I’ve recently published some stats about which Twitter users are listed in how many Project Management – related Lists on Twitter
There have been a few comments and questions about what this data means, and why it’s worth looking at.
So I’ve done a little basic correlation between the number of twitter followers that each user has and the number of lists in which the person is listed.
Although the data is noisy, there’s a reasonable correlation between the two, as you can see below.
See the notes after the chart for a bit more information on the chart.
List Membership an indicator of influence?
- bottom axis shows groups of list memberships of 30. i.e. I’ve grouped all users who are in 0-29 lists in one bin, 30-39 in the next and so on
- the bottom axis is inverse sorted, so the bigger numbers are closer to origin
- the LH axis is the average of all the user’s followers in that bin of 30.
- Tableau provides the following description of the Trend Line model.
Trend Lines Model
A linear trend model is computed for average of Followers given In # Lists (bin).
|Model formula:||( In # Lists (bin) + intercept )|
|Number of modeled observations:||10|
|Number of filtered observations:||1|
|Model degrees of freedom:||2|
|Residual degrees of freedom (DF):||8|
|SSE (sum squared error):||6.21867e+09|
|MSE (mean squared error):||7.77334e+08|
Individual trend lines:
|Followers||In # Lists (bin)||0.201747||8||In # Lists (bin)||3875.73||2786.68||1.39081||0.201747|