Some scenarios adapted from Gravetter, F.J. & Wallnau, L.B. (1996). Statistics for the behavioral sciences (4th ed.). St. Paul, MN: West Publishing.

Scenarios:

1. In a classic study examining the effect of environment on development, Kretch et al. (1962) divided a sample of infant rats into two groups. One group was housed in a stimulus-rich environment (ladders, tunnels, etc.) while the second group was housed in a stimulus-poor environment. Later, the rats’ maze-learning ability was tested to see if there is an effect of environment on maze-learning ability. Below are hypothetical data from this study (dependent variable = number of errors committed in attempting to reach the end of the maze). You found an obtained statistic of 2.21.

Rich rats: 17 10 25 39 38 41 29 29 42 (SS = 214, M = 30) Poor rats: 37 27 26 30 35 43 40 36 28 38 (SS = 321, M = 34) Use the 0.01 level of significance with one tail. A. Test and why it’s used. B. Concluding paragraph with APA statistical report and explanation. C. Graph D. Questions. 1. What is the independent variable? 2. Would the critical statistic be larger, smaller, or the same if larger samples were used? 3. What is one way you could redo the study with more Power?

2. Craik and Tulving (1975) gave subjects a list of words to remember. Each word was presented in the context of a sentence. Subjects in the experiment saw two different types of sentences: simple sentences and complex sentences. Below are hypothetical results from their experiment (dependent variable = number of words recalled at a later memory test). Note: obtained statistic = 1.62

subject simple sentences complex sentences

A 14 22 B 15 17 C 19 23 D 13 19 E 17 28 F 15 20

Use the 0.05 level of significance with one tail. Don’t create a graph for this one. A. Test and why it’s used. B. Concluding paragraph with APA statistical report and explanation. C. Graph D. Questions. 1. What type of statistical test would be used if, rather than measuring the number of words recalled under those two conditions, the experimenters measured two things: how many simple sentences participants could write in a 10-minute period, and how many complex sentences (out of 30) participants correctly comprehended? 2. What kind of error might you have made, and what is the probability you made it? (don’t need to explain the second part)
3. Several studies indicate that handedness is related to differences in brain function. Since different parts of the brain are specialized for specific behaviors, this means that left- and right-handed people should show different talents. To test this hypothesis, a psychologist tested pitch performance (part of music ability) for three groups of subjects: left-handed, right-handed, and ambidextrous. The data from that study are as follows:

left-handed right-handed ambidextrous

6 1 2 4 0 0 3 1 0 4 0 2 3 2 1

Use the 0.05 level of significance. The obtained statistic is 5.06. A. Test and why it’s used. B. Concluding paragraph with APA statistical report and explanation. C. Graph D. Questions. 1. What kind of statistical test would be used (be specific) if we did not look at an ambidextrous group? 2. What would you expect to happen to the obtained statistic if only music majors participated in the study? (assume the above involved students of all majors)
5. Is the amount a person reads associated with the quality of their own writing? To test this, a researcher (using alpha 0.01) asks participants how many books they read per year, and has them write a 2-page paper (which she then has someone read and assign a grade of 0 to 10 to). Here is what she found:

Participant Books/yr Writing score A 18 6 B 24 7 C 12 5 D 5 3 E 10 4 F 32 7 G 19 5

Note: r = + 0.93 A. Test and why it’s used. B. Concluding paragraph with APA statistical report and explanation. C. Graph D. Questions.

1. What would you expect to happen to the correlation if you measured number of errors made in the writing sample rather than writing score? 2. If your finding above was significant, what do you have the option to now do?