Guide to using the learning objectives
1.
Explain2
the advantages of using a multiple group experiment, rather than several
simple experiments, to compare the effectiveness of three or more different
treatments.
2.
Explain2
why a researcher examining the effects of two different treatments might
want to include a no-treatment control group.
3.
Explain2
why using three or more different levels of treatment allows you to
explore the functional relationship between the treatment variable and the
outcome variable. Then, explain2 why you would want to explore the
functional relationship between two variables.
4.
Explain2
why multiple-group experiments may have better construct validity than
simple experiments. In your answer, be sure to address hypothesis-guessing,
empty control groups, and confounding variables.
5.
Rank6
the following experiments on the effects of exercise and mood in terms of
how much construct validity you think each would have. Justify6 your
rankings.
a.
A 30 min of exercise group and two
control groups: a group that watched music videos and a group that took an
acting class.
b. Two levels of exercise (0 min,
30min)
6.
Rank6the
following experiments on the effects of exercise and mood in terms of how much
construct validity you think each would have. Justify6 your
rankings.
a.
Three levels of exercise (0 min, 15 min,
and 30 min)
b. Two levels of exercise (0 min, 30 min)
c.
Four levels of exercise (0 min, 15
min, 30 min, and 45 min)
d. Four levels of exercise (15 min, 30 min,
45 min, and 60 min)
7.
Define1
within-group variability. Define1 between-group
variability. Compare4 and contrast4 within-groups variability and
between-groups variability.
8.
Explain2
why, all other things being equal, the greater the variability between
group means, the greater the chance that the treatment had an effect.
9.
Defend4
using the term "error variance" instead of "within-groups variance."
10.
Defend4
the following statement: "It is misleading to call between groups
variance 'treatment variance.'"
11.
Explain2
why you should not use t
tests to analyze the results of multiple group experiments.
12.
There
are several terms (all of which end with "variance") for the top part of the F
ratio. Name1 those terms.
13.
There
are several terms (all of which end with "variance") for the bottom part of the
F ratio. Name1 those terms.
14. Justify2 the following comment, "if the null hypothesis were true, the F ratio would usually be close to1.0."
15.
Explain2
how to use an F table to determine whether an F ratio is
statistically significant.
16.
You
have an experiment with 40 participants and four levels of the independent
variable. Calculate3
a.
the degrees of freedom for the
treatment
b. the total degrees of freedom
17.
Describe2
what a statistically significant F for the treatment effect
indicates in a multiple-group experiment. State1 two important
questions that are not
answered by obtaining a significant F for the treatment effect.
18.
Contrast4
post hoc t tests with conventional t tests. Defend4
the use of post hoc tests for determining which groups' means differ from one
another.
19.
State1
the requirements for conducting a valid post hoc trend analysis and
explain the implications of those requirements on planning a multiple-group
experiment. Then, defend4 the use of post hoc test trend analysis.