| |
BGonline.org Forums
XBG 3ply RO: 53P21: Slot better as Split
Posted By: W Womack In Response To: XBG 3ply RO: 53P21: Slot better as Split (Andreas)
Date: Friday, 2 October 2009, at 1:17 a.m.
If the confidence intervals did not overlap you could determine significance, but with them overlapping you can't say for sure (see http://www.cscu.cornell.edu/news/statnews/stnews73.pdf). Does xg give you the standard errors in addition to the ci, if it does than you can determine significance. You could also attempt to back calculate the se from the ci, but you would need to know if xg used the normal or t distribution in calculating the ci.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.