Answers to the checkpoints of chapter 6
- If there are differences between the sample and test value, and if the sample has a normal distribution.
- The hypothesis is that the age is lower, so the hypothesis has a direction.
- The null hypothesis is not rejected (p value = 0.0668; conclusion p > 0.005).
- You would use a t test if the population variance is not known and (in some cases) for small samples.
- From 120 onwards, the shape of the t distribution is the same as a normal distribution.
- A critical value is the t value or the z value associated with the chosen significance level.
- If you have two dependent groups and the population variance is unknown.
- Because there is an association in the data that has been collected.
- In terms of whether the population variance is known or not known.
- From n = 120 onwards.
- That is a test in which the groups being compared have the same variance.
- At df = 199, the critical t value is 1.96; if you perform a z test, the critical value is also 1.96, because the sample for this is sufficiently large.
- No, you don’t reject it; p = 0.06.
- Yes, at 5% with a one-tailed test, you would.
- You could adjust the degrees of freedom.
- You have made a Type I error.