Dear Azubuike,

I am currently trying to solve the exact same problem.

When I find a solution to it, I'll let you know and post it here.

Best regards,

Vincent

PS: If you've already found one, please let me know.

Ok Vincent will keep you posted. 'hope to have a breakthrough soon.

Dear Azubuike,

I think I might have found a solution (although it does not include one of the programmes mentioned by you):

In the program "Mystat" (which is a free student-version of "Sysstat", downloadable here: http://www.systat.com/MystatProducts.aspx), you can create scatterplots and these have the function "confidential ellipse" in the "options"-sheet, where you can define the p level according to your needs.

I attach an example of the graph I created yesterday (p = 0.95) and hope that this is also what you were looking for.

I wish you all the best and good success with your thesis.

Vincent

Yes its what i'm looking for. But the data you used show wide variability that's why some of the ellipses are crossing the lower axis.

THanks alot for your efforts. I'll try and explore the mystat software. Cheers

With pleasure. The problem of axis crossing can be avoided by using a smaller p value than 95%. Have a great day.

In STATISTICA it's easy: graphs - scatterplots

Perfect, thanks.

Is there an equivalent way in SPSS?