I am planning to get a questionnaire filled from the students where they will rate their skills like communication skills, creative skills etc from 1 to 5. Can I call it Likert scale and can I use alpha value to show its validity and reliability upon the responses got from the 35 students
How many questions are there and what are the nature of those questions?
A difficulty of the lIkert scale is that it is arbitrary judgments, but given a strict numerical value. Is a "2' score truly twice as "good" as a "1" - a "4" twice a "2"? Conventional alpha scoring would be thrown off by this. There are ways to work non-parametrically or using binomial calculations that avoid this, and there is a method to convert subjective opinions (such as agree - neutral - disagree) to a continuous scale. I've used this last approach with a great deal of success, it was taught by the US Naval Postgraduate School and I do have a paper on it available.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability