Measuring some dependendant variables such as instrinsic motivation, autonomy, competence and relatedness perceived towards gaming for correlation and some "independent" such as hours played, gender, culture of gaming, etc.
My idea is that for inferential it would be easier to find evidence using a larger scale to avoid central tendency bias, but some people keep suggesting to stick to 4. Any (better if is supported with some literature/experience) light on this?
If you are using Likert Scales, Dawes' comparison of 5, 7 points scales https://papers.ssrn.com/sol3/Papers.cfm?abstract_id=2013613 and Chang's comparison 4 and 6 points scales http://apm.sagepub.com/content/18/3/205.short might be useful.
You probably should not use a 4-point scale as there is no mid-point. Here is an article on the topic: http://www.measuringu.com/blog/scale-points.php
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA