Hello,
I have a question about common method biases. I have collected data using an online survey for latent constructs each measured on several items. However, I have also manifest variables like gender, age and education. Should I include also those manifest variables in the empirical test for common method bias?