Suppose we want to assess the effect of music on concentration. We measure "concentration" by recording the time it takes a person to complete a simple word puzzle. Each person does a puzzle with and without music. A coin flip determines the order of the treatments. Our subjects are a sample of 18 volunteers from a statistics class. We test the hypotheses (at the 5% level): H 0: μ = 0 H a: μ > 0 where μ is the mean of the difference in times ("no music" minus "music"). The sample mean is 4 seconds with a standard deviation of 9 seconds. The sample is not random, but treatments are randomly ordered. In addition, the distribution of times is not strongly skewed, so we decide to use the t-test despite the small sample size. The t-test statistic is approximately 1.89 with a P-value of approximately 0.038. What can we conclude?

a. Music is not associated with statistically significant improvements in concentration as measured by time to complete a simple word puzzle.
b. Completion times only increased 4 seconds on average. For college students, music produces statistically significant improvements in concentration as measured by time to complete a simple word puzzle.
c. College students listening to music performed significantly faster on word puzzles than students who were not listening to music.
d. For college students, music is associated with statistically significant improvements in concentration as measured by time to complete a simple word puzzle.