Form the Delphi survey; we got the level of preference of the variable to retain it or not in our study. After getting the values on how to rank the variables, whether to take the mean or median value of the variables to rank them or some other methods or tools and techniques are available. Please suggest the ways to rate the categorical or interval data from a Delphi method.
Please suggest your answer with some references for which i can use them as citation.
Oridnal at-best.
See, Stone et al (1999) The Science of Self-Report
Hope this helps
Matt
Single Likert-scored items are ordinal data. You could look at the median score for each variable in your Delphi and rank order then from high to low, but this process is so basic that I doubt if you are going to find a formal citation for it.
Thanks Matthew Kerry for your suggestion. I am looking forward to find this book.
David L Morgan, Sir. I doubt weather I can use the mean value from the Likert scale where most of the research based on that. If I take median as my assessment criteria, then I will have only five values for ranking as the scale of Likert is based on 5 rating scale. Variables more than 10, most of the variables have the same rank which I do not want to use in my research.
When you say, "most of the variables have the same rank which I do not want to use in my research," do you mean that you do not want to use the ranking in your research or that you do not want to use those variables?
David L Morgan, I want to use both, however, I need a range of rank, so that I can use different rank weight methods. For that, if I take median I will have only 5 options for the ranks where numbers of variables are more than 40. Therefore, I need some citation where by the method of Likert scale mean value can be useful for outcome.
Thanks for your concern.
Partha Sarathi Mishra
Did you came across any other method other than mean or median for ranking of variables.
- Badges
- Science topic
- Similar topics
- Statistics
- Survey