I am finding this weird to ask but I will go ahead anyway;
I am conducting my PhD study in which I wish to collect pilot data from one person (representing an institution). My instrument is a checklist with 47 Yes/No items and an explanation/evidence of the response. I found that KR-20 could be used to estimate the reliability of this instrument. Is it common practice or does it even make sense to do that? Which other method would you propose for this?
Hi Daniel MAKINI Getuno
First of all, if I have to speak independently of the sample size, if the measurement tool you use is a questionnaire or checklist, it may be wrong for your measurement instrument using the KR-20. It is suitable for structured measurement tools such as achievement tests.
Coming to the other question, you cannot calculate this statistic or any statistics with 1 person. If there is no variance, there is no statistics. As a result, I suggest you get logical evidence (such as expert opinion), not statistical.
Good luck with your research
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments