A researcher wants to see if sleep can affect test scores. A party is thrown the night before an exam and all students are required to attend. No one is allowed to get more than 4 hours of sleep prior to the exam

Question

A month later, the researcher does a posttest and all students get at least 8 hours of sleep the night before the exam. This results in statistics of t = 2.84, degree of freedom (100 – 1) = 99, and p =.03. Which statistical test, if any, was conducted in this study? 1. An independent sample t test.
2. A dependent sample t test.
3. A nursing student t test.
4. An analysis of variance or ANOVA.

Answer 1

2
Rationale 1: The first clue should be the statistic of t =. However, this just tells you it is a t test, not which type of t test. There are two additional clues to help you figure out whether it is an independent sample t test or a dependent sample t test. The most obvious is the fact that there are not only two sets of test scores, but the same group took the test both times. The last clue is the degree of freedom. The formula for df is (n + n) -2 for independent sample t test but N - 1 for dependent or paired t test. In this case, a small n represents the sample of each group and the capital N represents the sample size of the entire study.
Rationale 2: As in this case, the paired or dependent sample t test usually has two measures on one group, a before and an after measure, and the degree of freedom would be 99 since the sample size is 100 – 1, which is the number of groups.
Rationale 3: There is not a common statistical test with this name.
Rationale 4: ANOVAs are reported as F values not t values.