MATH 10041 Chapter Notes - Chapter 8 .3: Type I And Type Ii Errors, Normal Distribution

• School: Kent State University
• Department: Mathematics
• Course: MATH 10041
• Professor: Rajeev Rajaram

Document Summary

Extreme value means something unusual has happened. Cannot use normal curve if the sample size is too small and/or if not randomly selected. We cannot make the significance level so super small (ex. 0 = 0%) because that increases the probability that we will mistakenly fail to reject the null. Ex. if the probability of convicting an innocent man is 0%, everyone goes free. But that means freeing a guilty person is 100% Statistically significant findings do not necessarily mean the results are useful. Fail to reject the null that you should actually reject. Don"t say you proved something --- say you demonstrated or solved .