Following on from my previous question on forming scale scores, my sample has 100 people who have answered 10 likert scale questions each on two academic subjects: Maths and English.

The likert scale is a 5 point scale, going from not important to absolute importance.

I have aggregated the scores:

  • The maximum score a person can get is 40 (which is 10 responses x 4 which is the maximum score on the likert scale e.g. 4 = absolute importance).

  • The minimum is 0 (which is 10 responses x 0 which is the minimum score on the likert scale e.g. 0 = not important)

If a person gets 20 points (total of 10 likert scale responses for Maths) and 30 points (total of 10 likert scale responses for English), my questions are:

1. Is 20 or 30 a composite score or an index score? (given that they are both based on subjective weighting aligned with the likert scale response)

2. Can I do a correlation analysis on the two scores? (given they are composite/index scores, hence represent a latent variable)

Many thanks

  • $\begingroup$ I have merged your duplicate accounts. Because you have multiple questions, please register and use only that account. $\endgroup$ – whuber Aug 3 '11 at 14:16

Composite score, scale score, and index score are terms that are often used interchangeably. As to your question 2, I'm not sure what makes you hesitate to use correlation; the usual assumptions involving correlation should apply here. Otherwise, you'll find more information at this page on this site


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.