Let's say if there are 2-6 proportions in a group of people or animal, what statistic test will be able to tell the number of one proportion is significantly different from other proportions?
Jos Feys Yes, there is no repeated measure. I will try chi-squared test. Thank you very much!
Honestly, I don't think there's enough information in the question to give a recommendation. With a small amount of data, it's best to just give similar, hypothetical data so that readers can see the arrangement of observations. ... Specifically, when you say proportions, are the data actually counts, as could be used in a chi-square test of association ? Or are they inherent proportions, just as "I estimate that 45% of the apple surface is blemished" ?
Sal Mangiafico Thanks for the clarification. The data are counts of patient and animal numbers.
You say: "what statistic test will be able to tell the number of one proportion is significantly different from other proportions?" Do you mean one specific proportion (say you have one group given a treatment and you want to compare it to, collectively, all others), or any one of the groups different from the collective of all the other groups, or even if any pair differ? Also, I didn't understand what you went by 2-6 proportions. And given people use the phrase differently, what do you mean "significantly different", something like more than a 10% difference in the proportions or something often less useful, like p < .05.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS