In the Spoelhof College Center of Calvin University, there hangs a poster regarding the
gender pay gap. This is a rather divisive topic, especially considering its context in recent
years. The poster makes interesting choices on how to break down the data, such as
saying that there is still an unexplained gap of 7% in pay after controlling for “college
major, occupation, GPA, age, region, and marital status, one year after college.” However,
the poster’s evidence has multiple issues, such as sample size and the inability to control
for all variables.
gender pay gap. This is a rather divisive topic, especially considering its context in recent
years. The poster makes interesting choices on how to break down the data, such as
saying that there is still an unexplained gap of 7% in pay after controlling for “college
major, occupation, GPA, age, region, and marital status, one year after college.” However,
the poster’s evidence has multiple issues, such as sample size and the inability to control
for all variables.
The posted cited “The Simple Truth about the Gender Pay Gap” published by the
American Association of University Women (AAUW) which, according to their website, is
“the nation’s leading voice promoting equity and education for women and girls.”(1) The
7% statement is taken from an AAUW report titled “Graduating to a Pay Gap: The
Earnings of Women and Men One Year after College Graduation” published in 2012.
According to figure 10 from this report, there is an unexplained difference in pay between
men and women of 7% after controlling for things that have been shown to affect pay.
“Notes: The chart shows the pay gap among 2007–08 bachelor’s degree recipients in
2009. The column on the left shows the pay gap among all bachelor’s degree recipients
working full-time one year after college graduation. The column on the right shows the
pay gap among bachelor’s degree recipients working full time or in multiple jobs after
controlling for factors found to affect earnings. Factors controlled for include occupation,
economic sector, hours worked per week, multiple jobs, months unemployed since
graduation, undergraduate GPA, undergraduate major, undergraduate institution sector,
institution selectivity, age, region of residence, and marital status. This analysis excludes
graduates older than age 35 at bachelor’s degree completion. Source: Authors’ analysis
of U.S. Department of Education, National Center for Education Statistics, 2008–09
Baccalaureate and Beyond Longitudinal Study data.”(2) The report uses this information
to say that approximately one-third of the pay gap is unexplained, as the general pay gap
is 20% so the 7% that remains after controlling for factors that affect earnings is 35%of
the gap.
American Association of University Women (AAUW) which, according to their website, is
“the nation’s leading voice promoting equity and education for women and girls.”(1) The
7% statement is taken from an AAUW report titled “Graduating to a Pay Gap: The
Earnings of Women and Men One Year after College Graduation” published in 2012.
According to figure 10 from this report, there is an unexplained difference in pay between
men and women of 7% after controlling for things that have been shown to affect pay.
“Notes: The chart shows the pay gap among 2007–08 bachelor’s degree recipients in
2009. The column on the left shows the pay gap among all bachelor’s degree recipients
working full-time one year after college graduation. The column on the right shows the
pay gap among bachelor’s degree recipients working full time or in multiple jobs after
controlling for factors found to affect earnings. Factors controlled for include occupation,
economic sector, hours worked per week, multiple jobs, months unemployed since
graduation, undergraduate GPA, undergraduate major, undergraduate institution sector,
institution selectivity, age, region of residence, and marital status. This analysis excludes
graduates older than age 35 at bachelor’s degree completion. Source: Authors’ analysis
of U.S. Department of Education, National Center for Education Statistics, 2008–09
Baccalaureate and Beyond Longitudinal Study data.”(2) The report uses this information
to say that approximately one-third of the pay gap is unexplained, as the general pay gap
is 20% so the 7% that remains after controlling for factors that affect earnings is 35%of
the gap.
This data is interesting when compared with more recently published research about the wage
gap. According to research by the economist Federico Anzil, “job market forces and gender
preferences in relation to marital status and parenthood could explain almost all of the pay gap.
Most of the gap is not the result of gender discrimination.”(3) Anzil found that “in most
occupations, the main source of the pay gap lies in the difference between the number of hours
spent at work by women and men, and marital status and parenthood explain almost all this
difference in working times.”(3)
This leads to the question of why such an apparent conflict of data exists. Some might go so
far as to argue that Anzil looked over certain parts of the data as a result of his male privilege
and as such cannot see the problem clearly. However, I would argue that there are two
important differences in these reports and that after considering these differences, one should
tend to agree with Anzil’s research as opposed to AAUW’s.
gap. According to research by the economist Federico Anzil, “job market forces and gender
preferences in relation to marital status and parenthood could explain almost all of the pay gap.
Most of the gap is not the result of gender discrimination.”(3) Anzil found that “in most
occupations, the main source of the pay gap lies in the difference between the number of hours
spent at work by women and men, and marital status and parenthood explain almost all this
difference in working times.”(3)
This leads to the question of why such an apparent conflict of data exists. Some might go so
far as to argue that Anzil looked over certain parts of the data as a result of his male privilege
and as such cannot see the problem clearly. However, I would argue that there are two
important differences in these reports and that after considering these differences, one should
tend to agree with Anzil’s research as opposed to AAUW’s.
The first major difference resides in the samples used: regarding both the size and the dates.
The AAUW study used data collected in a 2011 study published by the National Center for
Education Statistics titled “2008–09 Baccalaureate and Beyond Longitudinal Study (B&B:08/09):
A First Look at Recent College Graduates” that had a sample size of “approximately 15,000
graduates.”(4)(3) In contrast, Anzil’s research used data from the 2017 American community
survey and his sample “includes 1,509,403 cases” and each case “represents anywhere from 20
to 1000 people in the complete population” according to e-mail correspondence with the
author.(6) This means that not only is Anzil’s data newer but it is also larger. Therefore it is more
likely to represent the United States of America that exists today.
The AAUW study used data collected in a 2011 study published by the National Center for
Education Statistics titled “2008–09 Baccalaureate and Beyond Longitudinal Study (B&B:08/09):
A First Look at Recent College Graduates” that had a sample size of “approximately 15,000
graduates.”(4)(3) In contrast, Anzil’s research used data from the 2017 American community
survey and his sample “includes 1,509,403 cases” and each case “represents anywhere from 20
to 1000 people in the complete population” according to e-mail correspondence with the
author.(6) This means that not only is Anzil’s data newer but it is also larger. Therefore it is more
likely to represent the United States of America that exists today.
This appears to be an insignificant point to make, but sample size is crucial for statistical
analysis. It is understood that the larger the sample size, the better its likelihood of reflecting the
population. In conclusion, because of the drastic difference in sample size, Anzil’s research is
more likely to actually reflect the United States and thus should be held in higher regard than of
the AAUW’s report.
analysis. It is understood that the larger the sample size, the better its likelihood of reflecting the
population. In conclusion, because of the drastic difference in sample size, Anzil’s research is
more likely to actually reflect the United States and thus should be held in higher regard than of
the AAUW’s report.
Furthermore, a side by side comparison shows just how small the AAUW’s sample size
actually is. According to Graduating to a Pay Gap, “The sample of approximately 15,000
graduates who responded to the 2009 survey represents the 1.6 million students who completed
the requirements for a baccalaureate degree between July 1, 2007, and June 30, 2008, in Title
IV-eligible institutions in the United States and Puerto Rico.”(2) This means that AAUW is
extrapolating information onto the entire United States from less than 1% of the population.(5)
This is seen as less likely to represent reality when compared to Anzil’s report with a sample
size of ~1.5M.(6) Even though AAUW’s sample is more specific in covering only recent college
graduates, the fact that it is only 1% of Anzil’s sample would lead one to think that Anzil’s report
gives a fuller view of reality as it is today.
The second major difference is the researcher’s choice of statistical analysis. Anzil chose to
use weighted medians for his analysis and this fact was mentioned in each graph presented.
This is important because a weighted median is a very good method of finding a central
tendency as it is less influenced by outliers than other statistics. As such, a weighted median is
beneficial for trying to find the common definition of the “average” when it comes to wages.
“Graduating to a Pay Gap,” however, does not state that it uses methods of analysis that are
robust against outliers, choosing instead to simply do a regression analysis on their entire
sample. You can argue that AAUW’s decision better reflects reality because its analysis includes
the entire sample. However, the point that “Graduating to a Pay Gap” is attempting to make is
that the “typical,” or average, woman is paid 7% less than her male counterpart and that this is
due to gender discrimination. In order to do this, a statistician must use methods that are robust
against outliers so that they are more likely to remove the effects of an individual’s
characteristics, situation, and/or luck to find that which is most likely to be true about “the typical
example of the group under consideration.”(7)
From a conceptual point of view, one issue with “Graduation to a Pay Gap” is that
“Graduating to a Pay Gap” does not understand what “value” means. “Graduating to a Pay Gap”
argues that “graduates can become their own best advocates by understanding what they are
worth and negotiating their first post-college salary and benefits.”(2) “Graduating to a Pay Gap” is
positing that a person has a certain, tangible and fixed “value” that could be increased if one
does certain things, such as getting an education. This would mean that if women are paid less
than men on average, then either the women are being paid less than they are worth or the men
are being paid more than they are worth. This follows the thinking of classical economists before
the late 19th century, and as such, this thought process fails to properly answer the
diamond-water paradox; the idea that while water is necessary for life, it is much cheaper than
diamonds, an item that does not actively keep a person alive. If this theory is correct, then water
should be much more expensive than diamonds as its intrinsic value to a person is exponentially
higher than a diamond. This thought process has been rejected by the majority of contemporary
economists in favor of the subjective theory of value. The subjective theory of value is “the idea
that an object's value is not inherent and is instead worth more to different people based on how
much they desire or need the object.”(8) With this theory, the diamond-water paradox is solved
very simply as a unit of an object has worth only when it is inputted by a person. Labor is no
different. A contract is a written enshrining of an agreement between an employee and an
employer of their exchanging of labor for money and other benefits. For an employer, however
much they spend, in terms of both a salary and benefits, in order to gain an employee’s labor is
how much they value said employee’s labor. There is nothing about one specific employee
intrinsically that an employer values, but an employee can, and should, attempt to convince an
employer of why the employer should increase their subjective valuation of the employee’s labor.
This leads to another issue with “Graduating to a Pay Gap” which is its arrogance in thinking
that it managed to control for every possible variable in the discussion of payment. This reflects
a broader arrogance of social scientists, especially economists, who think they can fully
understand the complexity of human interactions. According to “Graduating to a Pay Gap,” their
analysis controlled for “occupation, economic sector, hours worked per week, multiple jobs,
months unemployed since graduation, undergraduate GPA, undergraduate major,
undergraduate institution sector, institution selectivity, age, region of residence, and marital
status..”(2) While this might seem to be all-encompassing, there is some important data missing.
Not only does this fail to recognize the inherent subjectivity of valuation and wages, but it also
fails to account for the fact that men, on average, are more likely to negotiate their starting wage.
Some might argue that this is a result of gender discrimination because there is some
evidence that women who aggressively negotiate their wages are discriminated against in the
workplace. However, this argument does not recognize what academia understands about
gender differences in negotiation. According to Lisa Barron, a professor at UC-Irvine, there has
been research that would argue that men and women negotiate in different ways. “Researchers
have suggested that men and women approach negotiation differently because they view the
relational aspects of the negotiation differently (Greenhalgh & Chapman, 1995; Greenhalgh &
Gilkey, 1986; Halpern & McLean-Parks, 1996; Kolb, 1993).”(9) Dr. Barron did research which
“suggests that negotiators’ orientation toward requests plays a fundamental role in salary
negotiation; those oriented toward requesting the same are likely to make smaller salary
requests than others, which contribute to differences in outcomes. Not all men in this study
indicated that they were oriented toward requesting more than others. However, most women
were oriented toward requesting the same as, but not more than, others.”(9) This research seems
to describe the difference in pay, but not the reason for the difference. Barron says “[i]t is likely
that both psychological processes and internalized societal constraints regarding entitlement
affect women’s salary negotiation beliefs and behavior.”(9) One could argue that this is a result of
women being conditioned by society to be less aggressive and as such is evidence of gender
discrimination. However, one could also argue that as a sexually dimorphic species, there are
biological reasons for these differences in negotiation tactics.
This topic is too far outside the scope of this essay and as such will not be discussed further,
though it was brought up in order to discuss the amount of confirmation bias involved in the
discussion of the gender wage gap. Every person who has researched this topic, whether
intentionally or not, brought their presuppositions about the world with them. Therefore, one’s
presuppositions would be used to analyze one’s empirical data. An example of this can be found
in “Graduating to a Pay Gap.” The authors took data, “controlled for all relevant factors,” found a
gap, and then, because of their presuppositions about how the world works, they came to the
conclusion that the gap was solely due to gender discrimination. They don’t mention any
potential weaknesses of their sample or their analysis because it confirms their worldview.
This brings up the question of why this matters. Should Calvin University care that a poster of
dubious truthfulness is hanging in a department? The remainder of this essay will be my
arguments for why this school should care. The simple answer is that in keeping this poster and
the mindset behind it, the Sociology and Social Work department is actively undermining
Calvin’s mission statement. Calvin University’s mission statement is “Calvin University equips
students to think deeply, to act justly, and to live wholeheartedly as Christ’s agents of renewal in
the world.”(10) If one believes that the wage gap is the result of gender discrimination, then, as
argued previously, one’s worldview does not conform with reality. As such, one who holds a
worldview that does not understand reality has not thought deeply enough about reality so as to
actually understand reality as it is. This conflicts with Calvin’s mission statement as one cannot
“equip students to think deeply” if one cannot think deeply themselves(9). Because students look
up to their teachers/professors as authority figures, students tend to take that which their
professors say as factual. Therefore, if a professor misunderstands reality, then their students
will tend to also misunderstand reality as they are being taught from a faulty point of view.
In a similar manner, if one does not understand reality, then one cannot act justly. In order to
act justly, you first have to have an understanding of justice. In order to understand justice, one
must first understand reality. This is because if one starts with a faulty premise, then one cannot
come to a proper conclusion. Therefore, if one does not understand reality, one cannot “equip
students to...act justly.”(10)
An unintended consequence of the Sociology and Social Work department’s worldview is that
it harms the psyche of every person who has to interact with it. It does this by distorting people’s
sense of self-worth. This worldview distorts a woman's sense of self-worth because it makes
them believe that no matter what they do, they will never be “valued properly” because the world
is out to get them. Their sense of self could either be deflated or inflated, depending on their
personality. On the one hand, a woman could see this information and have a very pessimistic
attitude, causing her to develop an inferiority complex and think that her worth is less than it
actually is. However, one could also develop an inflated ego because if one is not being paid
that which one is worth, then one, obviously, is worth more than that which one is given. This
could result in developing an over-inflated ego and an attempt to force people to think that they
have more worth than they have. As a result, they may become bitter and enraged that they are
not treated the way they think they deserve.
This worldview could also harm and influence a man’s sense of self-worth because it
encourages the mindset that men are overvalued. This worldview perniciously tells young men
that anytime they attain something, it was not done through a combination of hard work, luck,
and/or skill. Rather, this worldview would suggest that they were given these things because of
their gender. This can have serious repercussions, especially when said person wrestles with a
lack of proper self-esteem.
Even if the wage gap was real, implying that the wage gap is the reason a man gets far in life
can still have serious consequences. This would be similar to telling a minority student that the
reason they were accepted at a prestigious university was because of their skin color and not
their academic achievements. This, of course, would result in nothing but a dampening of the
student’s self-esteem. Even if it were true that, as a result of affirmative action, a student’s
acceptance at a school hinges on their ethnicity just as much as their academic capabilities,
telling a student that that was the reason for their acceptance would result in much greater harm
than any potential good.(10)
This is important because as a result of the subjective theory of value, people cannot be
“undervalued” or “overvalued” as their “value” is determined by how people think they are worth,
which, when it comes to wages, is normally affected by experience, education, and/or how
much their prospective employer thinks they will benefit the company. As such, by entering into
a contract with another person, you are making an agreement about how much you are being
valued by the other person.
In conclusion, because of the complexity of human interactions and the inherent subjectivity
of human valuations, you cannot claim that the wage gap is a result of gender discrimination.
Therefore, by perpetuating this falsehood, the Sociology and Social Work department at Calvin
University is failing to affirm reality and is actively working against Calvin University's mission
statement.
Works Cited
- This is because 15,000 / 1,600,000 = 0.009375 or 0.9375%
