I've investigated the preferences of students for English teacher personas by using the ranking scale questionnaire. There are 5 categories of personas to be ranked; each category contains different number of sub-characteristics ranging from 6 to 10 items. However, students were asked to rank each category only 1 to 5; 1 is the most preferred while 5 is the least preferred) which means some characteristics/items aren't selected or ranked. In this case, what statistic should be used to analyse the reliability of this ranking scale questionnaire? It is difficult for me to deal with this case since some students didn't rank all items provided. Kindly suggest me what statistic should be used to analyse its reliability.
Since you are ranking top 5, you can do two things:
1) simply present the total count of each item and then sort them using that total.
2) after doing 1), rank the numbers of 1st, 2nd, 3rd, 4th, and 5th in each item and then sort them using the count of 1st (because that is the most important).
Locky Law Many thanks for your answer. I did what you mentioned already. However, I still wonder if that is the way to check the reliability of my questionnaire. Are there any more statistic tools used (i.e. Cronbach's Alpha) for reliability check?
- Badges
- Science topic
- Similar topics
- Linguistics
- Philology
- Language
- English