Questionnaire Testing

Hi everyone,

I'm currently an undergraduate student trying to find validity and reliability of questionnaires. I have performed equivalence testing between the questionnaires for validity, test retest for reliability, and also have done Pearson correlations to identify any relationship.

I was told I should also do paired sample t tests between the different questionnaires. Although I don't understand why because this finds the significant difference when I am trying to find that the questionnaires are similar.

If anyone could help to explain this to me that'd be great!


Doesn't actually exist
Wouldn't the standard way to calculate something like this would be some type of measurement invariance procedure through Structural Equation Modelling?


Not a robit
You have two questionnaires that you are using to try and measure some phenomenon, correct? Is this a supervised setting where you know the true outcome classification for the phenomenon? If not, how do you know either questionnaire is accurate.