Hypothesis Testing
1. State H1 and H0
a. Hypothesis: This semester’s students will have systematically different anxiety
scores from the general population.
b. Null Hypothesis: This semester's students will not have systematically different
anxiety scores from the general population.
2. Determine the criteria and cutoff score at which we reject the null hypothesis.
a. We are given a 95% confidence interval for this problem. This means there is a
5% chance that a score would be in the rejection area without the null hypothesis
being false.
b. This gives cutoff scores of 1.96 and -1.96, with any z-score past these numbers
rejecting the null hypothesis.
3. Compute the sample statistic and identify its place in the comparison distribution.
a. Population Mean: 5.5 is given
b. Sample Mean: 258/49 = 5.265
c. Standard difference: 0.5/ 49 = 0.0714
4. Calculate the Z-score and determine whether or not to reject the null hypothesis.
a. Z-score: (5.265 - 5.5)/0.0714 = -3.3
b. As -3.3 is greater than -1.96 and within the rejection area, we can reject the null
hypothesis
Effect Size.
1. |5.265 - 5.5|/0.5 = d = 0.47
Power calculation
1. Find population and sample information
a. μ = 5.5
b. σ = 0.5
c. a = 0.5
d. d = 0.3
e. N = 49
2. Draw comparison distribution and predicted distribution for research hypothesis
a. Standard difference = 0.0714
b. Estimated μ = (0.3*0.5) + 5.5 = 5.65
3. Find raw score that corresponds to cut-off Z-score of distribution
a. M = (1.96*0.0714) + 5.5 = 5.64
4. Find z-score of predicted raw score
a. Z = (5.5 - 5.65)/0.0714 = -2.101
5. Use unit normal table to find the percent of the predicted distribution that is beyond that
newly calculated Z-score
a. Percent in body for Z-score: -2.101 = 98.21%
b. There is a 98.21% chance that we would find this score if we ran the study