forum.smartpls.com

Hi
I have a latent construct with tow manifest variables (reflective) highly correlated (.91). smartpls estimate the cronbach alpha to .55 while SPSS give .82. I d'on't understand why...could you help about this?
thanks
carole

I think this has been answered several times here in the forum: The difference comes from standardizing the variables in PLS. If you calculate Cronbach’s Alpha on the standardized variables in SPSS you should get the same result.

Another possibilities are untreated missing values or problems when importing variables with decimal signs (and not correctly choosing the decimal separator).

thanks a lot. I have searched on the forum without any success..I have probably entered an incorrect request...