| Statistics for the Behavioral Sciences, 4/e Michael Thorne,
Mississippi State University -- Mississippi State Martin Giesen,
Mississippi State University -- Mississippi State
Alternatives to t and F
Problems- For each of the following, tell which significance test would be most appropriate.
- two small, independent samples; ordinal scale data
- two dependent samples; interval scale data
- three or more independent samples; ordinal scale data
- two dependent samples; ordinal scale data
- two independent samples; frequency data
- Seven children from families in which there is only one child and six children with at least one sibling
are rated for willingness to share toys with another child. Each child is given a rating from 0 (no
sharing) to 10 (virtually complete sharing) during a 20-minute observation period. Use the appropriate
test to compare the groups, and tell what your decision means in the context of the problem.
Only Child | Child With Sibling(s) |
| 5 | 10 | 3 | 9 | 2 | 9 | 2 | 7 | 1 | 4 | 0 | 2 | 0 | |
- Thirty-one randomly selected rats are assigned to one of three different experimental diets. After 30
days on the diets, each animal is given a test of irritability to handling. In the test, the behavior is rated
from 0 to 15 with a higher score reflecting greater irritability. The scores are shown here. Do an overall
significance test. If a significant result is obtained, do all pairwise comparisons. Tell what each
decision means in the context of the problem.
Diet A | Diet B | Diet C |
| 6 | 14 | 4 | 5 | 12 | 4 | 5 | 12 | 3 | 4 | 11 | 2 | 3 | 9 | 2 | 2 | 7 | 1 | 1 | 7 | 1 | 0 | 5 | 0 | 0 | 4 | 0 | 0 | 1 | 0 | 0 | | |
- A trained speech analyst has received brief taped excerpts of the speech of 18 parents. Ten of the
parents have schizophrenic children, and the remaining 8 have nonschizophrenic children. Without
knowing whether the parent has a schizophrenic child, the analyst has rated the excerpts from 0 to 20
for defectiveness of speech. The groups do not differ on variables such as IQ, age, education, or social
class. Is there evidence for a difference in the speech patterns of the parents of the schizophrenic
children?
Parent of Schizophrenic | Parent of Nonschizophrenic |
| 16 | 12 | 15 | 11 | 13 | 10 | 12 | 9 | 9 | 9 | 7 | 5 | 5 | 4 | 3 | 2 | 3 | | 2 | |
- A self-rating scale was used to measure attitudes toward risk taking before and after alcohol
consumption for 12 persons. A high score indicates a positive attitude toward risk taking; a low score
indicates greater concern. Compare the before and after ratings.
Person | Rating Before | Rating After |
| A | 14 | 17 | B | 14 | 19 | C | 13 | 14 | D | 12 | 9 | E | 11 | 12 | F | 9 | 9 | G | 9 | 15 | H | 8 | 7 | I | 5 | 9 | J | 4 | 8 | K | 2 | 1 | L | 2 | 5 |
- Twenty-four students are selected randomly from a large introductory psychology class and assigned
randomly to one of two treatment groups. Half are given an alcohol-flavored drink, and the other half
receive a drink containing an ounce of alcohol. Ten minutes later, each student fills out a self-rating
scale measuring attitudes toward risk taking. Assume the data are ordinal scale at best. The results are
shown here. Compare the two groups. As before, a high score indicates a positive attitude toward risk
taking.
Alcohol Group | No-Alcohol Group |
| 14 | 19 | 14 | 17 | 13 | 15 | 12 | 14 | 11 | 12 | 9 | 9 | 9 | 9 | 8 | 9 | 5 | 8 | 4 | 7 | 2 | 5 | 2 | 1 |
- Matched pairs of parents have written letters to a child in a state mental institution. One member of
each pair has a schizophrenic child, and the other member has a nonschizophrenic child in the hospital.
One letter from each parent has been rated for double-bind statements (incompatible ideas and
feelings) on a scale from 1 to 7, with 7 reflecting a high incidence of double-bind statements. Use the
appropriate test to compare the groups.
Pair | Parent of Schizophrenic | Parent of Nonschizophrenic |
| A | 7 | 3 | B | 5 | 5 | C | 3 | 6 | D | 2 | 6 | E | 6 | 5 | F | 1 | 3 | G | 2 | 1 | H | 4 | 6 | I | 3 | 1 | J | 7 | 1 | K | 3 | 7 | L | 1 | 5 |
- An investigator wants to see whether creativity (divergent thinking) can be taught. In one class, the
teacher specifically rewards divergent responses during a 1-hour daily art period. In a second class, a
1-hour art period is held, but no effort is made to reward divergent responses. In a third class, a 1-hour
study hall is given while the other classes have the art period. At the end of the year, 10 students are
randomly selected from each class and given a standard test of creativity on which they receive a score
from 1 to 50. A higher score indicates greater creativity. Assume the data are ordinal scale at best.
Compare the classes with an overall test. If the result is significant, do all pairwise comparisons, and
tell what your conclusions mean in the context of the problem.
Class 1 | Class 2 | Class 3 |
| 48 | 41 | 42 | 46 | 40 | 25 | 43 | 27 | 24 | 40 | 25 | 22 | 39 | 13 | 18 | 38 | 11 | 17 | 35 | 10 | 9 | 28 | 10 | 8 | 27 | 9 | 8 | 15 | 7 | 5 |
Click here to view the answers (6.0K) |
|