I have two small samples (n

*1*=10; n

*2*=10) where I have success ratios of 3/10 and 1/10. The test is a Wason psychology test where I cannot estimate

*p*or

*q*, however I would like to compare said ratios and see if the difference is statistically significant. My textbooks all require samples at n≥30 to use Z values and normal distribution, so I don't know if this is possible. My H

*1*would be π

*1*> π

*2*.

Any help greatly appreciated, and I am sorry if this question is in the wrong place. Thanks!

Augusto