question

[Quentin L. Burrell]

Steven Morris raises an important point regarding any regression/correlation study - look at the data first!

  Another thing that would be interesting to see would be a scatter plot of the 467 articles with number of references on the x axis, and number of citations received on the y axis along with the line of fit.  Such a visual presentation really helps to assess the 'goodness' of the correlation that is being established, and usually shows if outliers are adversely affecting the analysis. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.asis.org/pipermail/sigmetrics/attachments/20070130/847491c1/attachment.html>