Inadequate questions leave statistics ‘flawed, incomplete’
This is my rebuttal to a letter to the editor published in last week’s Volante and written by Professor Emeritus of Economics Dennis A. Johnson, in which he attempts to tackle the “myth” of the gender wage gap. I believe that not only does his letter contain some pretty suspicious thinking, but worse, that it near-deliberately obfuscates an already tricky topic.
I will admit it is incredibly presumptuous of me, a junior-year English major, to be calling into question the economic acumen of a professor emeritus. But the tone of certainty in Johnson’s letter is the first place that I believe he went wrong.
Johnson did provide some numbers in his argument, however, they come from one source: The CONSAD Research Group. I have nothing bad to say about CONSAD; I read through the report and it’s not like there’s any explicit bias in there. However, Johnson’s reliance on the statistics of one group raises two questions: “How much research went into this letter?” and “How much do statistics really prove?”
The first question has a simple answer: If you are only looking at one source on a subject as complex and personal as a gender wage gap, you are not doing enough research. ‘But Sam,’ you ask, ‘The report has a foreword by the U.S. Department of Labor. Why would we have to look at anything else?’
The same Department of Labor also clearly states, and backs up with its own research, on its official blog: “Decades of research shows a gender gap in pay even after factors like the kind of work performed and qualifications (education and experience) are taken into account.” I am in no way assuming the superiority of one source over the other. What I’m saying here is for a subject as humanly intricate as the wage gap, less is not more.
My next concern is much more difficult to parse: How much do statistics really prove? We can get into epistemological arguments regarding statistics all day long but the problem here is not the numbers, it is the people interpreting them. Unlike Science, with a capital-S, which involves the creation and testing of hypotheses, the use of statistics instead looks at data that has already been measured. That means the only way to get a useful answer is to ask a good question. Statistics surrounding gender wages do not ask these questions.
The answers and, thus, questions that lead CONSAD and others to their conclusions are flawed, or at least, incomplete. It’s true, even a liberal estimate, like “The Gender Pay Gap: Have Women Gone as Far as They Can?” by Francine Blau and Lawrence Kahn, places things like “Occupation,” “Industry” and “Labor Force Experience” to blame for about 59.8 percent of the $0.23 difference in the raw data of the gender wage gap. However, it is important to not just stop at these statistics and let them justify the gap. Like “Inception,” we must go another level deeper. We must ask “Why do women seem to gravitate towards jobs that pay less?” This question would also do well to explain Johnson’s analogies of African-Americans in basketball and Jewish people in the garment industry.
The answer one will find on any site that claims to “debunk” the “gender wage gap myth” is women “naturally enjoy” child-rearing and staying at home and flexibility in work hours and most importantly letting men be the breadwinners. At the same time, men “naturally enjoy” dangerous jobs and being aggressive and competitive and protecting “their women.” An idea which comes up often is women consider their interest first and income second when looking for a job, while men do the opposite.
Perhaps I have a little too much faith in my readers, but I don’t think I have to explain why this sort of biological determinism is wrongheaded. Nor should I have to tell you allowing this sort of backwards thinking to influence economic and government policy is what got us into a mess like this in the first place.
Most importantly, even if one finds it totally cool to point to some sort of inherent preference for so-called “pink collar” jobs in women, there’s the fact that women in the United States still aren’t paid as much as men in the same field. According to the “Highlights of Women’s Earnings in 2011,” a report issued by the U.S. Department of Labor, out of 534 occupations surveyed, women only made as much as or more than men in seven fields. Seven. The percentages go lower and lower after that. At the lowest reported, “Property, real estate and community association managers,” women earned 60.6 percent of what men made in median weekly earnings.
There is a real chance I’m wrong in all of this, that I have been bowing to my confirmation bias and not doing enough research on the other side just to make sure. But if I am to be wrong, I’d rather be wrong on the side that doesn’t implicitly encourage hostility towards legitimate problems of inequality. Maybe that’s just me.
