My pre and post tests contain blend of questions such as fill in the blanks; mcqs; true false; short answers. How am I suppose to check its reliability? The results are not significant. With in treatment groups; post test results are better and significant as compared to pre test scores. However; post test scores are not significant between control and experimental groups. This research is quasi experimental. How am I suppose to address its reliability? The data is parametric though. I have used t-tests for comparative analyses. Results are not significant. Pearson coefficient correlation is less than 0.2 and is not significant. Nothing is helping. However, subject experts validated the tools. But its reliability has been questioned? How am I supposed to address this in my research report?
Your question about "reliability" is too vague.
Maybe this helps a bit:
https://www.socialresearchmethods.net/kb/reliable.php
http://www.open.edu/openlearn/science-maths-technology/mathematics-and-statistics/statistics/confusing-terms-statistics
I agree with Jochen, it is not clear what you are asking. What is the research question that you are trying to address and which instruments/samples are you assessing the reliability for? Or do you mean something different?
Hi, Anna,
The non-significant difference between your experimental and comparison groups indicates that your intervention was not the factor that resulted in the significant change from pre to post-test. It suggests to me that there is a factor that has not been controlled that is influencing the internal validity of your design.
Traditionally, "reliability" refers only to consistency of the instrumentation you used in your study. It is a very simple thing to check on the internal consistency using Cronbach's alpha is standard statistical packages. With a new instrument such as yours, I would also suggest looking at dimensional scaling with exploratory factor analysis. What you may find is that some of your items should be excluded as you develop summative factors/variables. These two measures will lend some small degree of validity to the face validity provided by your experts.
If you change the factor structure or develop new variables through this process, you can then re-calculate your pre/post and experimental/comparison groups.
Hello, what tools will I use to check the reliability of my pre and post-test? Thank you
- Badges
- Science topic
- Similar topics
- Psychology
- Social Psychology