Approximately a month ago, I was fated to come across this article in The Washington Post detailing how “The 10 richest women in the world aren’t entrepreneurs, but heiresses.” Indeed, that is the article’s exact headline. It sparked my curiosity, and caused me to wonder how far down the list of billionaires one had to go in order to find a “self-made” woman—a woman who was both a billionaire and an entrepreneur, rather than the fortunate heiress of a dead rich man. To find out, I went to the unquestioned authority on rich people: Forbes.
After paging through the tiny biographies of more than 500 billionaires consisting of both sexes, I discovered that of those 500 billionaires, only four women could legitimately be termed “self-made,” the first of which only cracked the list at ranking #240 (Chan Laiwa, $6.2B). The second appeared only eight slots later (Jin Sook, $5.9B), yet was co-listed with her husband, Do Won Chang. Dubious as this was, I gave her the benefit of the doubt. Afterwards, the next self-made woman did not appear until #360, and still the next did not appear until ranking #481, respective gaps of 112 and 121 ranking slots between them and the previous self-made female billionaire. The rest of the women in the top 500 rankings were invariably heiresses.
On discovering this, I ruminated on Facebook that there was a seemingly established pattern, that for approximately every one hundred and eight men, there was only one woman who could claim to have an equal measure of talent with the least talented man of the preceding one hundred plus men. Further, I hypothesized that while civilization depended on women, the quality of our civilization depended on men. Outrage ensued, primarily and unsurprisingly from women. Chalk it up to female insecurity.
I was predictably subsequently attacked and vilified for daring to insinuate that women were overall less talented than men. One attack went so far as to attempt to disqualify my observation based on my (relative) youth and apparent lack of success compared to women my age (both of which were not only irrelevant to the subject, but respective examples of ageism and a form of credentialism based on experientialism). Only one of my several attackers managed to address my observation with any intelligence rather than a purely rhetorical knee-jerk emotional reaction. She is to be commended, and it is primarily due to her that I took what little free time I have these days to compile a few of the studies I’ve read in the recent years and which eventually led me to make my controversial observation.
Here I must make a note. Women are not inferior to men or less valuable. I will never assert this position simply because it is not true. Yet I do and will continue to assert that women have naturally inferior skills in certain areas, and vice versa. I do not conflate “all women” and “all men” with women and men in a more general sense. This would be a grave mistake, and if I were to make it I would not blame anyone for discounting anything I might subsequently say on the subject. The fact remains, however, that I have not made this mistake, and therefore am worthy of your attention on this count. What follows is a more detailed defense of my position.
First, I would like you to look around. Note the things in your house, the things in the coffee shop you frequent, the bookstore you go to. Wherever you are at this moment, mentally calculate how many of those items were designed and created by men and how many by women. Obviously we cannot know for sure what was made by a man and what was made by a woman, but it is not too hard to make a guess. I would submit to you that 85% or more are products of men. The computer or phone from which you’re reading this article was created by engineers, a field comprised of 85% men. The walls, surrounding you were created and built by civil engineers and construction crews respectively, the former field comprised of 85% men and the latter comprised of 97% men (the construction field itself is 91%, but construction crews are 97% for obvious reasons). The tables and chairs, couches and sofas, bookshelves and beds which give you comfort everyday were created and designed by carpenters, comprised of 98% men. The dishwasher, electric or gas oven, refrigerator, microwave, blender, air conditioner and heater, the electrical circuitry which lights your home and lets you use all the aforementioned appliances… all designed and installed by men almost without exception. The car you drive, the petrol you purchase, the roads you drive on, the pots and pans and utensils you use for cooking and dining, the food you eat… all produced by mostly men.
Now ask yourself… what would you have without men in the way of comfort and convenience?
Probably Jack. Squat.
The main opposition to this conclusion is that many of these things were being created (furniture goes back hundreds of years, after all) before women began to enter the workforce in significant numbers, or that there is still progress which needs to be made—ergo, if women had been or were currently in the workforce in greater numbers, the credit could not be attributed so definitively to men. Unfortunately, this objection in its myriad forms is absurd, and is statistically and demonstrably false. We could say this for anything: IF X was instead Y, THEN the result of X would be a result of Y. Yet X is not Y, and there is no possible way to test for the purely hypothetical case. It is akin to saying that if men could give birth as women do, then more births would be attributable to men. The argument flies in the face of reality and is therefore dismissed immediately as the pure fantasy it is. And, too, statistically speaking, it simply has not occurred. To show this, I present Exhibit A.
Exhibit A is a study of women’s representation in 60 different occupations from 1972 to 2010, and the study’s results are astounding. Consider the following graph extracted from the study:
Figure 1. Mean percent of women working in 60 occupations as a function of year. doi:10.1371/journal.pone.0095960.g001
Starting from approximately the year 2000, the percentage of women in these 60 different occupations leveled out at around 42.5%, and has stayed there for an entire decade. One who believes women are identical to men in skill and ability will look at these numbers and find them grossly unjust—as half the population, women also ought to make up half the workforce! The Patriarchy is so evil!
But don’t be so quick to get upset. A Gallup poll in early 2012 discovered that 14% of women compared to 6% men are stay-at-home parents with children under 18, and 31% women and 24% men have no children under 18 but are still unemployed. These are differences of 8% and 7% respectively, consistent with the percentage of women needed to make up the missing numbers and make the workforce an exact 1:1 ratio of men and women.
This can hardly be a coincidence. Nor could I have gotten better numbers if I had forged them. I would even go so far as to make the prediction that if one were to examine the unemployment rates from 2000 to now, the average breakdown of men and women who are unemployed would have a difference of about 16%, with approximately 58% and 42% of unemployed men and women respectively, and which would also be consistent with the actual make up of the workforce.
The only conclusion one can make from this is that a significantly greater portion of woman are not looking for work, and an equally greater portion of men are looking for work, resulting in a workforce which reflects that status. From it, we can infer that the cultural wars which got women into the workforce to begin with have succeeded as much as they possibly can, and have in fact, stagnated from at least 2000-2010. The first question this raises is, why have they stagnated? The second question is, how does this relate to men and women’s differences in skill sets?
I’m not here to answer the first question. The second question, however, is of vital importance. By the above, we can see that the percentage of women in the workforce has not changed for at least the decade of 2000-2010. In other words, all the women who are either employed or looking for work have already entered the workforce. No more are forthcoming. If this is so, the credit for any work done by this workforce within that time period cannot be readjusted to fit a hypothetical or ideal model. More credit must by necessity go to men, simply for the reason there are more men in the workforce. Men do 58% of the work, and women do 42%.
I now present you with Exhibit B.
Exhibit B is taken from the same study as Exhibit A, and deals primarily with sex segregation within occupations. That is, what kind of jobs do men and women occupy? From the study:
There are undoubtedly other important job characteristics that contribute to sex segregation as well. Two fundamental dimensions of occupational variation that have been much studied by vocational interest and individual difference researchers are the people-things dimension and the data-ideas dimension –. The first dimension taps the degree to which occupations deal with people and their psychological dynamics versus inanimate things and mechanical systems. The second dimension taps the degree to which occupations entail routine record-keeping and data management versus creative thinking and the use of intelligence. While women and men do not differ much in their preference for ideas-oriented versus data-oriented jobs, they do differ substantially in their preferences for people-oriented versus things-oriented jobs, with women expressing greater preference for people-oriented jobs and men for things-oriented jobs , . This suggests that occupations’ positions on the people-things dimension may predict their degree of sex segregation, but occupations’ positions on the data-ideas dimension may not.
Later in the study:
People-things orientation accounted for slightly more variance than status did in 1972 (24 versus 19 percent, respectively). However, by 2010 people-things accounted for more than seven times as much variance as occupational status did (36 percent versus 5 percent, respectively). Thus, as women increasingly entered high-status occupations from 1972 to 2010, job status became an increasingly weak predictor of women’s participation in occupations, while occupations’ people-things orientation became an increasingly strong predictor.
I will not get into a detailed explanation of their methods, as that can be found in the paper itself. What is important for our purposes is the percent of variance which can be used as a predictor in the differences between men and women’s preferred occupations. In 2010, only 5% of total variance was attributed to occupational status. A high status, high paying job was a poor indicator of whether the person working that job was a man or a woman. In contrast, the people-things orientation was a strong indicator. The sex of engineers, machinists, scientists, mechanics, programmers, electricians, welders, construction workers, farmers, and mathematicians were easier to predict because they were usually men. Essentially any job involved in producing material goods, or advancing technology or scientific knowledge were dominated by the male workforce. On the other side of things, the sex of photographers, secretaries, librarians, nurses, receptionists, real estate agents, waiters, hairdressers, school teachers and social workers were more likely to be women.
And no, this is not because of discrimination or negative expectations in those fields (though those both exist to some small degree) and here is the reason why:
Multilevel linear modeling (MLM) analyses showed that women increasingly entered high-status occupations from 1972 to 2010, but women’s participation in things-oriented occupations (e.g., STEM fields and mechanical and construction trades) remained low and relatively stable.
The gender make up of construction crews hasn’t changed much in 40 years. Nor has the make up of workers in STEM fields. With the increasing number of women who were flocking to jobs for the last 40 years, 40 years of, we are told, terrible gender discrimination, there was hardly any increase in women in these fields whatsoever. In contrast, so many women have obtained high status jobs that it’s difficult to predict what gender fills them, when 40 years ago it would have been easy. If women are just like men, having the same talents to equal measure, then there is absolutely no reason to think that women could not break into these fields in the same way they’ve broken into the higher occupational ranks and high status jobs.
And yet they have not. Why?
There is a very simple reason, of course, but no woman really wants to hear it, and some men aren’t particularly keen on it either because they’re both overly invested in the idea that women are equal to men instead of complementary. The simple reason is that women are not as suited to those jobs as men are, and their suitability is derived from their natural state. While I have not absolutely shown that this is the case, I have, I think, made it abundantly clear that my second hypothesis was correct, namely, that our civilization depends on women and the quality of our civilization depends on men. Men produce goods, specifically food and shelter and technological improvements, and women produce children. Neither are inferior to other and neither are as suitable to the other’s natural tasks. The Forbes’ Billionaire list, as surprising as it is unlikely, indicates and supports this very conclusion.
Now to defend my first observation, that for approximately every one hundred and eight men, there was only one woman who could claim to have an equal measure of talent with the least talented man of the preceding one hundred plus men. Our first order of business on this matter is to push aside the obvious: that men are physically superior to women for nearly all functions except for child birthing and rearing. Any argument against this is futile as it is observably true and easily proven. There is a reason why construction crews are still 98% men, after all.
Our second task is to find out if men are more suited to STEM and related fields than are women. This is not so easily done, but it can be done. To begin doing so, I first present you with Exhibit C.
From a National Bureau of Economic Research (NBER) 2009 study, “An Empirical Analysis of the Gender Gap in Mathematics”:
Figure 1 plots the gender gap on the mathematics and verbal components of the Scholastic Aptitude Test (SAT) – over the past forty years. On the math section, female scores are on average 0.30 standard deviations lower than male scores; on the verbal portion there is no clear gender difference (College Board 2007). 1 An important shortcoming of the SAT data is that the population taking the test is not representative, and sample selection may occur differently across gender. For instance, since college attendance rates are presently higher for females, the female sample of SAT takers may be drawing more heavily from the middle or left tail of the ability distribution. Data from the National Assessment of Educational Progress (NAEP), a nationally representative sample that does not have sample selection problems, also shows boys consistently outperforming girls in fourth and eighth grade over the last two decades, though the magnitude of the gap is smaller (Lee, Grigg, and Dion 2007). The bulk of the evidence in the past 50 years suggests that the gender gap in mathematics does not exist before children enter school, but is large and significant in the middle school years and beyond. For instance, in a meta-analysis of 100 studies with a total sample of more than 3 million students, Hyde et al (1990) found a .29 standard deviation gender gap in math in high school.
Please make note of this statement: “The bulk of the evidence in the past 50 years suggests that the gender gap in mathematics does not exist before children enter school, but is large and significant in the middle school years and beyond.” Many have taken this to mean that boys and girls start out with equal potential with regards to mathematics which is later destroyed via gender-specific expectations and barriers. While that is not necessarily incorrect, I believe there is a more simple explanation: When everybody starts at zero, everybody is equal.
A simple analogy should suffice to make this explanation more transparent. At the beginning of a basketball game, both teams start off at zero points, at equal spots. No one has taken a shot yet. Nothing has happened. One team is made up of slightly shorter, slower players and the other team is made up of slightly taller, faster players. Everything else, shooting skill, ball handling, passing, etc., is equal except for disparity in speed and height. Then the whistle blows, and as the game goes on, the gap in score begins to widen from nothing to a little, to a lot.
This is exactly what happens for men and women with math. As schooling continues, men’s natural proclivities began to widen the gap, from zero all the way to 0.30 standard deviations. In a footnote, the NBER paper further states:
Among elite achievers, these differences are even more pronounced. Men outnumber women by more than two to one above the 99th percentile in SAT mathematics scores (College Board 2007). Males also score four percent higher on AP calculus exams and 6 percent higher on AP science exams (Freeman 2004, College Board 2007).
The paper continues:
The patterns on math tests are especially striking when one considers that females either systematically outperform males or have made enormous gains on many educational dimensions. The high school dropout rate is 28% for females compared to 35% for males (Greene and Winters 2006). As noted by Goldin, Katz, & Kuziemko (2006), in 2003 there were 1.35 females graduating from four-year colleges for every male. In stark contrast, in 1960 there were 1.6 males graduating from 4-year colleges for each female. In 1970, women made up only 9% of combined Medicine, Dentistry, and Law degree recipients. Thirty years later, women accounted for 47% of full time, and 44% of part-time students pursuing such degrees (Freeman 2004). Women make up 45% of all doctorate degrees (Freeman 2004). A 2000 study, commissioned by the U.S. Congress, found that “[t]he large gaps in educational attainment that once existed between men and women have in most cases been eliminated” (Bae et al. 2000).
With such awesome changes in education, how is it possible that science and math are the only areas where females cannot seem to catch up? There is a simple answer, which ironically is just outside the scope of the author’s paper. The emphasis is mine:
Due to limitations of the data, we can test only a subset of the possible socialization theories for the divergent trajectory of girls’ math scores in the early years of school, and none of the biological explanations. Among those hypotheses that we can test, we fail to uncover compelling support for any of them.
Among those possible socialization theories which they tested were: family background, school and neighborhood characteristics, teacher and parent assessments and expectations, parent educational prestige levels, and socioeconomic status. And there was nothing. No compelling support for any of them that would help explain the gender gap in mathematics and science. Conveniently left out of the inquiry, however, was a biological explanation. This concludes Exhibit C.
Finally, I give you Exhibit D.
In this study conducted by the Organisation for Economic Co-operation and Development (OECD), the researchers examine mathematics and reading gaps in different countries in an attempt to determine why girls tend to have better reading scores than boys, and why boys tend to have better math scores. From the Executive Summary:
Reading proficiency is the foundation upon which all other learning is built; when boys don’t read well, their performance in other school subjects suffers too.
Indeed. If reading is the foundation of learning, then why, if girls tend to have better reading scores than boys, do boys still outperform girls in mathematics? The study attempts to explain this by citing “math anxiety” in girls.
In the large majority of countries and economies that participate in PISA, among high performing students, girls do worse than boys in mathematics; in no country do they outperform boys at this level. In general, girls have less self-confidence than boys in their ability to solve mathematics or science problems. Girls – even high-achieving girls – are also more likely to express strong feelings of anxiety towards mathematics. On average across OECD countries, the score-point difference in mathematics performance between high-achieving girls and boys is 19 score points. However, when comparing boys and girls who reported similar levels of self-confidence in mathematics and of anxiety towards mathematics, the gender gap in performance disappears.
PISA reveals that girls tend to do better when they are required to work on mathematical or scientific problems that are more similar to those that are routinely encountered in school. But when required to “think like scientists”, girls underperform considerably compared to boys. For example, girls tend to underachieve compared to boys when they are asked to formulate situations mathematically. On average across OECD countries, boys outperform girls in this skill by around 16 PISA score points – the equivalent of nearly five months of school. Boys also outperform girls – by 15 score points – in the ability to apply their knowledge of science to a given situation, to describe or interpret phenomena scientifically and predict changes. This gender difference in the ability to think like a scientist may be related to students’ self-confidence. When students are more self-confident, they give themselves the freedom to fail, to engage in the trial-and-error processes that are fundamental to acquiring knowledge in mathematics and science.
It is no secret that stress and anxiety having impairing effects on reasoning, but the OECD stretches its effects when recalling that, (1) as the organization states, “Reading proficiency is the foundation upon which all other learning is built,” and (2) that girls outperform boys in reading by nearly 38 points.
As results from PISA have shown, girls do very well in school, too. In all countries and economies that participated in PISA 2012, girls outperformed boys in reading by an average of 38 score points (across OECD countries) – the equivalent of one year of school – as they have done consistently throughout all the PISA cycles since 2000. Boys, however, continued to outperform girls in mathematics in 38 participating countries and economies by an average of 11 score points (across OECD countries) – equivalent to around three months of school.
Hence, “when comparing boys and girls who reported similar levels of self-confidence in mathematics and of anxiety towards mathematics,” and who subsequently tested equally well on mathematics, the reading gap between girls and boys must also be taken into account—something OECD failed to do. Consider the following:
The data in Figure 6.2 suggest that trends in the gender gap in performance in different subjects are associated. Countries where girls became better readers between 2003 and 2012 are also generally the same countries where girls improved in mathematics during the same period. For example, in Finland, the gender gap in mathematics, in favour of boys, narrowed by 10 score points between 2003 and 2012. Over the same period, the gender gap in reading, in favour of girls, widened by 18 score points. In Greece, between 2003 and 2012, the gender gap in mathematics, in favour of boys, narrowed by 11 score points while the gender gap in reading, in favour of girls, widened by 13 score points. Similarly, in Sweden during the same period, the gender gap in mathematics, in favour of boys, narrowed by 9 score points while the gender gap in reading, in favour of girls, widened by 14 score points. Among partner countries and economies, similar trends were observed in Macao-China and the Russian Federation (Tables 1.2b and 1.3b).
These results, and the evidence developed in the context of Chapters 2 and 3, suggest that, in general, the gender gap in mathematics tends to be narrow when girls are good students in all subjects. But the factors that help to narrow the gender gap in mathematics also tend to enlarge the gender gap in reading, in favour of girls. Are gender gaps a “zero sum game”, in which education systems, schools and families have to choose whether to create an environment that promotes either boys’ performance or girls’ performance; or are there policies and practices that manage to narrow – or eliminate – all gender gaps in performance simultaneously?
From the first paragraph of the above quote, we can see that for each point scored in reading, we can add from 0.56 up to 0.85 points in math. Remember, girls scored on average 38 points higher than boys on reading. This should add anywhere from 21 to 32 points to their math scores. Only, boys still perform on average 11 points better than girls in math. Even when we eliminate “math anxiety” for girls, girls only manage to score on par with boys. By all accounts, they should be outperforming boys by at least 21 points. Yet they are not.
The OECD continues with their assessments:
Results from the PISA 2009 assessment of reading suggest that a large share of gender differences in reading performance may stem from disparities in how much boys and girls read for enjoyment and in how much boys and girls engage in reading activities. Indeed, the assessment found that if boys enjoyed reading to the same extent as girls do their reading scores would be 23 points higher, on average across OECD countries (Figure 2.11 and Table 2.9k).
Add another 13-18 points to boys’ math scores, if they enjoy reading as much as girls. This means boys, if they enjoy reading as girls do, would score 24 to 29 points higher on mathematics taking into account female “math anxiety” and 13 to 18 points higher sans female “math anxiety.”
There is no middle ground here. Boys are naturally better at math. Period. And by extension, they will also be better at math-heavy science and “thinking like a scientist.” The differences in scoring and thinking cannot be explained by the very real female anxiety which exists. The math simply doesn’t add up.
In summary, my initial observations extrapolated from the Forbes list of billionaires has large quantities of supporting evidence behind them. Does only one girl compared to a hundred boys have comparable talents? It is likely an exaggerated ratio due to the nature of billionaires, but the data on hand seems to point that way, and sometimes explicitly.
From Exhibit C:
On entry to kindergarten, girls make up 45 percent of the top five-percentiles in math test scores; by the end of fifth grade just 28 percent of the top five percent are female. Girls are underrepresented in the bottom tail of the math distribution in kindergarten, but overrepresented in the bottom tail by fifth grade.
From Exhibit D:
Among the top 10% of students in mathematics performance, the gender gap averages 20 score points; and among the top 10% in science, boys score an average of 11 points higher than girls.
Also from Exhibit D:
PISA finds that while boys outperform girls in mathematics, on average, in many countries and economies the gender gap is much wider among top-performing students than among low-performing students (Table 1.3a). In the large majority of countries and economies, high-performing girls do worse in mathematics compared to boys; in no country do they outperform boys at this level, and the magnitude of the gender gap is much greater than it is among students at an average level of performance.
However, even in science there is a sizeable gap in favour of boys among top-performing students. This is a troubling finding, as some believe it is responsible for the under-representation of women in STEM occupations (Summers, 2005; National Academy of Sciences, 2006; Hedges and Nowell, 1995; Bae et al., 2000).
Ultimately, this is in keeping with the list of billionaires, which is essentially the top performing percentage of the world’s population. Within that top percentage, there will be more men than girls, with wider gender gaps than among low performing members of the population. The cases are identical in nature, simply because it is the nature of men and women. I should add that there is nothing inherently wrong in this state of nature—it simply is.
Or you can chalk it up to female insecurity.
Either way, men have the natural advantage, and I stand steadfastly by my original statements.