Statistics on readership and posting habits for academic blogs

Question

Although this stackexchange seems to be a little hostile towards metrics (especially when they are about research productivity), it is still sometimes fun to indulge in a little bit of arbitrary measurement and quantification. Sometimes it can help you set targets, or let you know what is possible. In this case I am curious about blogs.

Having a web-presence is important, but how do you know if your academic blog is doing a good job?

From my own experience, I have noticed that my blog gets a lot more readership and mention than any of my papers. I usually find this encouraging, and at times it helps me increase productivity by incorporating blogging into my research work-flow and feeling like I am able to communicate with people before having complete results. Sometimes even receive feedback (although my blog is not at the level of regular commentators, and nowhere close to the comment activity I see on popular blogs that I follow).

However, getting more mention than my papers is not a fair standard. In fact, I have no standard by which to decide if I am doing an alright job blogging, and what I should aim for to improve the ability of my blog to engage other researchers or interested readers. Having some hard data is also useful for converting people new to blogging to the online community.

Are there any statistics on typical readership, posting rates, and commenting frequency for small (non-superstar) academic blogs? I would be especially interested in statistics that are broken down by area, since I expect a nutrition or cancer blog to inherently get more readership than one dedicated to Stone-duality. Of particular interest to me would be information about blogs in theoretical computer science and/or mathematical modeling.

Answer 1

I think it's interesting to consider the relative value to society of blog posts relative to more traditional forms of content distribution, such as book chapters, text books, journal articles and so forth.

Obtaining Benchmark statistics

• RSS counts: Many blogs, particularly popular ones, show their RSS subscriber count. You can use the Explore Search feature in Google Reader to search for blogs you know. This returns the number of Google Reader subscribers. This is less than the total reader count, but it can give you a rough ball park.
• Page views: Some blogs occasionally post their site statistics. Alexa can provide a very rough estimate of the popularity of a site.
• Comments: It is straight forward to look at other blogs to get a sense of how many comments they typically get.

My rough rules of thumb

I've been blogging since 2008 and have kept an eye on RSS feeds and page views over timeon my own blog. I've also picked up information from other blogs that I follow. My main observations are that it takes time to produce content, get indexed by Google, obtain RSS subscribers and so on. These would be my rough benchmarks for academic blogging. In the fields that interest me (e.g., psychology, statistics, R) I can think of specific blogs that fall in to one or other of these categories. This helps to inform the benchmark. Anyway, these are just my casual rules of thumb; of course, they aren't anything definitive.

• RSS subscribers:

  • 0 to 10: Not popular
  • 10 to 100: Just getting started
  • 100 to 500: Moderate levels of popularity
  • 500 to 1000: Relatively popular
  • 1000 to 10,000: Popular Blog
  • 10,000+: Superstar blog

• Annual Page Views

  • 0 to 1,000: Not popular
  • 1,000 to 10,000: Just getting started
  • 10,000 to 50,000: Moderate levels of popularity
  • 50,000 to 300,000: Relatively popular
  • 300,000 to 2,000,000: Popular Blog
  • 2,000,000+: Superstar blog

What does a page view mean

It is a little difficult to know what a page view means in terms of achieving broader blogging goals. Only a proportion of page views correspond to a person reading the entirety of the page. And only a proportion of those page views have any meaningful impact on the reader. In order to get a sense of what these proportions might be, I reflect on my own browsing. For example, I might be searching to diagnose a software error, do a tutorial on something, or get a review of a product. It might take a few search results to find what I'm looking for. That said, perhaps something between 1 in 10 and 1 in 2 search results provide useful results.

In summary, even if only 1 in 20 pageviews helped someone in some meaningful way, if you're getting a hundred thousand page views per year, that's still 5,000 instances of people being helped.

Answer 2

I used to track my stats compulsively, but no longer do so. this is mainly because I get lots of readers through an RSS feed, which doesn't directly impact traffic. Sometimes I'll monitor the relative hit rate of specific posts, and I have seen dramatic jumps (for example if I do business meeting blogging, or if I post on something controversial).

As a rule of thumb, the more technical the post, the less traffic it gets. The more buzzwordy, the more traffic. I had some thoughts on deep learning recently and that got huge traffic in comparison to some of my more technical posts.

Now, because of G+, twitter, and blogs, my "visibility" is diluted across all three media, and while I'm sure there's some way to monitor all of them, I haven't paid that much attention. Ultimately, I blog because it's fun, and the more I get distracted by audience response, the more I find myself distorting the things I post about.

Answer 3

I can't say much about statistics, as I have not come across any, but in my experience blog readerships are usually small and most likely by specialists in your area. However, I view a blog as free advertising for my research, as blogs are generally more highly ranked than academic papers by google and the like. I also view a blog as an ideal platform to put your research into laymans terms. From my blogs I have had an out of the blue invited talk, and also requests from numerous researchers who I hadn't met for copies of papers or general queries about my work.

I personally find maintaining a nice website and blog well worth the effort, and I try and spend a couple of hours a week on new content, but usually concentrated when new articles are published. Ultimately it's all about raising the profile of your research, and having more accessible material is always helpful, even for specialists.

Answer 4

I don't know if it is the answer you are looking for, but I would be cautious with looking at blog views (or even likes/tweets):

• First, they may be superficial. You don't even know if someone actually read it (maybe (s)he entered just because of a sexy title, or a nice picture, or - misleading keywords).
• Second, they is are measure of popularity, not necessary quality, with a lot of mechanisms making scaling exponential (e.g. snowball effect).

Personally, I often look at stats of page visits of my various sites... but I cannot make much sense of it. But what I find important is:

• How often I can send someone a link to my post, so it save my time of explaining something once again?
• How much I learn something from readers, or make new contacts through it?
• Do I hearing feedback, especially from strangers or people I don't know very well?

Moreover, then I can compare blogs to regular articles on this ground. Still it's apples to oranges... but now they are quantized fruits. :)