Skip to main content
16 events
when toggle format what by license comment
May 8, 2017 at 17:37 answer added HXD timeline score: 2
Mar 8, 2017 at 2:11 comment added Carl You need $n$ to determine the strength of the relationship. Also see stats.stackexchange.com/a/265924/99274.
Apr 2, 2016 at 17:27 answer added Lionel Benza timeline score: 14
Dec 1, 2015 at 21:46 answer added cd98 timeline score: 1
Jun 5, 2012 at 11:14 history edited Macro CC BY-SA 3.0
fixed formatting, some wording
Jul 20, 2011 at 12:30 comment added Glen I would just like to add a slope can be very significant (p-value<.001) even though the relationship is weak, especially with a large sample size. This was hinted at in most of the answers as that the slope (even if it's significant) says nothing about the strength of the relationship.
Jul 20, 2011 at 0:34 comment added Mog Ah, ok. That makes a little more sense than whether the two variables themselves are correlated...after all, I thought that's what I was trying to find using the regression. And that failing the test indicates I should be cautious interpreting the slope and p-value makes even more sense in this case! Thanks @whuber!
Jul 19, 2011 at 23:39 answer added shabbychef timeline score: 8
Jul 19, 2011 at 21:34 history tweeted twitter.com/#!/StackStats/status/93433431966494720
Jul 19, 2011 at 21:29 comment added whuber The Durbin-Watson statistic is a test for serial correlation: that is, to see whether adjacent error terms are mutually correlated. It says nothing about the correlation between your X and your Y! Failing the test is an indication that the slope and p-value should be interpreted with caution.
Jul 19, 2011 at 19:51 vote accept Mog
Jul 19, 2011 at 19:47 history edited Gavin Simpson CC BY-SA 3.0
typo in title
Jul 19, 2011 at 19:32 answer added jedfrancis timeline score: 9
Jul 19, 2011 at 19:28 answer added Macro timeline score: 27
Jul 19, 2011 at 19:19 history edited Mog CC BY-SA 3.0
added information
Jul 19, 2011 at 19:09 history asked Mog CC BY-SA 3.0