Hi, I'm a marketing PhD and having some time to spend these days, I would like to improve my skills on statistics. we all used those concepts cited in the title for our studies, but not everyone knows the real meaning of some of them.
recently, my friend has asked me to help him to analyse his data, but his data set contains very small groups (so... normality issues and violating the assumptions of performing anova tests...), I advised him to perform T tests or nonparametric tests instead of Anova. So, I would like to know exactly why, and I'm looking for papers which explain and justify everything related to calculating sample size (for means) and which explain accurately what is effect size, the statistical power, the relationship between p value and CI if there is one, margin error, etc.
My friend says it's related to the ratio Sample/population. it's sounds true, but I learned in my statistics courses that we cannot perform parametric tests when the variable is not normally distributed no matter how big the population is. If there is here some researchers who are constrained to a small sample size for many reasons can you tell me how you deal with this.
Thanks in advance.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA