After I posted my thoughts on big book causality, I received this question in my ask box:
The Short Answer
My short answer to the question of whether or not I find it disingenuous to suggest that being male might have had some effect on John’s success is… no. Not even a little bit. Not at all.
The Long Answer
My long answer is (as one might suspect from my the length of my last blog entry)…. long.
I want to start with the scientific method and the concept of the null hypothesis. In any experiment, there is a null hypothesis: the hypothesis that says that the effect you’re looking for or investigating does not exist. So for the “experiment” you’ve just run, looking at a data set of the top 100 bestselling books in 2012, the null hypothesis is that there is no effect of author gender. In order to reject the null hypothesis, the data would need to show that the distribution of male and female authors varies in a statistically significant way from chance. In this case, based on the numbers you cite, I am going to assume that it does not.[i]
This is called a null result. What does it mean when you get a null result?
It means that you don’t reject the null hypothesis.
It does not mean that you accept the null hypothesis, or that the null hypothesis has been in any way proven or confirmed.
The reason for this is that you can get null results for a huge variety of reasons. In a laboratory setting, you can make pretty much any effect go away with a poorly designed study. A null result might mean that the effect you’re looking for doesn’t exist, but it also might mean that there’s something wrong with your logic or design, that you’re not getting at your big question the right way, that your measures aren’t sensitive enough or that they don’t measure what you think they do, that your sample size isn’t big enough…or any hundreds of things.
Null results are notoriously hard to interpret, and there is probably not a social scientist on the planet who would agree that saying “well, I analyzed a single data set, and the results don’t appear to be significant, so no need for further exploration! THE CASE IS CLOSED!”
Science doesn’t work like that—and neither should your day-to-day reasoning. “I did not see evidence of privilege in this one thing I looked at” is a far cry from “there is definitely no privilege at play in this entire domain!”
What stands out to me the most about your question is that you seem to believe that the evidence that you present is sufficient to robustly conclude not only that there is no effect of male privilege within the literary world (especially the young adult literary world), but also that anyone who so much as suggests there might be is being “remarkably disingenuous” and “silly.”
This is particularly striking to me, because as you pointed out in the question, “male privilege exists in many areas of life.” If you come into this knowing that male privilege—and white privilege and so many other kinds of privilege—do exist, in many different arenas, why does it take so little evidence to convince you that not only is it unlikely that male privilege exists in this domain, it is remarkably unlikely?
The existence of male privilege out there in the world should be enough to make a person look at a null result and think “this merits further investigation” rather than “case closed,” especially given that you can’t conclusively interpret an isolated null result anyway.
For the sake of the rest of my response, I’m going to translate the question from “aren’t you being disingenuous and silly, forcing something that isn’t there?” to “I’m curious, given the composition of this end of year bestseller lists and the existence of female writers of mega-hits, like J.K. Rowling, Stephenie Meyer, and Suzanne Collins, why would you think male privilege plays a role in the reception of literature and especially young adult literature?”
I could give a variety of answers to this question, but today, I am going to focus on one: reception by literary critics and review outlets, a factor John himself credits for contributing to his success.[ii]
First, I’ll give a brief overview of the question and its importance. Then, I will outline some theoretical reasons to think that male privilege might exist in this arena and that it might be particularly strong for stories that, like The Fault in Our Stars, are set in our real world, rather than a fantastical or futuristic world. And finally, I’ll examine some actual data on the topic.
Reception among adult literary critics and review outlets
The questions “could being written by a male author have affected the reception and treatment The Fault in Our Stars received?” and “do some books by women become huge bestsellers?” are not actually the same question, and it’s a mistake to treat them like they are. One of the many food-for-thought questions I posed in my last post regarded the interaction between gender and genre, in terms of success. The idea behind this question is that there may be particular genres in which being male is more of an advantage than others—for example, in books considered to be “literary,” or in realistic contemporary YA. There are a variety of theoretical reasons to think this might be true.
For example, there’s an idea in our culture that male stories are somehow more universal than female stories—we’re a culture that expects young girls to be willing to read stories about boys, but actively discourages boys from reading stories about girls. We’re also a society in which prominent female bloggers in the real world are frequently met with rape and death threats and told to shut up. Empirical studies have shown that male students perceive females as “dominating” class discussion even if they talk only 50% of the time.
In short, there is reason to believe that when women talk, they’re met with more resistance than men are; they’re seen as overstepping their bounds, even when they’re not talking any more than men are; and their experiences and stories are seen as holding less relevance for the opposite sex.
What does all of this have to do with predicting that literary acclaim might favor books written by men, particularly for books set in the real world—books that deal with things like break-ups and falling in love and infatuation and loss? Philosophers have suggested that worlds further removed from ours in fictional space—those that feature more fantastical elements, for instance—may trigger less “imaginative resistance” to counter-cultural norms. So there is a theoretical reason to predict that the “guy stories are for everyone; girl stories are for girls” rationale—as applies to both author gender and character gender—may apply somewhat more to stories set in our world than to stories set in a fantastical or dystopian world.
Here’s an anecdotal example illustrating how this overvaluing of contemporary male stories and undervaluing of contemporary female stories can play out in terms of character gender: I’ve seen it suggested that the reason that “An Abundance of Katherines” might have been received differently (e.g. coverage in the New York Times Book Review, Printz Honor) than a female-POV “An Abundance of Kevins” would have been is that obsessing over someone you’ve broken up with is such a female thing to do, that it’s more interesting and somehow a more worthy story when a teenage boy does it than when a teenage girl does it.
But is there any actual evidence to suggest that teenage boys take being dumped by someone they love any better than teenage girls do, that obsessing over a break-up is actually a “girl” thing to do? A google search for “why did my boyfriend break up with me?” (quotation marks included) yields 21,700 results. A search for “why did my girlfriend break up with me?” yields 118,000. If anything, it seems as if guys might obsess (on the internet) about their lost loves more. It’s also not the case that “guy gets depressed because his girlfriend dumped him” is somehow a unique or underplayed narrative in our culture. The Social Network, 500 Days of Summer, and Silver Linings Playbook all feature guys who don’t deal well with break-ups, and that’s just off the top of my head.
A story about a guy obsessing over an ex is not, a priori, a more noteworthy story than the story of a girl doing the same. But people treat it like it is. You sometimes see the same line of reasoning used in the adult literary world with respect to author gender—books written by men that deal with emotion tend to be seen as somehow deeper, because everyone knows that women are emotional creatures anyway, so it’s the male emotions that are really interesting.
You can see how this might put contemporary books by females—particularly those that, as The Fault in Our Stars does, feature romance and feelings—at something of a disadvantage compared to those written by males, in terms of the way they are evaluated for literary merit.
In the next section, we’ll look to see if there is any evidence to support the idea that the literary world may be biased in this way.
One of the contributing factors to his success that John cites in his original post on the success of TFIOS is the fact that “someone got a bunch of adult literary critics to read a YA novel that those adult literary critics really liked.” He actually bolded that phrase in the blog entry, because it was an important point to his analysis in the success of TFIOS!
So what do we know about getting adult literary critics to read stuff and feature it in prominent outlets?
Well, there is a wealth of data suggesting that adult literary criticism and review outlets are skewed in favor of male authors. You can see the round-up of 2012 review percentages here, but for a bigger sample size, you can also look at the numbers in 2011 and 2010. Or you could look, for instance, at this round-up of 2009’s “best books of the year” from various outlets.
This isn’t just one analysis from one list. This is looking at multiple review outlets, multiple best of lists, across multiple years, and there is a fairly unmistakable trend here: books by male authors are given more attention in these outlets than books by female authors.
So let’s return to the causality chart from my previous post (which, remember, is far less complicated than what big book success probably looks like). I’ve added in just this one consideration of author gender.
The dashed line here is meant to emphasize that gender is exerting some weight on press coverage. It does not have to affect all types of press coverage to be incorporated in the model; if it exerts a statistically significant effect on a subset of coverage that John identifies as playing a key role in his success (attention from adult literary critics), that alone is worth inclusion in the model.
Time Magazine Best Books
Just in case you aren’t yet convinced, let’s take a look at a specific major milestone for TFIOS that John mentioned in his exploration of the book’s success: being named TIME MAGAZINE’S BEST FICTION BOOK OF THE YEAR, the first YA book ever to hold the title.
Could there be an effect of author gender on the likelihood of being named to—and topping—this list?
I found a list of the books named to Times Best Books of the Year lists here.
This list covered the books recognized on Times’ Best Book lists from 1996-2010. I then went in manually added in the 2011 and 2012 titles, and I went through and classified each title by author gender (using Google to help me make the correct determination for unisex or ambiguous names). Then I ran some pre-planned statistical tests on the resulting numbers.
Disclaimer: all of this was done around 2 in the morning and is subject to some margin of human error. If you run the statistics for 7 out of 10, the result is not significant—meaning that 7 out of 10 does not significantly vary from 5 out of 10. So even though the cut-off point of the top 10 appears to favor women, it does not do so in meaningful way.
I’m using TFIOS as an example here, but as I discussed in my last blog post, this isn’t really about John or TFIOS. It’s part of a wider discussion, made easier to have by the fact that John has provided us with a thoughtful analysis of his book’s success to use as a reference point for factors that might be worth investigating.
This may be particularly true for YA, given that certain adult genres, such as science fiction, have a history of being more unwelcoming to women and women writers that could be an additional confounding variable here.
I ended up with 249 titles total, which tells me that somewhere along the way, a single title was lost; however, as you will see once I show you the numbers, one book would not affect the statistical significance of these results. I’ve included the source of my data and details about the method so that others can check the math if they wish to do so.
You know, it’s one thing to know a thing (ie, gender matters to the way one is perceived as an author) and it’s another to watch a scientist break it down. Jen, this is amazing stuff.