Sexism in Reporting

Today I read yet another article on sexism in STEM fields, written by someone with extreme bias and poor research skills.  The article used political incorrectness and gender statistics as evidence to support the claim that a vast majority of women in computer related fields have experienced workplace gender discrimination.*  Unfortunately, this dramatic misrepresentation of facts has yet again weakened the argument that women are facing more discrimination in technical fields than men.

To start with, I want to verify that women in software and engineering legitimately face unfair discrimination.  Sexual harassment against women is more common in these fields than in those where women are more evenly represented.  Women are typically paid less, even when they are equally productive.  Some men in these fields and some clients of these fields treat women as inferior.  American culture in general tends to discourage women from entering these fields.  The situation is legitimately bad.  Unfortunately, it can be hard to see this though all of the obviously wrong propaganda suggesting discrimination where it does not exist and implying there is more discrimination than there really is.

Sexual harassment is a problem.  In computer and engineering fields, mild sexual harassment in the workplace is not terribly uncommon.  In fields where women dominate, it is certainly far less common.  The evidence, however, seems to suggest that in computer and engineering fields, sexual harassment against women is still less common than it is outside of the workplace, and it is far less common than it is on American college campuses.  The fact is, U.S. culture has a problem with sexual harassment, and when you get a higher concentration of men in the workplace, sexual harassment seems to increase proportionally.  In short, sexual harassment in computer and engineering fields is not unusually high.  Yes, we should still be doing something about it, and it is still a form of discrimination against women, but it is stupid to suggest that this is a problem unique to these industries, and it is even more stupid to suggest that something that is so much worse outside of the workplace is solely the responsibility of managers of a very small percentage of the population.  If we want to eliminate sexual harassment in computer and engineering workplaces, we need to focus on U.S. culture, because that is the source of the problem.

It turns out that there is a legitimate reason women are paid less in computer and engineering fields.  This is not always fair, but most of the time it is!  It is true that some employers routinely pay women less without justification.  It turns out this is much rarer than most of these news articles suggest though.  A number of recent studies have shown that women approach work differently from men.  The most consistent finding is that women are far more likely to work only their scheduled hours and take more time off.  For many, this is about child care, but this is not always the case.  Women have reported reasons for this that include wanting more general family time (where child care is not a factor) and wanting more personal time.  Many women have said that they have little loyalty to their employers, putting friends, family, and other commitments far above their job.  Men tend to rate their employment much closer to family, and some even put it above family.  Men are far more likely to work when sick, skip vacation time entirely, and work longer than required.  It has been suggested that men tend to spend more than 20% more time working than women in the same position, on average.  Now, there are women who work as much as men on average, but they are fairly rare, especially in computer and engineering fields.  So why are women reporting lower pay as discrimination, if they clearly work less?  The fact is, it is not clear that they work less.  Most businesses don't publish hours worked for every employee, for privacy reasons and because they may not actually track hours for salaried employees.  Those who leave a 5pm on the dot are not there to see others working several hours later.  Those who take more time off may not realize that other employees are always still working during that time.  Women who take holidays off to be with family may not realize that most of their male coworkers are working through the holidays.  Men are also more likely to do work during their time off at home, which is also difficult to see.  An employer may easily notice that the male workers are more productive than female workers but not fully understand that it is because the men are spending more time working.  This can lead to a subconscious bias favoring paying men more and hiring more men.  And this can lead to unfair discrimination, in the rare cases where a woman does work as much as men and is equally productive.  Recognizing this bias can help employers make sure they treat employees more fairly, but most activists would rather bury the evidence that women tend to work less, instead of admitting to it and eliminating the resulting bias against women who work as much as men.

Male workers and clients sometime treat women as inferior in these fields.  This is certainly a bad thing, but there may be reasons for it.  This may be the result of male workers noticing female workers leaving several hours before they do.  The lower productivity of women who work less could be noticed by clients.  I strongly suspect, however, that most of this is just pure discrimination.  Even if they do work only the required hours, there is no call for male employees to treat their female coworkers poorly.  And if female workers are not living up to the expectations of clients because they are only working scheduled hours, employers need to make sure those projects have enough hours worked on them, even if it means assigning additional workers to them.

A lot of the blame is on U.S. culture though, not specifically on workplace cultures.  It is pretty dumb to expect the computer and engineering industries to fix a problem with U.S. culture that goes far beyond their control.  These fields should be expected to discourage workplace discrimination against women, and they should be expected to discipline obvious workplace discrimination, but they should not be blamed for the culture that is creating it.  On the other side, we need to make sure women in their fields are aware of how their work habit affect their productivity.  It might bring up privacy concerns for businesses to publish hours worked by individual employees, but there is no issue with publishing averages.  Perhaps if female employees saw that the average time worked was 60 hours a week, she would not be so bothered that she is getting paid 10% less for her 40 hours (33% less work) a week.

This might sound like I am suggesting women should work more if they want fair treatment.  This is not what I am saying!  What I am suggesting is that we need to do our research so we can understand where claims of discrimination are legitimate and where they are based on incomplete information.  Personally, I tend to be more like women when it comes to work/family balance.  I would prefer to work a 20 hour a week job than even 40 hours.  I am even willing to do so for $40k a year, which is less than half of the $100k that is common for my field (and more like $120k that I am actually worth).  I cannot find a job that fits these requirements though, so instead I teach advanced undergrad college courses as an underpaid adjunct professor.  The fact that I cannot find a job that allows for the work/family balance I want is evidence that part of this problem is our society's work culture.

In my opinion, the best solution to the pay problem is to return salaried pay to its original purpose.  Salary was invented for part owners and upper managers in a company.  There were two things that separated salaried workers from hourly workers.  One was a vested interest in the success of the company, and the other was some degree of representational power for the company.  The idea was that people who get paid based on the success of the company have enough interest in its success that overtime work will be inherently compensated through dividends and increasing value of the company.  This is why laws protecting unions and union workers do not protect managers and often do not protect salaried workers (representational power for the company is a conflict of interest with being a member of a union of workers for that company).  (This is beginning to change.)  President Obama even suggested that we need to legally enforce the original assumptions applied to salary.  If this was done, a vast majority of computer and engineering jobs would become hourly.  This would hopefully eliminate the pay discrepancy between men and women, because most employers in these fields seem to be trying their best to pay employees fairly.  On average, men would still be paid more (in fact, much more, because the extra hours would be subject to overtime pay, and because the current pay difference is far less than the difference in hours worked), but it would become very clear that the higher pay is the direct result of working more hours, not because the hourly rate is any different.  This would also bring a few other benefits.  Many employers would reduce the extra hours worked by male employees, forcing them to have better work/family balance and opening up new positions, reducing unemployment.

The fact is, right now most arguments about sexism in certain STEM fields is filled with red herrings and lies designed for shock value.  In reality, most of the things that are commonly discussed are not actually legitimate problems.  Things like pay discrepancy can easily be explained by differences in average hours worked, but when they get all of the focus, they overshadow the real problems.  The result is, every article on this topic gets filled with comments easily discrediting the claims of the article, and unfortunately that discredits the valid argument that women are treated unfairly in the first place.  As long as we are focused entirely on problems that mostly don't exist or that are part of U.S. culture, not work cultures exclusively, there will continue to be so many holes that few people take the real problems seriously.  We need two things.  First we need better research, and second, we need to quit letting political correctness get in the way and start discussing the real problem, instead of using lies and incomplete information for shock value.


* It turns out that if you assume computer and engineering fields have less women because fewer women enjoy that stuff, reports of discrimination are about equal to the national average of all workplaces.  In other words, to prove that fewer female employees is evidence of discrimination, you first have to prove that women who want to go onto those fields are not able to due to discrimination, and so far no one has successfully proved this.  The best evidence we have is that in the '90s, those fields were 30% female and now they are only 25%, but that only suggests 5% additional discrimination, not the 25% that is based on the assumption that women are equally interested in these fields as men.