Chapter 1 – How large is big?
2008/05/12 § Leave a comment
How large is big?
Writing for the students’ magazine of the University where I performed my doctoral studies in physiology I got myself into the crossfire of a student committee. It was a committee whose task it was to make sure that the boys and girls at University were treated the same way. Or to put it in somewhat fairer terms, their job was to make sure that no one was discriminated against.
I had started to write the column in this magazine because I always found writing fun and I was hoping to gain more experience in doing so. Mainly because I was hoping to increase my chances to become a science writer after my PhD studies. Fortunately, all I had to do was write a column that was under my total control. The only restriction being of course the amount of words I was able to use.
Right when I had started I had warned a close friend of mine that I might be outlawed at some point. The reason being that I intended to write about something that had to do with “feminism” and “gender issues”. I was not going to be too obvious about it and so I had decided to write about some more harmless topics for a while and open pandoras box once I had been established at the magazine and gained respect or at least some sort of acceptance.
Eventually the day came and I wrote about the topic that has thrilled me since earliest puberty: what makes a man want to be a man and what makes a woman want to be a woman. Even that question is a provocation already for your everyday Swedish feminist of course, since we are forced and we do not choose. But my article was much worse than that. The outrage was boundless.
I had dared to question the feminist dogma that we are indoctrinated and manipulated brainless-like subjects under the control of “the society” and “the media”. I had dare to argue and provide some evidence for the point that we might be biologically different in a way that makes us behave differently. This question is at the very core of the whole topic of “sex and gender” and “feminism and equality” of course.
I am not claiming that my argument was especially good at the time. However, it was not that bad either. But one thing for sure the counter argument was my first lesson in “feminism”. The lessons I learned can be described as follows:
1. Something that should not be true can not be true
2. If there is scientific evidence supporting something that should not be true, then the evidence must be questioned by questioning the motives of the individuals presenting and/or producing the evidence
3. We need not provide evidence, we need only question the opponent
I do feel like going on but I think that summarizes the source of the biggest nuisances. While there might be some curiosity on the readers’ part now as to what the conversation contained which occurred (as article, reader’s letter response, response to the reader’s letter… etc) I can assure you that I will cover all main arguments dealt with.
In a way, this conversation triggered the wish to write this book. Because I saw that there is real need to educate people in the biological evidence that we already have.
After being interested in the topic and studying it for many years I had become aware of an overwhelming amount of well performed studies which mostly pointed in one direction. What I had not come across was an ideology that would not be interested in evidence, but, as ideologies seem to require, rather just postulate and ignore or denounce evidence.
However, one question that was really relevant and that I had not really bothered addressing sufficiently up to that point was: how large is big?
What triggered the question was a very simple yet insanely provocative statement on my behalf. I had claimed that an individuals interest in something could influence their abilities to perform the subject of the interest. Furthermore, the abilities to perform a task could influence the choices we would make when it comes to choosing a profession, I argues. Or more clearly; if I am good at learning languages and translating from one into the other then maybe I am keen on doing something in my professional life that lets me utilize this faculty.
I can only guess, but do have the gut-feeling that I would have been on the safe side with my reckoning had I not gone done the slippery path and mentioned that this may hold true not only for human beings but for male and female too.
Of course instantly my argument was unacceptable because I might have been saying, though no one bothered if I was saying, that this was the core of more women than men in many professions requiring verbal skills. It was fair enough no one actually asked if I was trying to say just that. I was indeed implying that and am still doing so.
The real question arising from it was twofold: are there such differences and in case there are, are they large enough to make any difference?
The argument of the committee members for gender-justice basically argued that there were no such differences. And they said that the differences that some claimed to have observed in this respect were due to “society” influencing young people in their choices. Furthermore, they claimed that even the observed differences were too small to account for the differential amounts of male and female in certain jobs.
Well, I love to point out of course that there either are differences or there are none. The above argument reads in translation: there are no differences, and besides they do not matter. I am aware that I come across as mocking my opponents. But in fact, I only do so because I can.
Seriously though, this is a common strategy in bad arguments as you can experience them with people defending any ideology. However, while I will address this more later, let us for now stick to the indeed interesting question how large does something have to be to be big.
Some number crunching
In order to make things more comprehensible let us start with an example as we may find it in a real life situation. Consider the following situation. A company intends to employ 10 individuals for a certain job that requires some specialist knowledge. Let us assume that we have 100 individuals applying consisting of 50 male and 50 females. The company has all applicants perform a test and it turns out that the men have an average of 100 points in the test, while the women have and average of 110 points. If the company now decides to only take the 10 best performing applicants, would we expect them to male or female? Gut feeling may lead us to assume that the in average better performing women comprise the group of applicants.
However, the correct answer is: it depends.
What it depends on is what the distribution of results in each group looks like. If there is a “wider spread” of good and bad results in the group of males, then out of ten chose candidates for the job the likelihood is higher for them to be male, even though the females are the ones scoring higher on average. To put it in other terms; in the above example we assume that the group of women does perform better in average than the group of men. However, most women perform very closely to the mean of 110 points, e.g. in a range of let us say 90-120 points. The group of males on the other hand has an average of 100 points only but consists of individuals performing as badly as 70 points but also has a number of individuals who “balance” the bad performance by achieving 130 points. Now we can see how there can be more men getting the job. Simply due to the fact that we have some with more than 120 points outperforming any female in the group of applicants.
This result is certainly counter-intuitive. And it shows that we have to know much more than just mean results of tests.
Resulting from this consideration it is possible that small differences have a great impact. It is even possible that seemingly no difference has a great impact. And finally it is even possible that the outcome of a difference results in a consequence that is contrary to our initial expectation.
I cannot overstate the importance of this short paragraph. When we are dealing with groups of individuals not just minute differences of mean performance of each group can have an impact on who is hot and who is not. Even the distribution, i.e. how many under- and overachievers are in each group is important.
The overlaps between male and female groups are large for many investigated faculties and abilities. However, ultimately the differences that we are going to investigate in chapters to come, are sufficient to lead to effects large enough to be observed by anyone.
What has to be kept in mind is obviously that due to the overlaps in the groups the predicting power of studies and investigations on gender differences on an individual’s capacities is low. That means that while the differences can predict that for example men are physically much stronger than women on average the individual man can be much weaker than some women.
This fact has the effect that one statement or argument reoccurs in discussions about gender difference: “why group people in the first place? Maybe two randomly assigned groups of men (or females) would show similar differences?” The short answer is “no”. But maybe you are not content with the short answer, so here we go:
Sense and nonsense of categorization
It is somewhat startling that the feministic ideology is at work so strongly in some that they have convinced themselves into believing that men and women are equal. Their credo tells them that if we would group randomly assigned groups of men and perform a test, the outcome would be a difference between the groups that is similar to what can be observed in studies on gender difference.
I have not found any study investigating this. But that is no evidence for a conspiracy, rather for the fact that this is so obviously nonsense that no one bothers looking into it. However, a simple experiment of thought will help us out.
Imagine that you gather a couple of hundred kids from a school to perform 50 meter dashes. You assign the kids into groups via a simple coin tossing exercise. If you now record the average running times the groups performed you will not see much of a difference other than random variation.
This variation has no predicting power though. It can simply mean that you have coincidently assigned a larger number of younger pupils into one group than the other for example, so that the group with older pupils might end up with faster runners due to their age. It might also mean that you ended up with kids who perform more sport in their spare times than in the other groups.
Without any knowledge of these differences you could not predict in advance though who will perform better. If however, you decide to group the pupils by age it is going to be a rather straightforward exercise to predict that the older pupils will outrun the younger ones. A similar effect will occur if you group by sex.
That is one aspect of what science attempts to accomplish, i.e. to categorize in a way that helps predict outcomes. Scientists do that in medicine for example by grouping healthy people into one group and people with a certain disorder into another when testing a treatment. Anything else would not lead to the development of an applicable medical procedure.
The only reason I can see why some believe this rational should or could not be applied to sex and gender is that it does reveal differences. If you are interested in closing your eyes from how the world around you really is and prefer the realm of fantasy and ideology I suggest you stop reading now.
If however you prefer to know how things “really are”, then read on and enjoy the journey.
And now that the ignorant folks are gone I let you in on a secret: grouping by male and female has turned out to be suboptimal! It would make much more sense in many instances to group by hormonal content than by penis and vagina. However, the primary sexual organs have proven to be good enough predictors in almost all instances for the hormonal profile of a person. So for now, we accept this factum but will look into this more in the chapter about testosterone.
In conclusion we can say that it is well worth to look at differences, because they may make a difference. The seeming “size” is not the most important factor. What matters is if these differences produce observable differences.
As in the above example; at the end of a long day of grouping and running we assemble a list of the top ten runners. We might find that the times that separate the best male runner and the best female runner is just, let’s say, a second on 50 meters. That is a small difference. But to our surprise all the names on the list of the top ten runners indicate male runners.
On an individual basis we often might merely observed seemingly minor differences, but the outcome looking at things on a group level might be tremendous.
The neurologist Professor Sandra Witelson: “I think of myself as a scientist,” she says, “not as a male or female scientist.” Witelson, thinks there now is persuasive evidence that men’s and women’s brains “are actually different in some of the ways they are put together, anatomically and chemically. “That will upset some people”, she says, “because they assume that biology means things are immutable. But”, she adds, “the fact is that upbringing and other environmental factors play a tremendously important role in shaping the mind – a reminder that biology alone is not destiny”.