Bernard,
the answer is yes. Chi square test is robust or is a non-parametric test because it does not assume normality or symmetry of data.
Hope this helps
However, what is meant by "chi-squared test" in the initial question ? This is also unclear to me. I suspect that Bernard is referring to the Pearson's chi-squared test for count data. Am I right ? If yes, the purpose of this test is to analyse a contingency table crossing the categories of 2 qualitative variables. For scored items, the first step should be to build this table, i.e. counting individuals in each cell of the table crossing the scores for each variable.
In order to evaluate the effect of fertilisation (four fertilizers) in the severity of cassava mosaic virus, I have decided to collect my data by using four scores according to the level of infestation (0%, 25%,50% and 75%). At the end of my experimentation, I have constructed the contingency table with the total number of plant for each of scores. then, I performed Pearson's chi-square test followed by post-hoc chi-square test( by using "stats" and "fifer " packages of R software). So I want to know if I'm in the right direction ???
- Badges
- Science topic