Study Guide for the Psychology 138 Final
The previous 2 study guides are still good guides as to previously
tested material.
Exam 1 Study Guide
Exam 2 Study Guide
Here are concepts from the new material that you
should know and have mastered:
1. Be able to examine the results of a z-test, a t-test (all 3 types),
a correlation, or a regression and be able to interpret what they mean.
For example,say that I am comparing the average heart rate of people
given Drug X to the heart rate of people given a placebo. Suppose I say
that
the mean for the experimental group is 70bpm and the mean for the
control group is 60bpm. The independent samples t-test has 16 degrees
of freedom and t = 2.7 with p = .015. Is the bpm of the experimental
group significantly higher than that of the control group?
What is the null hypothesis? (Drug X
has no effect on heart rate and
thus the mean heart rate will be the same in both groups.)
Should you reject the null hypothesis? (Yes. p < .05)
State your conclusion in non-statistical language: It appears that Drug
X increases heart rate.
2. Be able to know which test should be given in different situations.
3. What assumptions are made in t-tests?
A)
Normality - The
distribution of sample means is approximately
normal.
B)
Homogeneity of Variance- The variability of the
sample is
approximately the same as the variability of the population. In an
independent-samples t-test, there is the additional assumption that the
variance in both groups is approximately equal.
C)
Independence - The
probability of any score being selected does not depend on the
selection
of any other score.
4. Be able to answer questions like this:
A) A researcher randomly selects
a number of people to be included in a study and also says, "If you
know anyone else that might want to participate, bring them along with
you." When the researcher runs whatever hypothesis tests that were
planned, what assumption has been violated? (The Independence
Assumption).
B) In order to do an independent-samples t-test, one first has to check
the results of the ______ test to see if the Homogeneity of Variance
assumption holds. (Levene's)
5. Be able to tell why a z-test is better than a t-test. (A z-test uses
the population standard error whereas the t-test uses the estimated
standard error. Thus, the z-test is more accurate and more powerful.)
6. What is the relationship between a 1-sample t-test and a
paired-samples t-test? (A paired-samples t-test is a special kind of
1-sample t-test. For each pair of scores, a diference score is created.
A paired-samples t-test is a 1-sample t-test on the difference scores
with μ = 0).