Dr. Goddard developed a technique that he thought would increase IQ in adults. He obtained a random sample of 52 adult Americans and randomly assigned half of them to a control group and half to the experimental group. He did nothing to the control group, but he administered his IQ-increasing treatment to the 26 in the experimental group. Afterward, he measured IQs and found that the mean for the control group was 100, while the experimental group had a mean IQ of 102. The standard deviation in both groups was 15.

Dr. Goddard found sM1−M2=4.16, t = 0.48, r 2 = 0.46, and that the 95% for the difference between population means ranged from −6.36 to 10.36.

(a) Did Dr. Goddard reject the null hypothesis?

(b) What conclusion should he reach about whether his treatment works to increase IQ?

(c) How big is the size of the effect as determined by r

2?

(d) What information does the confidence interval give on how sure we are about the impact of the IQ-increasing technique? (e)

How worried are you that Dr. Goddard made a Type II error?

(f) Do you recommend replicating with a larger sample size?