Exam 15: More Nonparametric Tests

The null hypothesis in nonparametric tests is to the null hypothesis in parametric tests as

A group of 10 identical twins were assembled for a weekend of testing by a research laboratory. As part of the testing, one member of each pair was assigned to Group 1, leaving the other member to serve in Group 2. Both groups viewed a silent film which showed several "crime scenes."Afterward, both groups filled out a questionnaire in which high scores indicate that the judge thought the victim was more responsible for the crime. Group 1 viewed the film with a musical sound track; Group 2 heard no sound track. Determine if the sound track made a difference in the scores. Write a sentence summary of the results. Last Name Group 1 Group 2 Blick 12 14 Presbie 21 25 Fox 18 19 Hitt 21 20 Carlson 17 20 Boswell 13 16 Howell 15 20 Hendricks 24 26 Dunlap 10 11 Ward 9 8

Suppose 25 cardinals and 25 bluejays were ranked for resistance to air pollution with the rank of one going to the most resistant bird. The smaller sum of ranks, 738, was for cardinals. The conclusion from these data is that

For the Wilcoxon matched-pairs signed-ranks T test, one difference of zero causes a pair of score to be dropped from the analysis.

If you have a difference score of 0, it should be eliminated and N reduced by one for

Which nonparametric test should be used to analyze data from the following studies? a. When two people form a group, social psychologists call it a dyad. Suppose 30 volunteers who didn't know each other were randomly paired as dyads. Each dyad chose a topic to discuss and the amount of time A looked at B's eyes was recorded. At the end of 20 minutes, A rated his or her own attraction to B. The question the psychologist wanted to answer was whether you could predict attraction by amount of eye contact. b. In an experiment on memory, Winter and Uleman (1984) told participants to "remember what is read" and then read sentences describing people. (The plumber slips an extra $50 into his wife's purse.) Later subjects recalled what they could when given one of three cues. One kind of cue was a "disposition statement"- e.g., generous; another was a semantic cue, e.g., pipes; and the third was no cue at all. The dependent variable was memory for the original statement. c. An anthropologist observing a troop of baboons noted each animal's position in the dominance hierarchy. The anthropologist was interested in whether males or females were more dominant. d. Once the solar collector engineers finished the work described in Chapter 15 (finding that 14 gal/hr was optimal for their design), a competing manufacturer challenged them to a duel at high noon (actually at several high noons). Every day at noon for 14 days the manufacturers set up their collectors at a shopping center and obtained a measure of "temperature rise in one hour." Of course, some days were sunnier than others but both collectors got the same amount of sun on a given day.

For a Spearman r_s, a difference of zero causes a pair of score to be dropped from the analysis.

Suppose you wanted to show the degree of relationship between autokinetic movement experienced by subjects when they are alone and the movement experienced by the same subjects when they are being influenced by others. The nonparametric test that will show the degree of relationship is

On the question of when to use a parametric test and when to use a nonparametric test, your text discussed two issues. (This discussion has not yet resulted in a recommendation.) The two issues were

A Spearman r_s can be used to

require that the dependent variable scores be normally distributed.

Suppose a Wilcoxon matched-pairs signed-ranks test with N = 60 produced a sum of thenegative ranks of 643. A z value of leads to the conclusion that the null hypothesisshould be .

Your text discussed two issues for which hard and fast rules don't exist on the choice of para-metric or nonparametric tests. The issues were

Chapter 15 of your textbook explained how to analyze two-group designs that have two dependent variables.

a.An aerobics teacher believed that physical conditioning is an important factor in the ability to perform skills, even if the skill does not require strength or endurance. To test this hypothesis, she selected 10 high BMI (body mass index) women enrolled in her aerobics course. Before the first class, she tested them on a button-sorting task. After a full semester of regular exercise, she tested the women on the same button-sorting task. Scores represent the number of seconds required to sort the buttons. Analyze the data with a nonparametric test and comment on the aerobics teacher's hypothesis. Subject Pretest Score Posttest Score 1 29 31 2 33 28 3 36 39 4 37 37 5 40 44 6 42 41 7 45 51 8 47 50 9 50 56 10 56 49 b. What is the correlation between the pretest and posttest rankings of the women on the button-sorting test? Find the correlation and explain what it means.

Power is the likelihood of

The null hypothesis for testing the significance of r_s is that the population correlation coefficient is

For large samples the normal curve is used to give probability values for

In comparing parametric and nonparametric tests, which of the following is(are) the same for both tests?

Which nonparametric test should be used to analyze of the data from the following studies? a. Julia and Davi were interested in the effects of music on reading comprehension in 5th graders. They played pop-rock, easy listening, or no music to different groups while they were reading. Later the students took a test over the material they read. b. Kelly and Aubrey gave students a test of Machiavellianism (manipulativeness of others) and divided them into high and low scorers. Students then learned a motor skills test. High math scorers were compared to low scorers. c. Carla compared the attitudes of African-American and Caucasian students on the Cultural Attitudes Inventory.