Message/Author |
|
|
Hello, Where can I get significance levels for the between-level (Level 2) correlations? I have to report them in a manuscript. I get Correlations matrix as part of the "ESTIMATED SAMPLE STATISTICS FOR BETWEEN GROUP" output but not their significances. Thank you! |
|
|
You would have to model these correlations using WITH and then find them in the regular output. |
|
|
Thank you very much Bengt. When I model those correlations I get an error message: *** ERROR One or more between-level variables have variation within a cluster for one or more GROUP clusters. Check your data and format statement. Between Cluster ID with variation in this variable Variable (only one cluster ID will be listed) BULLIEDT 19 Here is part of my syntax: %BETWEEN Group% anx on GrsizeM ageM bulliedt1p; s1 on GrsizeM ageM bulliedt1p; s1 with anx; anx with GrsizeM ageM bulliedt1p; GrsizeM with ageM bulliedt1p; ageM with bulliedt1p; plot: type = plot2 plot3; output: sampstat StdYX tech1 tech3; Here bulliedt1p is a proportion of bullied kids per group, and doesn't vary within group... Can't figure out what is wrong with me syntax. Thank you again. |
|
|
We need to see your data, input,and output - send to Support along with your license number. |
|
|
Thank you, Bengt. I think I figured out why I get the error message- there are missing cases coded as -999 at the group level. So the group- level data looks like: 72.7 72.7 72.7 72.7 72.7 -999.0 -999.0 72.7 72.7 72.7 How can I code them so I don't get the error message? Thank you again! |
|
|
You just have to use the option for missing data flags - see UG. |
|
Back to top |