the manufacturer is fudging the data even a little, the FDA can easily be both over-strict and over-trusting. All it needs to do is demand an excessively expensive study sequence and then not look too carefully into the data it gets back from it.
The incentives check out, since the more expensive the study is, the higher the sunk costs are. If you sink a billion dollars into human trials of a drug, it’d better perform to FDA specs or you’ll shortly be out of a job. And you didn’t think the FDA was going to repro every billion-dollar study themselves, did you?
The article (and again, this is second-hand science reporting) seems to say that the FDA accepted the data that at dose 1, the drug was effective, and then didn’t require further testing that the drug had the same effects at dose 2 which was higher.
I am broadly sympathetic to the FDA, because they are being blamed for insufficient caution here yet the same writers would be just as happy to write a critical article about the FDA holding up a drug with unnecessary tests and that the lower dosage test results should be perfectly acceptable. Look at the way a media opinion campaign was mobilised for Addyi – so I think the outraged critical tone is a little harsh.
The FDA really are between a rock and a hard place – hold up a drug for testing and you get the drug companies feeding press releases to sympathetic media columnists about how this means suffering and delay in a cure or treatment; take the manufacturer’s data on the face of it and let the drug be marketed, then get castigated by the same media for using the public as guinea pigs.
Speaking of Addyi – I am pleased to say that its launch has not resulted in the torrent of sales to desperate women seeking horniness that they anticipated.
I take they used the normal amount of Prozac?
I don’t know – turning up late to a presentation on “How to Muck Things Up” sounds perfectly in the spirit of the thing 🙂
Your SSC meetup is in exactly the same small classroom, on the opposite end of campus from CS, where I got assigned to teach my graduate class this semester! (Is there some scheduling algorithm that hashes “Scott A” into “5-134”?)
Which is to say: I look forward to coming on Sunday, and I will not get lost on the way.
I dearly hope Scott Adams won’t be coming as well. Legend has it that when all three Scott As meet in one spot the world will be destroyed.
I’m pretty confident that I wrote a first year law school exam solely about your dad.
Can I get some context on this?
Presumably Dror Bar-Natan.
That’s a mathematician, why would he have anything to do with a law school exam?
Being said dad, I can guess the context. But I’m truly interested in the content! Do share.
Ok, well now I HAVE to attend this. BRB, shifting Sunday schedule.
Well, nope. I am stuck in Chicago, in a huge snowstorm, for at least another day. I will miss the meetup. I am frustrated and miserable about it.
…can we have an SSC meetup in Chicago, then?
Is there a risk of events reaching capacity? (Is there a max capacity?)
I’m nowhere near New England but I really want to hear those lectures. Will recordings be posted anywhere?
Probably not, but I’ll probably post the written version online. It won’t be anything you guys haven’t heard before.
Can you post lecture notes after the talk? As a longtime SSC reader I’m very interested
Curious if you got invited to these events (well the first two) due to professional contacts or SSC/LW-ish internet contacts?
Also interested in this.
Heads up that the MIT meetup is right at the end of Splash (giant educational program/nerd-con for high school students). Campus in general will be sort of swarming with them, and Lobby 7 in particular will probably be packed, because everyone will be saying goodbye and waiting for their parents to pick them up. I recommend trying to go in a side door to avoid this. There’s also some chance a Splash volunteer will try to clean up that room while you’re in it, in which case you have the choice of letting them do it and politely telling them to get lost.
http://www.learningu.org/becoming-a-chapter/create-a-new-splash
If you like the idea of high school students being able to come to your university for a weekend and take classes on basically anything taught by undergrads or grad students, you should get in touch
See you at Yale!
dammit, the one week I’m in Mexico.
I am coming out to Boston on Monday evening, what time is your talk?
FYI, the name of the Harvard building is pronounced like the Hall of Fame pitcher, not what you do to a finger.
/ˈsoʊfæks/?
/ˈsiːvər/
*irrelevant nitpick*
/ˈsiːvəɹ/
(I’m really sorry, I couldn’t help myself)
I can’t attend any of these things, but will the lectures be recorded somewhere ?
Seconding request for recording.
What time is the Harvard talk?
Any chance that “How To Ruin A Perfectly Good Randomized Controlled Trial” will be posted after the event? Geographical distance prevents me from attending.
Seconding the request to post your talk! I don’t have the excuse of geographical distance, but I do have the excuse of ludicrous amounts of homework. Would you be okay with an audience member or two filming it? Would any potential audience members be willing to film it for the rest of us?
Thirding this request! Any chance of a YouTube post?
#2 — time?
5PM. Added.
I’ll be at the MIT event and I might be able to make it to the presentation on Monday.
Is this talk based on a paper you wrote? It sounds interesting.
Let me know if you are in the Baltimore/DC area and want to chat.