There are no valid cases for VAR when GRP = .. Statistics cannot be computed for this

[tonia bird]

When I perform a normality test for my variables using the grouping variable named "Group", I got a warning message telling me that there are values in the grouping variable named "Group" equal ".." ! knowing that I revised all the values in it and they were all with specific values .Also the number of observations "N" in some of the variables is lower than the true number I have . Why does this happen ? and how can I fix this ?

 

[Dason]

Re: There are no valid cases for VAR when GRP = .. Statistics cannot be computed for

You should start by telling us which software you're using...

 

[Dragan]

Re: There are no valid cases for VAR when GRP = .. Statistics cannot be computed for

You should start by telling us which software you're using...

That certainly makes sense to me as well ("sounds" like S+ or SAS, but who knows for sure).

 

[GretaGarbo]

Re: There are no valid cases for VAR when GRP = .. Statistics cannot be computed for

If the software warns that there is a data value that equal "..". Then there is probably such a problem. A value "3.14" can easily become "3..14".

 

[tonia bird]

Re: There are no valid cases for VAR when GRP = .. Statistics cannot be computed for

sorry,I forgot to tell you that I am using SPSS 16.
also i revised all the data in the variable ,all are without any decimals( 1,2,3...)