1
$\begingroup$

I am writing a paper to find the relationship between motivation and the comfort level in learning English. I prepared a questionnaire with two sets of questions (one for motivation and one for comfort level using Likert scale) to find the relationship between the two. I collected data from 64 participants. Then in order to find the correlation between them, for each participant, I summed up their motivation and comfort level scores in order to have two total scores, one for motivation and one for comfort level for each participant. Now, my question is why I should sum them up? what if I take scores for each question separately? what are the pros and cons of each method? Also could you please introduce a source so I can learn more about it.

Thank you in advance for your help

$\endgroup$
  • 1
    $\begingroup$ Who said that you should sum them up..? You can use many different approaches including IRT models that use individual items to estimate underlaying latent variables. Such models can be easily extended to regression models. $\endgroup$ – Tim Jul 16 '18 at 10:33
0
$\begingroup$

You might be interested in canonical correlation analysis, which will tell you linear combinations of two sets of variables (which in your case would be a set of scores for motivation and a set of scores for comfort level) that are maximally correlated with each other.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.