Exam 10: Hypothesis Testing, Effect Size, and and Confidence Intervals: Two-Sample Designs

Data Set 10-3: An experimenter matched by weight four pairs of volunteers. One member of each pair was randomly assigned to one of the two groups. One group fasted for 24 hours and the other for 48 hours. Scores below represent the number of ounces of Yummi (very vanilla) consumed during the first 10 minutes after the fast. Test for a difference at the .05 level. 24 hours 48 hours 26 18 35 24 23 17 30 14 -The effect size index for Data Set 10-3 is

Data Set 10-3: An experimenter matched by weight four pairs of volunteers. One member of each pair was randomly assigned to one of the two groups. One group fasted for 24 hours and the other for 48 hours. Scores below represent the number of ounces of Yummi (very vanilla) consumed during the first 10 minutes after the fast. Test for a difference at the .05 level. 24 hours 48 hours 26 18 35 24 23 17 30 14 -The standard error of the difference for Data Set 10-3 is

"Achievement test scores are declining all around us,"brooded Professor Probity. "Not in my class,"vouchsafed Professor Paragon. "Here are my final-exam scores for last year and this year on the same exam. Run your own t test on them. Calculate the effect size index."What did Probity discover? Last Year This Year 29 27 27 28 26 25 22 24 19 22 15 20 14 18 13 16 10 12

A social psychologist wanted to determine whether attitudes of men toward abortion were different in rural and urban areas. He prepared a questionnaire and administered it to a group of city dwellers and a group of country dwellers. Each city man was matched with a country man on age, income, and education. High scores indicate positive attitudes toward abortion. Here are the questionnaire scores. Rural Men Urban Men 22 25 19 23 18 20 17 18 15 12 10 9 6 4 4 3 a. What can the social psychologist say about differences in the attitudes of rural and urban men? Why? b. Justify the statistical method you used to analyze the data.

In which of the following would an increase from a smaller number to a larger number produce a more powerful statistical test?

In a paired-samples design, the Apple mean was 6 and the Orange mean was 7. An $\alpha$ level of .05 was adopted; the data produced a t value of 2.15. Which conclusion is appropriate for a two-tailed test?

List the two characteristics of the dependent variable that will assure you that an independent-samples t test will produce accurate probabilities.

For a two-sample t test, an effect size that qualifies as small is

If a 95 percent confidence interval about a difference between means is -2.31 to 24.62, the null hypothesis may be

Cowboys from the mythical Stats Bar-X Ranch competed in a singing contest at the Dry Gulch Saloon where they sang against cowpokes from the Dragging y Ranch. The songs were psychowestern (which are like psychedelic songs, but with a twangy ranch dressing). A Stats Bar?X researcher counted the number of people clapping after each song over the hour of the contest, during which the two groups alternated singing. a. Compare the two groups with a t test, find the effect size index, and write an interpretation of what you find. b. Find the 95 percent confidence interval about the mean difference and write an interpretation. Stats Bar-X Dragging y 12:00 set 21 23 12:10 set 25 18 12:20 set 20 16 12:30 set 17 14 12:40 set 14 10 12:50 set 8 3

The null hypothesis is

Data Set 10-6: Drugs used in treating schizophrenia all block the reception of dopamine by neurons. (Dopamine is a neurotransmitter, which, when released by the axons of one nerve, inhibits the firing of the next nerve.) This fact led to the idea that schizophrenia occurs when too much dopamine is produced. Suppose the following data on dopamine production were obtained. Schizophrenics Control Group 42 33 31 27 29 18 -The conclusion from Data Set 10-6 (which is the same arrived at by some of the actual research) is that schizophrenics

1 - $\beta$ is the of a statistical test.

For a two-sample t test, which of the following would be considered a small effect size?

For a two-sample t test, an effect size that qualifies as small is

p < .05 means that the difference between sample means

Both $\alpha$ and N help determine the amount of power a t test has.

72 An experimenter found sample means of 12 and 16. A test on 9 df gave a t value of 5.61. The design

The formula for power is 1 - $\mu$ .

Indicate whether each design below is independent samples or paired samples and where possible, give the degrees of freedom. -From a large General Psychology course the instructor obtained the SAT scores of all the first-born and all the second-born students. The mean SAT score of the two groups was compared.