Talk:Linear correlation

From Wikiversity

[edit] Venn diagram

Suggest removal of the venn diagram, as it has nothing to do with the topic. (Or, if it has some relevance that I am not aware of, explain that relevance in a caption.) (The preceding unsigned comment was added by 139.80.123.36 (talk • contribs) 14:41, 17 July 2008.)

Thanks for the feedback. I've added a caption to the Venn Diagram. Ideally we would add a series of bivariate scatterplots to this resource. -- Jtneill - Talk - c 11:45, 17 July 2008 (UTC)

I think now that there are bivariate scatterplots, that the venn diagram is not very useful, and it more likely to be confusing (for example, what is the event space for this diagram?). It should be removed.Wsntw 05:26, 19 March 2009 (UTC)

[edit] Remoe "correlation explore" link

The demo linked shows different correlations in such a way as to suggest that 'r' indicates both the slope of the correlation line as well as the strength of the correlation. I find this confusing. The main graphic presented here and on wikipedia's correlation article (Correlation_examples.png) does an excellent job of clearing up the issue, by showing perfect correlations with different slopes as well as various strengths of correlation all with the same slope (and, the wikipedia caption for the image is more informative). The linked demo does the opposite, by increasing the ambiguity.