There are several very basic misunderstandings of how things work when it come to engendering and encouraging diversity, and I’d like to make a few comments with the aim of clearing them up, at least partially. One example of a misunderstanding came up a while back when some of us were complaining about the number of Y-chromosomes represented in Richard Dawkins’ otherwise excellent science anthology, and I have been reminded of it more recently by the inexplicable blogarrhea coming from the general direction of the former John Loftus, who can’t stop complaining about (… oh never mind, it does not matter). In the end, it is all about how we make selections, which are samples of a larger population, and we make selections quite often.
1) All selection efforts are making a small aquarium from a large sea of fish.
This guideline was well underscored during the discussion of Dawkins’ book. Dawkins put together a volume of science writing that included almost no female representation. A similar book covering a somewhat different time period that I randomly pulled off my shelf had twice or three times as many women represented, and that volume was put together years earlier, during what may have been a less enlightened time. There were relatively few science papers by women that could have been used from the old days, because few women were allowed in science and those that did exist were either discouraged or disallowed from publishing. In order to advance our collective goals of supporting women in STEM, Dawkins’ anthology should have purposefully contained more women than an even sample would have represented, though of course it could not have contained a lot more. But, the anthology actually contained fewer. It seems as though there was never a moment’s consideration given to the question of gender diversity when compiling that volume.
During the commenting and conversation about this, several individuals (who did not understand what the anthology was or how it was constructed) made rather annoying statements such as “It was the top 100 writers … if you added a women in there that didn’t belong, then someone who deserved to be on that list would be bumped off,” or words to that effect. (In fact, there was not an arbitrary cut-off of 100 authors.)
The truth is that when putting together lists, anthologies, awards, or other similar things there is no such thing as an absolute scale. There is a great deal of subjectivity as to where an individual ranks in almost any comprehensive or achievement-based ranking scheme. Also, there is a much larger universe to draw from than people often realize. For any selection among a large enough category of individuals, there will be people you never heard of, didn’t consider while making your selection, and who were in the end ignored the entire time but who should not have been. Assuming this is true, and it is, then how can one assume that a selection of a hundred, or thirty, or a dozen, or whatever, of people to represent the “best of” something is an absolute immutable perfect subset? You can’t. The criteria are not as solid as one may think, and the population is not as perfectly sampled as one might think.
Therefore, an extra effort to locate and include underrepresented groups does not violate some golden rule of quality and quantity.
2) Underrepresented groups get under-represented even more during selection processes.
This is very important and in some cases it may be the main problem with enhancing diversity. I recently ran into this when I tried to put together a list of “diversity enhancing” bloggers to supply to a blog network to help guide future selections of bloggers to invite. Before you assume that I was making this list for the blog network you are now reading, I should mention that I’ve made about six of these lists for Scienceblogs.com and the list to which I refer may well have been one of those.
Let’s say you want to increase ethnic diversity on a list of potential bloggers. The best way to do this, and this is not my idea but, rather, how a lot of extant hiring and other selection processes already work, is to make a more diverse “mid list” (or long list) from which individuals will be selected based on their abilities, fit, availability, and so on. But if you look at bloggers in a given area, you might find that “ethnic” bloggers are more often part of existing networks, and therefore may seem less likely to be available to join a different network. In other words, if your goal is to recruit “indies” … individuals who are blogging at WordPress or a University site or some other platform, into a network like Scienceblogs.com, you might miss those bloggers who are already “not indie” because they are in a network. Now, suppose that being “ethnic” in a certain milieu tends to result in being part of a network. this can happen for a lot of reasons, including the fact that you are already part of a pertinents group that then goes and starts blogging, or perhaps it is a matter of security and comfort in an unfriendly context. There is one “black” blogging group that is about to start up as a bloggerific version of its non-bloggy self. Groups like Skepchick and Feminista may be this phenomenon manifest in the context of gender.
You see the point … a selection process seeking independent bloggers will systematically miss underrepresented or repressed groups because individuals in those groups self-select to be not indie as often as straight white guys do, and the reason you are looking for more diversity is the same as the reason that it is harder to find.
This phenomenon is a sub type of a larger set of phenomena that tend to make individuals we seek to promote diversity harder to find.
Another example, having to do with gender, is simply lifespan. Have you ever heard of a blogger starting out blogging, and then running into frequent harassment with the occasional scary event (like a credible or very obnoxious threat) happening now and then? It happens all the time. Have you ever heard of a blogger either turning off to blogging for longish periods of time because of this, or simply quitting? That happens now and then. Have you noticed that when this happens it is very often a female blogger? It is true it often is. I am engaged in a at least a dozen ongoing conversations with women who will blog, are blogging, or did blog where this is an issue. I’m not having this conversation with any guys.
If you want to develop a selection of individuals in any milieu including the blogosphere, to invite to a network, to give an award to, to put on your blogroll, and you want it to be diverse, you have to take this into account. Which brings us to the next item …
3) If you want diversity (and you do) you have to make it happen.
You can’t just say “I’m not biased” or “I’m not sexist” or “I’m not racist” and then expect the world to adjust to this noble sentiment and sort everything out for you. If you are in charge of making a list of people, to consider to give talks at a conference, to be on a panel, to invite to a radio show, to be in a blog network, to have their work chosen as part of an anthology or show, or for that matter to be your acquaintance or friend, even a facebook friend or a plussy, if you don’t pay attention to the composition of your list, the list will not manage attention to diversity for you.
You have to do it on purpose. You have to make it happen.
4) In this day and age, a selection process that does not address these issues IS a sexist or racist (or both) process, like it or not.
In the blogosphere, the world of atheism, and the world of skepticism, we are a little behind in these areas. Try sitting on a hiring committee of a modern University or College, or get a job where you are involved in graduate or undergraduate admissions, and you’ll learn that attention to important details such as diversity and balance is not only old hat but very much ingrained in the system. I find it astonishing when I run into people busily hiring, selecting, choosing people, and when the issue of diversity comes up they act like it has never been an issue before, or otherwise give signals indicating that this is something that was never part of the process for them before. There are large parts of this world where consideration of diversity and mechanisms for enhancing diversity are very much commonplace.
There are even systems which, for many years now, have required gender parity. The process of selecting delegates for political representation, that attend conventions and pick candidates, is often a gender-balanced one. At the very first stage of selecting delegates for the Minnesota DFL party, one male and one female is put forth for each of the smallest political units, and as the delegates continue from one step to the next in the primary, caucus, and convention process, fifty-fifty parity is maintained the entire time.
When you are involved in a selection process, in a decision making role, and you don’t account for diversity, then you are being a bad person! Or at least, if you are operating in a milieu where this has become normal, you’re defiantly doing it wrong. It is no longer acceptable in the skeptical and atheist community to put together lists of speakers for conferences that include only white males. As recently as a year or two ago, it apparently was, which is rather shocking. Despite living in a backwards and unenlightened community (skeptics and atheist!), the truth is that the importance of diversity has been known since before you were born, if you are of an average age. Approaches to developing diversity have existed and been widely used for long, long before it ever occurred to the organizers of many well known skeptics and atheists conferences to use those approaches. And I have to admit that I’m a little annoyed at About.com right now. Austin Cline is over at About.com, and has a reader’s choice award for “Favorite Agnostic or Atheist Blog“, “Favorite Agnostic or Atheist Book” and “Favorite Agnostic/theist to follow on twitter.” These involve reader nominations, some sort of narrowing down based on votes, and then a short list from which readers can chose by clicking on a web form. If you look at the lists at this stage, they tend to be very white and pretty much male in the first two categories (though interestingly very female in the hipper but less well established, twitter category). In other words, they tend to lack diversity.
Yes, yes, yes I know, I know what some people are going to say … this is what the readers picked, therefore it represents what is “out there” and that is just the way it is. If you want diversity, Laden, get it some other way because this is a simple direct sampling of who is out there in the world to be chosen from.
That sentiment, quaint as it is, is wrong for several reasons. First, we want to increase above background levels the representation of underrepresented groups, even just a little, for a number of reasons that are so blindingly obvious I won’t extend this already too long post to enumerate them (if you really don’t know, ask in the comments). Second, see above; the selection process is likely to underrepresented underrepresented groups more than they are already underrepresented, and I strongly suspect this happens during the public voting, nomination, and polling methods that today plague the Internet.
Then there is the argument that this is just the way it is because only a proper voting process can be used or freedom of speech, due process, and several other constitutional amendments will be violated and our basic freedoms will vaporize and civilization will collapse. But there is a problem with that argument. Two problems, really.
First, the system is broke with respect to diversity enhancement, so please stop using it. There is no reason in the world that any particular process of nomination, voting, or polling MUST BE USED or the world ends. If we recognize that on one hand we want increased diversity and at the same time we want some sort of community based consensus, then surely we can figure out how to get there using technology and conversation and ideas and stuff! I mean, really, this is the 21st century. We can solve this problem. The second problem with this excuse is that the system is broke anyway. On-line voting systems are all broken and stupid. To be honest, I’m not sure of all of the technical details of how Austin did this, so maybe he’s got a less broke system. But I know this kind of system is broken in general when the science category of the once venerable “webbies” is populated almost entirely with science denialist sites. There are certain things for which crowd sourcing is not the way to go.
I don’t mean to pick on Austin Cline or About. I think they do a great job at getting the word out, and these problems with diversity (or lack thereof) are widespread within our own communities. Its just that this particular set of selections is out now, current, and a friend of mine (who will also be blogging about this) has pointed them out.
Can we start thinking of better, creative, and more productive ways to carry out these selection processes that we all seem to be so interested in?