I was ready to do a bit more analysis of whether or not my schedule is helping me write more. This time I decided to get real and use a formula to figure out if the change I noted previously was actually statistically significant. It’s been so long since I’ve had a statistics class that I had to turn to the web for answers. One web search later, I found THIS, which I adapted quite easily to work with my daily log of word counts.
I had noted the start date for the schedule in my daily log, so it was easy to make the formulas work with my own spreadsheet. Here’s what I came up with.
|658||average value of data before change (daily word counts)|
|741||average value of data since change (daily word counts)|
|952||standard deviation of data since change|
|133||number of data points since change (how many days I’ve been on the schedule)|
|15.73%||probability that the change is only due to chance|
So, I wouldn’t say the change in word count is significant. After a bit of further reading on Wikipedia, the problem I’m seeing is that I have no idea what to set as the significance level. I can’t deny that it’s quite possible the increase in my word count was/is temporary and could have other causes. The truth is that my word counts often improve after I make a change. And honestly, if I look at a rolling 7 day average, I can’t see any patterns at all.
Other than the pattern that sometimes I write a lot and sometimes I don’t write anything! Consistency isn’t something I’m good at.
I did have several more 3,000 word days recently than usual (I’ve never had three of them so close together before), but on the other hand, it had to happen sometime, right? I now have 1,112 days worth of entries and I’m adding a new one every day.
All I’ve really done is give myself something else to think about, unfortunately, while I try to pull myself out of this funk and get on with the writing.