 |
 |
Hi Thais, I'm a bit surprised that it gave significant results in this case. I'm more compelled to believe there is an error somewhere in your analysis. Are you sure the nuisance variables have all been mean-centered? This could be an issue when testing the intercept (the overall average, across groups). All the best, Anderson On 19 June 2017 at 16:04, Thais Minett <[log in to unmask]> wrote: > Dear Anderson, > > I tried the new model you suggested and it gave me a significant result as > I was expecting. > > I don't understand where I messed up in the previous model. I did enter > the correct number of zeros/EVs in the real model. Basically, even if I > increased the range of display of results in fslview from min 0 and max 1 , > nothing shows up as if it was an empty file. Is there any explanation for > that? Is the 0.5 0.5 values from my matrix correct? > > Regards, > > Thais >