The basic argument isn’t that complicated though: A) there are women who attribute their lack of involvement in the STEM fields to extremely bad experiences at an early age and B) there’s an obvious way this would be causally related.
There are several problems with that theory.
1) A lot of people who deice not to go into STEM had bad experiences. (In fact bad experience may very well mean wasn’t good at it).
2) The kind of things they wind up pointing to as “sexual harassment”, e.g., wearing a bad 50′s sci-fi shirt with ‘ray-gun-babes’ or happening to overhear a not-quite g-rated conversation between two men, don’t seem like the kind of things people should be too bothered about.
3) Women have less variance on IQ scores then men and thus we would expect fewer of them to show up in at high levels in IQ-intensive fields.
(Feminists dispute the last point, but they’re arguments tend to boil down to “you’re sexist for even suggesting this”).
As for 2, sure there’s a range of behaviors and it is worth discussing that which ones do or don’t matter. At the same time, the mild behavior is the one set of behavior that we actually do have studies showing it has an impact. In particular, women who have been stared at by men then perform more poorly on math tests(PDF).
3) Women have less variance on IQ scores then men and thus we would expect fewer of them to show up in at high levels in IQ-intensive fields.
Yes, up to a point. No one here is asserting that this is the only cause or the primary cause of differences in gender ratious. That’s not the same thing as asserting that it isn’t a cause. And IQ variance is clearly insufficient: different disciplines requiring similar needs have radically different gender ratios. And there’s real evidence that in at least some cases, cultural issues are having much more of an impact than IQ- look at how the percentage of women in IT and computer related fields was steadily going up and then started dropping when personal computers appeared. See discussion here.
(Feminists dispute the last point, but they’re arguments tend to boil down to “you’re sexist for even suggesting this”).
This is now the second time you’ve made a comment like this- bringing up an argument that hasn’t been made so you can knock it down. That might be rhetorically fun but it isn’t helpful. Bad arguments are made for pretty much any position possible. The fact that such arguments are being made somewhere isn’t relevant for fairly obvious reasons.
In particular, women who have been stared at by men then perform more poorly on math tests(PDF).
That’s a paywall, so I assume you have not read it. Here’s a jailbroken copy: “When What You See Is What You Get: The Consequences of the Objectifying Gaze for Women and Men
”, Gervais et al 2011 (Libgen; PDF.yt; Dropbox).
This paper inherits the usual defects of the ‘stereotype threat’ literature. It takes place in no-stakes situations, while stereotype threats have failed to generalize to any situations that actually matter, and blinding is questionable (they bring the subjects in, then “They also learned that they may be asked to report their feelings about themselves and others and to complete word problems”, and do math problems? Gee, I’m sure none of these undergrads recruited from psychology classes figured out what the real experiment was!) The results are also a little bizarre on their face: ”...the objectifying gaze also increased women’s, but not men’s, motivation to engage in subsequent interactions with their partner...the objectifying gaze did not influence body surveillance, body shame, or body dissatisfaction for women or men”. Huh?
And IQ variance is clearly insufficient: different disciplines requiring similar needs have radically different gender ratios.
That does not follow. If different disciplines have non-identical needs, then depending on the exact differences in distribution shape, the correlation between IQ, and the cutoff for success (see for example the table of r vs cutoff in “What does it mean to have a low R-squared ? A warning about misleading interpretation”) - not to mention the other variables which also vary between gender (Conscientiousness; degree of winner-take-all dynamics; expected work hours) - may well be sufficient to explain it. You’ll need to do more work than that.
And there’s real evidence that in at least some cases, cultural issues are having much more of an impact than IQ- look at how the percentage of women in IT and computer related fields was steadily going up and then started dropping when personal computers appeared. See discussion here.
This paper inherits the usual defects of the ‘stereotype threat’ literature. It takes place in no-stakes situations
Sure. It is extremely difficult to test these situations in high-stakes situations for obvious reasons.
Gee, I’m sure none of these undergrads recruited from psychology classes figured out what the real experiment was!
This is an intrinsic problem in almost all psychology studies. Is there anything specific here that’s worse than in other cases?
The results are also a little bizarre on their face: ”...the objectifying gaze also increased women’s, but not men’s, motivation to engage in subsequent interactions with their partner...the objectifying gaze did not influence body surveillance, body shame, or body dissatisfaction for women or men”. Huh?
I don’t see what your point is. What do you find is bizaare about this and how do you think that undermines the study?
And finally, this is social psychology.
That’s a reason to be skeptical of the results, not a reason to a priori throw them out.
That does not follow. If different disciplines have non-identical needs, then depending on the exact differences in distribution shape, the correlation between IQ, and the cutoff for success (see for example the table of r vs cutoff in “What does it mean to have a low R-squared ? A warning about misleading interpretation”) - not to mention the other variables which also vary between gender (Conscientiousness; degree of winner-take-all dynamics; expected work hours) - may well be sufficient to explain it. You’ll need to do more work than that.
You are correct. The word clearly is doing too much work in my comment. At minimum though, the fact that other similar disciplines don’t have that situation even though they historically did is evidence that IQ variance is not all that is going on here. And that’s especially the case when many of those disciplines are ones like medicine that have taken many active steps to try to encourage women to be interested in them.
And there’s real evidence that in at least some cases, cultural issues are having much more of an impact than IQ- look at how the percentage of women in IT and computer related fields was steadily going up and then started dropping when personal computers appeared. See discussion here.
See discussion here.
Now seen. Having read that discussion, I agree with Kaj there. Do you have any additional point beyond which you said to Kaj there?
It is extremely difficult to test these situations in high-stakes situations for obvious reasons.
It is not so difficult as all that: high-stakes tests are conducted all the time and gender is routinely recorded. I refer you to the WP article for how stereotype threat evaporates the moment it would ever matter.
This is an intrinsic problem in almost all psychology studies. Is there anything specific here that’s worse than in other cases?
It is not that bad a problem in most studies, and stereotype threat studies are particularly bad.
What do you find is bizaare about this and how do you think that undermines the study?
Their results make no sense in almost any causal model of how stereotype threat would work. What sort of stereotype threat has no effect on attitudes and body images and increases interest in co-workers, and how would you expect this to support the argument you made with regard to co-workers in the real world?
That’s a reason to be skeptical of the results, not a reason to a priori throw them out.
Indeed. So why did you cling to a weak reed?
Do you have any additional point beyond which you said to Kaj there?
No. I stand by the sum of my comments: that it is blatant post hoc rationalizations which contradict any theory a feminist would have made before seeing the actual data, which clearly supports an economic rather than pure bias account, and makes false claims about new CS students as well.
It is not so difficult as all that: high-stakes tests are conducted all the time and gender is routinely recorded. I refer you to the WP article for how stereotype threat evaporates the moment it would ever matter.
Which studies there are you referring to as being relevant? Note by the way that the study in question isn’t quite the same as stereotype threat in the classical sense anyways.
What do you find is bizaare about this and how do you think that undermines the study?
Their results make no sense in almost any causal model of how stereotype threat would work. What sort of stereotype threat has no effect on attitudes and body images and increases interest in co-workers, and how would you expect this to support the argument you made with regard to co-workers in the real world?
I’m not completely sure what model would actually do this but it could be something that causes them to think of themselves less as people doing math and more as people who are socially or sexually interested in others. But the fact that it didn’t have an impact on body image is strange, and needs further investigation. In the short version though, that should suggest that this study actually is more reliable: one of the most common criticism of psychology as a discipline is that the studies have way too high a confirmation of hypotheses rate. That’s been discussed on Less Wrong before. In this case, the fact that part of the study went against the intuitve hypothesis and went against what the authors explicitly hypothesized is a reason to pay more attention to it.
That’s a reason to be skeptical of the results, not a reason to a priori throw them out.
Indeed. So why did you cling to a weak reed?
Because this is one of the very few studies that have looked at how sexualization impacts performance. There are a lot of stereotype threat studies (as you noted) but they don’t generally look at this. I’d be happy to rely on something else or change my opinion here if there were that many more studies.
Do you have any additional point beyond which you said to Kaj there?
No. I stand by the sum of my comments: that it is blatant post hoc rationalizations which contradict any theory a feminist would have made before seeing the actual data, which clearly supports an economic rather than pure bias account, and makes false claims about new CS students as well.
So in your view, what precisely is the reason for the fact that the percentage of female CS students was consistently rising and then took a sharp drop-off? Also, why do you think Kaj disagreed with your position?
Which studies there are you referring to as being relevant?
...So in your view, what precisely is the reason for the fact that the percentage of female CS students was consistently rising and then took a sharp drop-off?
If you aren’t going to read the links I provided*, I’m not going to bother continuing. Both of those questions were answered.
* please note I have already gone above and beyond in not just reading your source material while you have not, but jailbreaking & critiquing that study, and also excerpting & linking contrary opinions & surveys
If you aren’t going to read the links I provided*, I’m not going to bother continuing. Both of those questions were answered.
I read the conversation with Kaj and I read the links thank you very much. In that conversation you brought up a variety of different issues, focusing on the “practicality” issue but you give multiple different versions of that claim and I’m not completely sure what your primary hypothesis is. The primary claim there seems to be that the ups and downs on the graph mirror ups and downs in the market, but the primary link justifying that claim is this one you gave which doesn’t make any claim other than the simple claim that the graphs match without even showing that they do. The only bit there is there that is genuinely interesting evidence is the survey showing that women pay more attention to job prospects when considering fields which is not at all sufficient to explain the size of the drop there, nor the fact that law didn’t show a similar drop in the last few years when there’s been a glut of lawyers.
please note I have already gone above and beyond in not just reading your source material while you have not
I don’t know where you are getting the second part of that claim from. But it is true I didn’t read every single link in the Kaj conversation, and I’m not sure why you think reading a single study is on the same scale as reading an additional long conversation and every single link there. So if you want to point to which of those links matter there, I’d be happy to look at them.
Bad arguments are made for pretty much any position possible. The fact that such arguments are being made somewhere isn’t relevant for fairly obvious reasons.
On the other hand, the fact that such arguments are used to intimidate anyone who dares question a certain position is relevant (possibly successfully remember what happened to Summers). In particular it affects what arguments we expect to have been exposed to.
Furthermore in Lewin’s case we have no idea what he actually did, thus the only evidence we have is that a committee at MIT decided what he did was bad. Thus to evaluation how much we should trust their conclusion it is necessary to look at the typical level of argument.
On the other hand, the fact that such arguments are used to intimidate anyone who dares question a certain position is relevant (possibly successfully remember what happened to Summers). In particular it affects what arguments we expect to have been exposed to.
It isn’t at all relevant. To use a different example (coming from the other side of the poltiical spectrum)- one argument made against releasing the recent torture report was that anyone wanting it released was “anti-American” which is essentially the same sort of thing. The presence of such arguments is in no way relevant to any actual attempt to have a discussion about whether releasing the report was the right thing. No matter what position you discuss someone will be using bad arguments to intimidate people into silence. Rise above it.
Furthermore in Lewin’s case we have no idea what he actually did, thus the only evidence we have is that a committee at MIT decided what he did was bad. Thus to evaluation how much we should trust their conclusion it is necessary to look at the typical level of argument.
The typical level of argument isn’t that when it comes to sexual harassment though. The typical level is a massive mix with some universities overreacting, and other’s underreacting. For every example of a university overreacting there’s an example of it underreacting. For example here.
But this also isn’t relevant for another reason: this entire subthread isnt even discussing the specifics of the Lewin case but a more general question of whehether such issues matter and are worth discussing. It is a red herring to go back to the original situation. But if you really do care about that situation, it might be worth looking at what Scott Aaronson has said on it, I’m curious if this adjusts your estimate at all that this is a minor situation being overblown?
The typical level of argument isn’t that when it comes to sexual harassment though.
I’ve been somewhat following the situation, and yes it is. The fact that you would claim otherwise cause me to update away from trusting other claims or judgements you make on the subject.
But if you really do care about that situation, it might be worth looking at what Scott Aaronson has said on it, I’m curious if this adjusts your estimate at all that this is a minor situation being overblown?
I didn’t see anything in the article that would adjust my estimate. The only thing there is that some who know told Aaronson that “this isn’t a borderline case”, given the kinds of things feminists consider “not borderline cases” these days that isn’t strong evidence.
The typical level of argument isn’t that when it comes to sexual harassment though.
I’ve been somewhat following the situation, and yes it is. The fact that you would claim otherwise cause me to update away from trusting other claims or judgements you make on the subject.
I’m not sure a polite response to that, so let me just ask, given that I just pointed to an example that went in the other direction, maybe it is worth considering, just maybe, possibly, that you are vulnerable here to a combination of confirmation bias and what media sources you are using? Let’s as a start focus on a simple example: were you aware of the example I linked to above before I linked to it?
I didn’t see anything in the article that would adjust my estimate. The only thing there is that some who know told Aaronson that “this isn’t a borderline case”, given the kinds of things feminists consider “not borderline cases” these days that isn’t strong evidence.
At this point, I think we may be having problems with radically different priors, part of which is that I give Aaronson enough credit that I don’t think he’s going to go the most radical end of the women’s studies department and ask for their analysis to get some idea of what happened.
At this point, I think we may be having problems with radically different priors
Aaronson’s post also states that the incidents occurred online, and for that matter on the MITx platform, which caters to MOOC users, not actual MIT students. Given these factors, I just can’t see how MIT’s Damnatio memoriae towards Walter Lewin could be anything but an outrageous overreaction.
I’m not sure at all the relevance of your comment in the context of what we are quoting. The fact that this was not the regular MIT students has been known for a while. I’m also not sure what that has to do with my comment, since everyone here is in agreement that the removal of the videos was an overreaction. (However, I’m not at all sure how the fact that it was with MOOC users rather than regular students makes any difference whatsoever unless you are talking about a very marginal difference in legal liability.) What is the connection between your comment and the part of my comment that you are quoting?
At this point, I think we may be having problems with radically different priors, part of which is that I give Aaronson enough credit that I don’t think he’s going to go the most radical end of the women’s studies department and ask for their analysis to get some idea of what happened.
He can only get information from the people who handled the case, who are likely to be SJW-types.
He can only get information from the people who handled the case, who are likely to be SJW-types.
These issues are handled in general by university committees. Does your lack of knowledge on this fact cause you to update at all about how good your judgment is for such issues?
Also it is worth noting that “SJW-types” in most contexts is a group which is by and large restricted to certain parts of the internet or some parts of certaind departments on campuses.
There are several problems with that theory.
1) A lot of people who deice not to go into STEM had bad experiences. (In fact bad experience may very well mean wasn’t good at it).
2) The kind of things they wind up pointing to as “sexual harassment”, e.g., wearing a bad 50′s sci-fi shirt with ‘ray-gun-babes’ or happening to overhear a not-quite g-rated conversation between two men, don’t seem like the kind of things people should be too bothered about.
3) Women have less variance on IQ scores then men and thus we would expect fewer of them to show up in at high levels in IQ-intensive fields.
(Feminists dispute the last point, but they’re arguments tend to boil down to “you’re sexist for even suggesting this”).
I’m not sure what you mean by 1. Can you clarify?
As for 2, sure there’s a range of behaviors and it is worth discussing that which ones do or don’t matter. At the same time, the mild behavior is the one set of behavior that we actually do have studies showing it has an impact. In particular, women who have been stared at by men then perform more poorly on math tests(PDF).
Yes, up to a point. No one here is asserting that this is the only cause or the primary cause of differences in gender ratious. That’s not the same thing as asserting that it isn’t a cause. And IQ variance is clearly insufficient: different disciplines requiring similar needs have radically different gender ratios. And there’s real evidence that in at least some cases, cultural issues are having much more of an impact than IQ- look at how the percentage of women in IT and computer related fields was steadily going up and then started dropping when personal computers appeared. See discussion here.
This is now the second time you’ve made a comment like this- bringing up an argument that hasn’t been made so you can knock it down. That might be rhetorically fun but it isn’t helpful. Bad arguments are made for pretty much any position possible. The fact that such arguments are being made somewhere isn’t relevant for fairly obvious reasons.
That’s a paywall, so I assume you have not read it. Here’s a jailbroken copy: “When What You See Is What You Get: The Consequences of the Objectifying Gaze for Women and Men ”, Gervais et al 2011 (Libgen; PDF.yt; Dropbox).
This paper inherits the usual defects of the ‘stereotype threat’ literature. It takes place in no-stakes situations, while stereotype threats have failed to generalize to any situations that actually matter, and blinding is questionable (they bring the subjects in, then “They also learned that they may be asked to report their feelings about themselves and others and to complete word problems”, and do math problems? Gee, I’m sure none of these undergrads recruited from psychology classes figured out what the real experiment was!) The results are also a little bizarre on their face: ”...the objectifying gaze also increased women’s, but not men’s, motivation to engage in subsequent interactions with their partner...the objectifying gaze did not influence body surveillance, body shame, or body dissatisfaction for women or men”. Huh?
And finally, this is social psychology.
That does not follow. If different disciplines have non-identical needs, then depending on the exact differences in distribution shape, the correlation between IQ, and the cutoff for success (see for example the table of r vs cutoff in “What does it mean to have a low R-squared ? A warning about misleading interpretation”) - not to mention the other variables which also vary between gender (Conscientiousness; degree of winner-take-all dynamics; expected work hours) - may well be sufficient to explain it. You’ll need to do more work than that.
See discussion here.
Sure. It is extremely difficult to test these situations in high-stakes situations for obvious reasons.
This is an intrinsic problem in almost all psychology studies. Is there anything specific here that’s worse than in other cases?
I don’t see what your point is. What do you find is bizaare about this and how do you think that undermines the study?
That’s a reason to be skeptical of the results, not a reason to a priori throw them out.
You are correct. The word clearly is doing too much work in my comment. At minimum though, the fact that other similar disciplines don’t have that situation even though they historically did is evidence that IQ variance is not all that is going on here. And that’s especially the case when many of those disciplines are ones like medicine that have taken many active steps to try to encourage women to be interested in them.
Now seen. Having read that discussion, I agree with Kaj there. Do you have any additional point beyond which you said to Kaj there?
It is not so difficult as all that: high-stakes tests are conducted all the time and gender is routinely recorded. I refer you to the WP article for how stereotype threat evaporates the moment it would ever matter.
It is not that bad a problem in most studies, and stereotype threat studies are particularly bad.
Their results make no sense in almost any causal model of how stereotype threat would work. What sort of stereotype threat has no effect on attitudes and body images and increases interest in co-workers, and how would you expect this to support the argument you made with regard to co-workers in the real world?
Indeed. So why did you cling to a weak reed?
No. I stand by the sum of my comments: that it is blatant post hoc rationalizations which contradict any theory a feminist would have made before seeing the actual data, which clearly supports an economic rather than pure bias account, and makes false claims about new CS students as well.
Which studies there are you referring to as being relevant? Note by the way that the study in question isn’t quite the same as stereotype threat in the classical sense anyways.
I’m not completely sure what model would actually do this but it could be something that causes them to think of themselves less as people doing math and more as people who are socially or sexually interested in others. But the fact that it didn’t have an impact on body image is strange, and needs further investigation. In the short version though, that should suggest that this study actually is more reliable: one of the most common criticism of psychology as a discipline is that the studies have way too high a confirmation of hypotheses rate. That’s been discussed on Less Wrong before. In this case, the fact that part of the study went against the intuitve hypothesis and went against what the authors explicitly hypothesized is a reason to pay more attention to it.
Because this is one of the very few studies that have looked at how sexualization impacts performance. There are a lot of stereotype threat studies (as you noted) but they don’t generally look at this. I’d be happy to rely on something else or change my opinion here if there were that many more studies.
So in your view, what precisely is the reason for the fact that the percentage of female CS students was consistently rising and then took a sharp drop-off? Also, why do you think Kaj disagreed with your position?
If you aren’t going to read the links I provided*, I’m not going to bother continuing. Both of those questions were answered.
* please note I have already gone above and beyond in not just reading your source material while you have not, but jailbreaking & critiquing that study, and also excerpting & linking contrary opinions & surveys
I read the conversation with Kaj and I read the links thank you very much. In that conversation you brought up a variety of different issues, focusing on the “practicality” issue but you give multiple different versions of that claim and I’m not completely sure what your primary hypothesis is. The primary claim there seems to be that the ups and downs on the graph mirror ups and downs in the market, but the primary link justifying that claim is this one you gave which doesn’t make any claim other than the simple claim that the graphs match without even showing that they do. The only bit there is there that is genuinely interesting evidence is the survey showing that women pay more attention to job prospects when considering fields which is not at all sufficient to explain the size of the drop there, nor the fact that law didn’t show a similar drop in the last few years when there’s been a glut of lawyers.
I don’t know where you are getting the second part of that claim from. But it is true I didn’t read every single link in the Kaj conversation, and I’m not sure why you think reading a single study is on the same scale as reading an additional long conversation and every single link there. So if you want to point to which of those links matter there, I’d be happy to look at them.
On the other hand, the fact that such arguments are used to intimidate anyone who dares question a certain position is relevant (possibly successfully remember what happened to Summers). In particular it affects what arguments we expect to have been exposed to.
Furthermore in Lewin’s case we have no idea what he actually did, thus the only evidence we have is that a committee at MIT decided what he did was bad. Thus to evaluation how much we should trust their conclusion it is necessary to look at the typical level of argument.
It isn’t at all relevant. To use a different example (coming from the other side of the poltiical spectrum)- one argument made against releasing the recent torture report was that anyone wanting it released was “anti-American” which is essentially the same sort of thing. The presence of such arguments is in no way relevant to any actual attempt to have a discussion about whether releasing the report was the right thing. No matter what position you discuss someone will be using bad arguments to intimidate people into silence. Rise above it.
The typical level of argument isn’t that when it comes to sexual harassment though. The typical level is a massive mix with some universities overreacting, and other’s underreacting. For every example of a university overreacting there’s an example of it underreacting. For example here.
But this also isn’t relevant for another reason: this entire subthread isnt even discussing the specifics of the Lewin case but a more general question of whehether such issues matter and are worth discussing. It is a red herring to go back to the original situation. But if you really do care about that situation, it might be worth looking at what Scott Aaronson has said on it, I’m curious if this adjusts your estimate at all that this is a minor situation being overblown?
I’ve been somewhat following the situation, and yes it is. The fact that you would claim otherwise cause me to update away from trusting other claims or judgements you make on the subject.
I didn’t see anything in the article that would adjust my estimate. The only thing there is that some who know told Aaronson that “this isn’t a borderline case”, given the kinds of things feminists consider “not borderline cases” these days that isn’t strong evidence.
I’m not sure a polite response to that, so let me just ask, given that I just pointed to an example that went in the other direction, maybe it is worth considering, just maybe, possibly, that you are vulnerable here to a combination of confirmation bias and what media sources you are using? Let’s as a start focus on a simple example: were you aware of the example I linked to above before I linked to it?
At this point, I think we may be having problems with radically different priors, part of which is that I give Aaronson enough credit that I don’t think he’s going to go the most radical end of the women’s studies department and ask for their analysis to get some idea of what happened.
Aaronson’s post also states that the incidents occurred online, and for that matter on the MITx platform, which caters to MOOC users, not actual MIT students. Given these factors, I just can’t see how MIT’s Damnatio memoriae towards Walter Lewin could be anything but an outrageous overreaction.
I’m not sure at all the relevance of your comment in the context of what we are quoting. The fact that this was not the regular MIT students has been known for a while. I’m also not sure what that has to do with my comment, since everyone here is in agreement that the removal of the videos was an overreaction. (However, I’m not at all sure how the fact that it was with MOOC users rather than regular students makes any difference whatsoever unless you are talking about a very marginal difference in legal liability.) What is the connection between your comment and the part of my comment that you are quoting?
He can only get information from the people who handled the case, who are likely to be SJW-types.
These issues are handled in general by university committees. Does your lack of knowledge on this fact cause you to update at all about how good your judgment is for such issues?
Also it is worth noting that “SJW-types” in most contexts is a group which is by and large restricted to certain parts of the internet or some parts of certaind departments on campuses.