That seems like kind of a weird thing for a person to say, doesn’t it? I mean, now that I think about it. But it’s true. I dig stats.
I’ve mentioned before that one of the main reasons I chose Word Press for my blog site is because of all the statistics that come with the package. I can keep track of which posts are the most popular and so can know what the readers like and what they could do without (more or less).
So today I was wondering which type of ATM post is the most popular to date. To find out, we need to look at the data and then crunch some numbers.
First, we need to know the number of hits for each of post. Okay, here you go:
Okay, next, I removed from the analysis the posts listed in red because they’re not really posts per se but are menu buttons in the banner, just below the header image of the blog.
Then, I put the posts into 4 main categories: Narratives, News, Science/Biology, and Stuff About This Blog. The one originally categorized as “Rebuttal” (Getting Some Facts Straight About Matt Rice and Thomas Beatie), plus the most recent one that was originally categorized as “Commentary” were both put into the “Narratives” category for the purposes of these analyses.
|Category||Total Hits||No. of Posts|
|Stuff About This Blog||768||10|
Now, if we divide the number of posts into the total hits per category, we get the average number of hits per category:
But wait — we still have to take into account the amount of time each post has been available for readers to see. A post that has been published for, say, four months could have more hits than one that’s only been published for four days simply because it’s been available for a longer period of time rather than because it’s more popular.
Well then, what if we run a correlation between the number of hits an individual post has received versus the number of days it has been published? If the number of hits is directly correlated with the number of days published, we’d see something that looks like this example of the correlation between shoe size and foot length:
Well, that doesn’t look like a direct correlation. The number of hits for each post is not necessarily dependent upon the number of days each post has been up on the blog.
I wonder what the correlation would look like if we split the posts into the four categories we used above. Let’s check it out:
And there you have it folks! From this graph, we can see that the length of time that a post is up on the ATM blog has little or no effect on the number of hits it gets when it’s in the categories of Narratives, News and Stuff About This Blog. It’s as though those posts receive a certain number of hits and then no one looks at them again.
On the other hand, the Science/Biology posts just keep accumulating hits over time, so the rough correlation would be that the longer the scientific posts are up, the more times people look at them over time. It’s not a perfect correlation, but the trend is there.
Perhaps this means that I should concentrate on the science stuff and leave out the news completely. The narrative posts and the information about the blog look to be okay to toss in once in a while, but it seems that you, the readers, want SCIENCE!
Can’t say that I blame you. Science is good stuff! Well, okay — next post, science it is!