I have considered a particular response from sample sets of two populations. Now I have the proportions. Except for Two Proportion Z Test, what test should I use? The sample sizes are above 100 for both the populations.
You can use the Wald test statistic which identical to the Z-test anyway, If your sample data are two variables both having values of 0 and 1, then testing for equal proportions is equivalent to testing for equal means of the populations and you can use the t-test. You can also present your data in a 2X2 cross-classification table and use the chi-square test.
The easiest test to grasp logically and explain to others would be to conduct a chi-square analysis of a two by two table. The columns would be the two conditions and the rows would be the number of success and not successes. It should be noted that the chi square analysis in this case would produce the exact same result as the z-test as they are mathematically equivalent in this case. Thus, my recommendation and the first two answers should yield the same results.
What I understood, you collected response from samples from two different populations. Selection of test depends on what type of variable is under study. If the data is qualitative type you can conduct Z-test for the equality of proportions of two population.
The Z-test can used for the equality of proportions of two populations in data of qualitative type and for the equality of means of two populations in data of quantitative type too.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size