In: Statistics and Probability
If we used a standard where H0 is rejected at p < .10 (as opposed to p < .05), we would decrease our rate of [(a) Type 1 or (b) Type 2] error compared to a standard where we reject at p < .05. Explain.
Type 1 error is the error in which we reject H0, when H0 is true. If the range of p is increased from 0.05 to 0.10, then it will decrease the rate of type 1 error. For example, consider the case where value of p is greater than 0.05 and less than 0.10 i.e., 0.05 < p < 0.10, then H0 is rejected in this case as compared to previous case in which H0 is not rejected as the range of p was 0.05 to reject the H0. Therefore, increasing the range of p decrease the chance of rejecting H0, where H0 is true. It is the case of weak evidence of rejecting H0.
Type 2 error is the error in which we do not reject H0, when H1 is true(i.e., H0 is false). If the range of p is increased from 0.05 to 0.10, then it will decrease the rate of type 2 error. For example, consider the case where value of p is greater than 0.10 i.e., p > 0.10, then H0 is not rejected in this case as compared to previous case in which H0 is not rejected if the value of p is greater than 0.05. Therefore, increasing the range of p decrease the chance of not rejecting H0, where H1 is true. It is the case of weak evidence of rejecting H0.