For a survey using Likert measured items, is it necessary to conduct two separate pretests and run reliability test for each before conducting the actual study.
I want to conduct a survey to assess the perceptions of people on using mosquito nets. The sample size is 400. I am using a Likert measured items and want to be sure of adequate consistency of the items.
I know of no reason why you should run two separate pre-tests. Are you thinking that you need to have two data tests to conduct tests of reliability? If so, you need to look into Cronbach's alpha as a test of internal consistency, which requires only a single data set.
Because there have been so many questions here about LIkert-scored items, I have put together a set of recourses on this topic:
https://www.researchgate.net/post/How_can_one_help_people_create_simple_scales_from_Likert-scored_items
Thank you David. I do agree with your response. Sure the link will be very helpful to everyone on this platform.
Usually two data sets are not required for the pretest, but if you have two different groups such as providers and users, then it may be useful to pretest with the separate groups. But for homogeneous sample this is not desirable.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size