People do too like to read science

I don't worry too much about the subjects I choose to write about here. I write mostly about science because it follows my workflow, and honestly because people seem to like it.

But various elements among the "science blogging" crowd are wringing their hands about the sad state of affairs among "top-rated" science blogs. It started at Bayblab, where the complaint was that many so-called science blogs write little about science. In the comments (and elsewhere, such as Sandwalk, Uncertain Principles, and Blog Around the Clock), actual science bloggers raised an empirical point: they get more readers and more comments for their non-science content. Judging by results, they're better off writing about creationists, religion, politics, their favorite fiction, or the ever-present lolcats.

It seems a little silly, but I want to correct that misconception. Blogging about actual science is not a turn-off for readers. If it were, people like me or Carl Zimmer, or Cognitive Daily wouldn't have many readers. In fact, I get vastly more traffic for science-related posts than for anything else.

Now, it is fair to say that there is an ascertainment bias at work: I really don't do cat posts, or Dawkins-worship, I never write about politics, and I keep creationist-bashing to a minimum. I used to do a share of white supremacist-bashing, but I can't say that it really drove much traffic. No, I pretty much do science, and, well, science. In fact, my front page right now has nothing except science -- except this post, of course.

Sure, I have the occasional post questioning panda conservation tactics, and some quotes and whatnot. But those things hardly get any traffic on their own; they mainly fill in the cracks between the long posts on the front page. My experience is that people appreciate clear writing about science and will find blogs that do it consistently.

Here are my top five individual posts for February:

1. Last year's post, "The appendix: not just for appendectomies anymore?"
2. The mutation-child-development post, "Hunting for your child's DNA doppelganger"

3. The Icelandic gene-pedigree study, "Third and fourth cousin marriages more fertile"

4. Genetics and English history, in "Viking ancestry, surnames, and medieval genetics"

5. The review of last week's HGDP SNP surveys, "Serial founder effects, again"

I say "individual posts" because the vast majority of readers go straight to the main blog page. I tend not to hide the content of posts "under the fold," and regular readers don't generally need to view pages other than the main. So when I see a post show up in my log, I can be pretty sure that most of those page views come from outside referrers. That gives me a rough picture of who is finding my site from posts linked elsewhere. Likewise, Technorati gives a more-or-less direct picture of who is linking to what posts -- as long as the referring site is registered.

A second post from Bayblab suggests that bloggers write mostly for other bloggers, making the medium an inbred conversation. But I have to counter based on my statistics that the audience depends on the blogger:

1. The vast majority of traffic that finds pages on my blog from links at another site, comes from outside the usual "science blogs" sources. Yet, almost all of my inbound links are to science content. More than half my inbound visitors are finding the blog from Google or Yahoo because they are searching for some topic I've covered. A large fraction now comes from news stories, for which editors or authors have graciously provided links to more specialist content. I get significant inbound traffic from newsgroups, as well as bloggers outside of science.

2. None of the top 30 referrers to my blog are from ScienceBlogs. Developing a following among other blogs is great for links, but does not particularly drive many readers. I link outbound to blogs in probably half my posts, but I wouldn't characterize this as a conversation; it is generally more like a toast after a good meal. Two of my top 30 inbound referrers are science blogs (GNXP and Dienekes).

3. A dedicated following of readers is an awe-inspiring, humbling thing. My readership includes some of my scientific colleagues, and I think it's wonderful. I also have a number of other bloggers who read and are kind enough to link into my posts. But if I were primarily aiming for these groups, I would be happy with 200 readers. I have vastly more readers from outside the field, who are interested in anthropology, and who are looking for more detail than the usual news stories about the field. It seems to me that this is the core audience for a science blog, and if a blog doesn't have much success with science content, it is because it is missing this audience.

4. I don't have comments. Bloggers may like getting a lot of comments, because it gives the appearance of a conversation. But it doesn't take more than a glance at comment sections like the one at Tierney Lab to see that getting lots of comments is not an indicator of the value of a post. If people are commenting heavily on a post, it usually means that they think they're at least as qualified as the blogger to vent about the topic.

Personally I love to hear from readers -- and if you write to me with an idea, there's a decent chance I'll follow up with a post. But I don't think most of you are looking for a free-for-all with any anonymous schmo who wanders in off the internets. It doesn't have to be that way, though -- for example, the comments section at Cognitive Daily is superlative. As I continue to update the cogs and wheels that run this site, I may experiment a bit with commenting to see how it might go.