Some months ago I found that I had been put on an email list for Linda Stein, an artist and warrior for gender justice. I am not sure how I got on the list; I suppose I had expressed my pro-gender-justice attitudes somewhere, and someone noticed and put me on the list. In any case, I enjoy the occasional emails I get from Stein and I found myself particularly intrigued by the subject line of the most recent email, "Are Men More Opinionated than Women?"
It turned out that this subject line referred to an article Stein recently published in the Huffington Post, "More Female Voices Must be Heard: New York Times Letters to the Editor." The article began as follows,
"On Monday, January 2, 2017, the New York Times presented an all-male lineup in its Letters to the Editor section – not one female.
"I had been noticing this gender imbalance in the Op-Ed section of the Times, and was prompted, some years ago, to send the newspaper a letter decrying this inequity. It was not published or acknowledged."
Stein's article went on to describe her investigation of three months' worth of Letters to the Editors in the NYT. Her data-gathering confirmed her impressions: Of the 794 letters published in January through March of 2017 where gender could be identified, 63% were written by males and 37% by females (the gender of 5 letter-writers could not be determined).
Like Stein, I was troubled by this imbalance: These statistics violate my intuitions about fairness. In a country where 49% of the population is male and 51%, female, it seems to me that female opinions in one of our leading newspapers are seriously underrepresented. Like Stein, I began thinking about possible reasons for the imbalance and how it might be corrected.
However, I was interested in Stein's study, not just as a social issue, but as a case of what has been called "the psychology of science" (Singer, 1971; Feist, 2011). Allow me to explain what the psychology of science is before looking at its application to understanding the gender imbalance in opinion letter writing.
An Interlude about the Psychology of Science
The psychology of science recognizes that scientists, as human beings, are subject to the same general psychological laws that describe all human thought, motivation, and behavior. Scientists are not simply unbiased observers and rational calculators, although science as an enterprise strives for unbiased, rational accounts of reality. If we want to understand how scientists actually behave, we sometimes need to include personal, psychological factors (Johnson, Germer, & Efran, 1988).
I was first introduced to the psychology of science in an undergraduate course where I was required to read Thomas Kuhn's The Structure of Scientific Revolutions. From this book and another book that apparently inspired Kuhn (Norwood Russell Hanson's Patterns of Discovery) I learned about the power of a theory to shape a person' observations and thinking. Prior to the work of Hanson, Kuhn, and Feyerabend, philosophers claimed that theory-neutral observation of facts precedes any theorizing. Furthermore, they claimed that observations always have the last word: If an observation contradicts a theory, the theory is abandoned and new theorizing beings.
But that is not the way scientists actually observe, think, and behave. There is no such thing as pure, theory-neutral observation, because the nervous system already contains theories about the world before we step foot in our first science class. Actually, our brains have a theory about the world when we are born, and different species have different theories about the world because nervous systems have evolved to favor perceptions, thoughts, and motives for dealing with each species' unique requirements for survival and reproduction. The theories in our brains of course develop as we grow and experience the world, but there was never a time when the brain was a "blank slate" that did not interpret the world. Theories exist prior to observations, and Kuhn et al. have given us examples of how theories guide and create observations in science.
When a scientist becomes very fond of a theory (and scientists do use words such as "beautiful" and "elegant" to describe theories), this emotional attachment can lead to what psychologists have called confirmation bias (Wason, 1960). Scientists, like all human beings, seem to be more likely to search for find evidence that confirms their beliefs rather than disconfirms their beliefs. History shows many interesting cases of scientists who stubbornly cling to a favorite theory despite contradictory observations, which disproves the claim from some philosophers of science that contradictory observations always lead to abandonment of a theory.
The power of theories to guide our thinking and perceiving tells us something important about the formation of hypotheses, which is a crucial part of doing science. Textbook accounts of the scientific (hypothetico-deductive) method often gloss over where hypotheses come from. These accounts often say that science begins with the observation of some aspect of the world, for example, that a majority of letters to the editor published by the NYT are written by males. Next, some hypothesis for the observation is generated. (How a scientist comes up with a hypothesis is generally not discussed.) Then, a study is designed to test the hypothesis. Data relevant to the hypothesis are gathered and then examined to see whether they support or contradict the hypothesis. The results obtained from examining/analyzing data represent something that a scientist observes, so whole new cycles of hypothesis-formation and testing can continue, keeping the scientist gainfully employed for life.
Some accounts of the "scientific method" will say that hypotheses are logically deduced from the scientist's currently-accepted theory. I think that this notion is on the right track, but that the part about "logically deduced" is incorrect. Don't take my word for it. Go around and interview some scientists. Ask them how they get their hypotheses and whether their hypotheses are logically deduced from a theory. I think, more often than not, that the scientists will tell you that hypotheses simply "pop into mind" as opposed to being deduced by symbolic logic from theoretical propositions. But they will tell you that their theories do somehow guide their thinking as they try to come up with hypotheses to test. Let's see how that might work with the observation of gender imbalance in NYT letters to the editor.
Hypotheses about the Gender Imbalance in Opinion Letter-Writing
Now, I do not want to insult Stein by calling her scientist. She is actually an artist, and a very good one. But, as personality psychologist George Kelly observed, everyone is a scientist, whether formally trained or not. All of us have theories about how the world works, especially the human part of the world that is so important to us. These theories guide the way we interact with the environment, especially our social environment. Using our theories of human nature and individual differences, we predict how other people will react when we say or do something to them and use these predictions as guides for our own behavior. Sometimes we revise our theories when our predictions fail. And sometimes we stubbornly cling to our theories when they do not work, which interferes with our ability to function in the world.
When you read Stein's article, you can see how her theories of human nature helped her to generate testable hypotheses. One of her theories, which can be found in some form in most technical theories of human motivation, is that people are most likely to engage in activities that interest them. Perhaps women are simply less interested than men in writing letters to the editor. If so, they would submit fewer letters than men, which means that more of the published letters would come from male writers. Stein tested this idea by emailing the NYT with her data and asking whether the imbalance might be due to women submitting fewer letters than men. The editors emailed her back, saying that although they did not record statistics on submission rates by gender, they were certain that males constituted "a large majority" of letter-writers. That might partially explain the imbalance in published letters.
But Stein made a more interesting, fine-grained prediction about the men's and women's interest in writing letters to the editor. She hypothesized that the topics covered in the NYT Letters to the Editor might be typically less interesting to women and more interesting to men, which would explain the greater number of male letters. But what would happen if the topic were "gendered," that is, the topic might be one that is especially interesting to women. Would we see more published letters from women on such gendered topics?
To help test this hypothesis, Stein created a chart that shows the number of published letters from male and female authors for each day of January, February, and March, 2017. In one column, she marked the days where the topic is a "gendered theme" that might be of more interest to women (e.g., "Feminism after Election," "Women's March," "Numb after Breast Reconstruction"). Stein writes:
"For research purposes, when published letters were sorted by the Times editor into topics which interested women specifically, the topics were noted in the chart to see if more women’s letters were published."
But, oddly, after setting up this chart, Stein did not actually add up the number of male and female letters to examine the sex ratio for the gendered themes. So, I counted the number of letters published on gendered themes myself and found that 35 were authored by males and 56 were authored by females. If we assume that males and females submitted equal numbers of letters to the editor, the probability of the Times publishing 56 or more letters written by females out of 91 letters is p < .02. But in reality, those odds are certainly far less than 2%. As the article says later, an editor from the Times indicated that "a large majority of our letter writers are men." The editor did not specify what "a large majority" is, but if it is 60% of all letters submitted, then the odds of seeing 56 or more letters published out of 91 by chance alone is p < .00003. This evidence supports the idea that women are more likely to have their opinion letters published when the topic is one of special interest to women. Women do not submit and have accepted as many opinion letters as men overall because they have little interest in most of the topics covered on the NYT opinion page. This all follows from the informal theory that people are more likely to engage in behaviors that interest them.
There is something shallow, however, or at least incomplete, about explaining what people do in terms of their interest or lack of interest. This level of explanation fails to explain why people are or are not interested in various activities. (This is also the problem with simplistic versions of behaviorism that "explain" behavior in terms of rewards and punishments. Simplistic behaviorism fails to explain why people find certain things rewarding or punishing.) I want to get into a deeper discussion of interests in a moment, but first I would like to consider alternative theories and hypotheses about the gender imbalance in letter-writing.
What are the alternative theories that might generate alternative hypotheses about the gender imbalance in letters to the editor? Unfortunately, alternative theories and hypotheses are not discussed explicitly in Stein's article. Nonetheless, her article contains links to other articles that do try to explain lack of female participation in activities involving self-expression. A common theory that seems to me to run through those explanations is a version of feminist theory that emphasize the power struggle between women and men and the oppression of women by men. This kind of theory would lead to the hypothesis that women's letters are not being published on the opinions page, not because women are uninterested in writing opinion letters, but because men in power are somehow preventing them from publishing on the opinions page. Even if lack of interest seems to be involved, that lack of interest in expressing opinions might be due to male oppression. Stein refers to articles that mention how, at an early age, girls are afraid to express opinions in school for fear of appearing unfeminine (where standards of femininity are arbitrarily set and controlled by men).
Evidence that Stein's thinking might be guided by a version of feminism that emphasizes oppression can be found throughout the second half of her article. She provides a link to another article she wrote on how men in the art world bully and exclude female artists. She discusses an article provided to her by the NYT editor that cites misogyny and abuse on the Internet as a force that can discourage women from participating in the comment section of articles.
Stein also quotes at length from an article by Emma Pierson:
"A world of all men is one in which Dylan Farrow’s story [that she was sexually assaulted by Woody Allen] is doubted as often as it is believed; a world of all women is one in which it is accepted almost without question. We see the consequences of this not just when women comment in the New York Times, but when a woman who has been sexually assaulted remains silent rather than reporting the crime to a male-dominated police department; when a historically male Congress lags on legislation to protect sexual assault victims on college campuses; when women in the male-dominated military are more likely to be raped by fellow soldiers than killed by enemy fire."
Stein then writes about the scarcity of women in positions of power at the NYT. She hopes that men will be willing to help correct the imbalance of power in this area, then concludes by writing, "But women can’t sit around and wait for men to give up power. We must be active, assertive, ambitious and, even (gulp) aggressive."
In Stein's article, her story shifts from the specific statistics on gender imbalance in letters to the NYT to larger issues of oppression and a need for equity. I have no problem with that because I am also against oppression and for equity. There is no doubt that some men abuse, bully, harass, and oppress women. Men also do this to other men, and some women abuse and bully others as well. None of that is right. (I also think that most of us are oppressed by the very rich, and that this problem affects more people than abuse based on gender. But that is a topic for another time.)
But is any of this relevant to the original question on gender imbalance in letters to the editor? I doubt it. The (female) editor who replied to Stein's inquiry about the imbalance writes:
"We pick letters based on merit, not gender. Typically we decide that we want to use a letter well before we reach the signature line, so we pick them gender-blind. That said, we do sometimes try to get a better gender balance if we find that we’ve picked only letters from men on a given topic. "
So, there is no obvious oppression here. In fact, the editors, after selecting letters based on merit, will occasionally include a few more letters from women that did not make the first cut, in order to improve gender balance.
The editor continues:
"I’d like to point out that the time frame that you are studying may not be representative. Politics has of course dominated our pages in the last few months, and our politics letter writers skew even more disproportionately male than the norm. We find that certain topics, like education and health, tend to draw a larger response from women."
So, perhaps there are more letters from men on politics and more letters from women on education and health because of men and women have different interests? We are back to interest-theory again. (Of course, oppression theorists might argue that women were forced to be interested in the stereotypically-female topics of health and education by the powerful men who control society. These theorists have already said that the patriarchy forces women to be interested in low-paying, female dominated fields such as nursing, human services, and teaching, and discourages interest in high-paying fields such as science and engineering. Again, an interesting topic, but one I will leave for a different time.)
My Alternative Hypothesis on the Basis of the Gender Imbalance in Opinion Letter-Writing
At this point, I want to share my own theoretical preference and how it affected my hypothesizing about the opinion letter writing. My theory is not such much an alternative to "interest theory" as it is a particular version of interest theory.
I am an evolutionary personality psychologist. An assumption common to most evolutionary theorizing is that the things we are interested in doing often increase our ability to survive and reproduce. Both men and women must be interested in food and sex to some degree, because those two things are required for anyone's survival and reproduction. But when men's and women's interests differ, leading them to behave differently, this could be because men and women need to do different kinds of things to pass on their genes. This is a simple idea, but not simple-minded. This perspective allows us to perceive sex differences in terms that would never have occurred to a non-evolutionist in a million years. That is what makes evolutionary theory powerful.
As an evolutionary psychologist, my first inclination when I read about the imbalance in opinion letters was to think about what opinion-expression means in the struggle to get one's genes into the next generation. What occurred to me is that opinion letter-writers are often trying to demonstrate how smart, creative, and/or moral they are. If a person writes a letter that is cleverer than the article he or she is responding to (and cleverer than the other letters), the person is successfully advertising the high quality of his or her genes to the world. The Letters to the Editor section can therefore be seen as an arena where participants battle each other to establish their dominance and attract mates. In most human societies, men more often battle each other in order to be chosen as a woman's mate than the other way around. From this perspective, the male-female imbalance in letters to the editor is not surprising.
Of course, this is just a hypothesis, and I am more interested here in demonstrating how different theoretical orientations generate different kinds of hypotheses rather than trying to convince you that this particular hypothesis is correct. The hypothesis would have to be tested, of course. One could set up an experimental study in which participants had an opportunity to engage in writing opinion pieces, manipulating variables such as whether potential romantic partners would be reading the opinion piece. I will leave the design of such a study to anyone who is interested in conducting it.
I want to end with some thoughts about the importance of a good theory. There are good theories and bad theories. Good theories lead to hypotheses that are likely to be found to be supported by evidence when tested. Bad theories either fail to lead to testable hypotheses, or lead to hypotheses that are not supported when tested. It takes time and resources to test hypotheses, so following a bad theory will lead to wasted time and money. Evolutionary theory now has a great track record for generating hypotheses supported by data, which is why I encourage people to think in evolutionary terms. And here's the real kicker: If you want to solve a social problem such as gender inequity, you need to know the real causes that underlie the problem. Not alleged causes from an ideological position, but the real causes, revealed by research. I think we have a much better chance of finding the real causes of gender inequity and correcting them by conducting research guided from an evolutionary perspective than by any other perspective.
Feist, G. J. (2011). Psychology of science as a new subdiscipline in psychology. Current Directions in Psychological Science, 20, 330-334. DOI: 10.1177/0963721411418471
Johnson, J. A., Germer, C. K., Efran, J. S., & Overton, W. F. (1988). Personality as the basis for theoretical predilections. Journal of Personality and Social Psychology, 55, 824-835. doi:10.1037//0022-35188.8.131.524
Singer, B. F. (1971). Toward a psychology of science. American Psychologist, 26, 1010-1015. DOI: 10.1037/h0032255
Wason, Peter C. (1960), On the failure to eliminate hypotheses in a conceptual task, Quarterly Journal of Experimental Psychology, 12, 129–40. doi:10.1080/17470216008416717