Week 7_Extra practice_independent samples_KEY

Independent samples t-test worksheet KEY 1. Describe what the null hypothesis is conceptually testing with an independent samples t-test. The null hypothesis states that there is no difference between the means of the two populations (or treatments) from which the samples were obtained [EDIT]. 2. A company would like to assess the impact of using a professional trainer to conduct a confidence-building workshop with their salespeople. A sample of 32 workers was obtained. Half (n = 16) attended the workshop and the other half (n = 16) served as a control group. A few days later, each of the participants is given a questionnaire measuring their level of self-confidence. The data are as follows: Controls Workshop M = 17.04 M = 22.19 SS = 90 SS = 270 a. Do these data indicate a significant difference in self-confidence between those who attended the workshop and those who did not? Assume an α = .05 and a two-tailed test. s p 2 = SS 1 + SS 2 df 1 + df 2 = 90 + 270 15 + 15 = 12 s M1-M2 = √ 12 16 + 12 16 = 1.22 df =( 16 − 1 )+( 16 − 1 )= 30 t crit = ± 2.042 t = 22.19 − 17.04 1.22 = 5.15 1.22 = 4.22 Reject the null hypothesis (4.22 > 2.042); this represents a significant difference . b. Calculate r 2 to measure the size of the treatment effect. r 2 = 4.22 2 4.22 2 + 30 = 17.81 17.81 + 30 = .37 c. Write up the results in APA format. People who attended the workshop showed higher levels of self-confidence than those in the control group, t (30) = 4.22, p < .05, r 2 = .37.