Review Sheet for Exam
Two
- You will be asked to complete the
major types of computations that you have completed on
the quizzes. You will be asked to do the following:
- calculate confidence
intervals for the mean using the t or z
distributions (depending upon the sample size)
- complete formal hypothesis
tests about whether a sample mean is different
from a specific value a, using the t or z
distributions (depending upon the sample size)
- calculate the exact p-value
when using the z distribution (you won't have to
calculate it for t).
- You will be asked to complete two
formal hypothesis tests, using the steps discussed in
class. Be sure you know
- how to formulate hypotheses
(using Greek letters and in words)
- what the assumptions of the
tests are and how we know if the second one is
met
- how to draw the sampling
distribution graph and label important things on
it (like the critical value(s) and the test
statistic)
- how to figure out what the
critical values are
- how to specify the decision
rules (formally, as in class and on quizzes)
- how to determine whether to
reject or accept the null and what your decision
means in words for the specific problem.
- Many of the true/false and multiple
choice questions tap conceptual aspects covered in class
and often on quizzes. Know the following:
- Type I and II errors (know
definitions and how to describe them in the
context of specific problems; know that Type II
errors sometimes occur b/c of low power)
- power and what influences
it
- alpha (& know that this
IS the type I error rate!)
- how we resolve the tension
between wanting a low type I error rate and high
power
- how to interpret a
confidence interval
- how to test hypotheses
using a confidence interval
- what a p-value is and how
that's related to whether we reject the null
hypothesis
- when we reject the null
hypothesis, dependent upon whether or not the
test statistic falls in the rejection region
- upper- and lower-tailed
tests, one- and two-tailed tests, directional and
non-directional tests
- what factors influence the
width of a confidence interval and how they do so
(i.e., N and confidence level)
- how the t and z
distributions differ and under what conditions
they look the same