Psychology 3180 Assignment
I - In each of the problems that follow, you are to use your computer and SPSS as your tools to analyze and interpret quantitative data.
An industrial psychologist sought to evaluate the effects of a new profit-sharing incentive plan on productivity in her firm, a computer assembly plant. The plant had 10 assembly teams. The number of units produced by each team during the 8 weeks prior to the beginning of the new incentives and for 8 weeks afterwards a e presented below. Bearing in mind the non-experimental nature of the study, test the hypothesis that the new incentive plan had no effect on productivity (the null hypothesis).
|
Team
|
Before
|
After
|
|
1
|
85
|
80
|
|
2
|
79
|
82
|
|
3
|
70
|
82
|
|
4
|
80
|
78
|
|
5
|
81
|
85
|
|
6
|
102
|
100
|
|
7
|
76
|
68
|
|
8
|
70
|
60
|
|
9
|
78
|
71
|
|
10
|
72
|
80
|
1. The correct test for t e example above is
a. an independent samples T test.
b. a one-sample T test.
c. a paired samples test.
2. The degrees of freedom (df) for this problem is
a. 18
b. 10
c. 9
d. 2
3. The mean for units produced "before" is
a. 79.30
b. 78.60
c. 9.37
d. 10.76
4. The mean for units produced "after" is
a. 10.46
b. 10.20
c. 78.60
d. 80.80
5. The dependent variable is
a. level of incentive, i.e. in this case their presence or absence
b. units produced
c. number of assembly teams
d. time, i.e. eight weeks
6. The independent variable is
a. level of incentive i.e. in this case their presence or absence
b. units produced
c. number of assembly teams
d. time, i.e. eight weeks
7. The observed value for t is
a. 2.96
b. 9
c. .013
d. .305
8. The p-value for t is
a. .013
b. .748
c. .700
d. .767
9. Maintaining a Type I (alpha) error rate of .05, this difference in means is significant.
a. True
b. False
10. The null hypothesis was supported (be careful)
a. True
b. False
11. The effect size measure d for this problem is
a. .767
b. .013
c. 2.93
d. .096
12. Should you expect the two sets of scores above to be collelated?
a. Yes
b. No
13. What is the value for Pearson r as an indicator of the association between the two sets of scores above?
a. .00
b. .16
c. .748
d. .305
14. What would you write for the p-value for the Pearson r above?
a. .16
b. .013
c. .000
d. <.001
15. Is this r statistically significant?
a. Yes
b. No
16. For a test statistic to be "statistically significant" requires that the p value be
a. above .05
b. below .05
c. above .01
d. below .01
17. The regression coefficient in linear regression analysis corresponds to, the ____ in non-statistical (i.e., as in normal math courses) descriptions of linear equations.
a. X term
b. Y term
c. slope
d. y-intercept
18. One measure of effect e that can be used for bivariate correlation is
a. r2 or the coefficient of determination
b. Pearson r
c. eta squared
d. the mean on Y
19. In the context of an independent samples T test, an eta square of .24 may be interpreted as
a. 76% of the variance on the dependent variable associated with the independent variable
b. 24% of the variance on the dependent variable associated with the independent variable
c. a very mild effect
d. being of very little statistical significance.
20. In the context of simple linear regression analysis, the independent variable is sometimes referred to as the
a. criterion variable
b. the predictor variable
c. the constant
d. the intercept
II - Use the scatter plot below to answer the following questions.
21. This plot indicates a negative correlation.
a. True
b. False
22. This plot appears to contain an outlier.
a. True
b. False
23. This plot contains clear evidence of a nonlinear relationship.
a. True
b. False
24. Pearson r also alerts us to the presence of nonlinear relationship in the data.
a. True
b. False
25. The scatter plot above indicates very little correlation between the two variables.
a. True
b. False
26. Non-experimental research rarely uses correlation or linear regression.
a. True
b. False
III - A researcher wants to predict college GPA on the basis of high school GPA scores. She collects the following data and performs a regression analysis.
|
High School GPA
|
College GPA
|
|
1.50
|
1.30
|
|
2.20
|
2.00
|
|
1.30
|
1.70
|
|
2.50
|
1.80
|
|
4.00
|
3.50
|
|
3.20
|
3.50
|
|
1.70
|
2.00
|
|
3.00
|
3.21
|
|
3.00
|
2.90
|
|
2.00
|
3.00
|
27. What is the unstandardized regression coefficient?
a. .572
b. .828
c. .786
28. What is the constant?
a. .786
b. .572
c. .828
29. The equation for predicting Y-values on the basis of X is
a. Y = .828X + .002
b. Y = .572X + .786
c. Y = .786X + .572
30. What would be the predicted college GPA for someone with a high school GPA of 3.50?
a. 2.38
b. 2.73
c. 3.32
31. What is the correlation between high school GPA and college GPA?
a. .828
b. .815
c. .685
32. Is this correlation significant?
a. Yes
b. No
33. How much of the variance on Y (college GPA) is explained by X (high school GPA)?
a. 82%
b. 100%
c. 69%
34. Random assignment to groups is typically associated with which kind of T test?
a. paired samples
b. independent samples
c. one-sample
d. Pearson Levene test
35. The coefficient of determination (or effect size estimate) for a Pearson r of .82 is
a. .67
b. .82
c. .24
36. In general, a p value in excess of .05 indicates that
a. the experimenter did not have enough subjects.
b. the results are statistically significant.
c. the results are unreliable and too likely to be due to chance.
d. the experimenter as sloppy in analyzing the data.
37. Type I error is the same as
a. beta error
b. a lack of power
c. alpha error
38. A researcher randomly assigns 10 rats to a control group and 10 to an experimental group. The study has only one dependent variable, but the researcher also collects information on the age and sex of each rat. The SPSS data file will have ________ rows.
a. 10
b. 20
c. 30
d. 40
39. The researcher above will need at least ____ variables in the SPSS data file.
a. 2
b. 6
c. 4
40. A colleague of the above researcher uses an additional experimental group. According to material covered in this course, an acceptable statistical analysis would be
a. ANOVA
b. Chi Square
c. paired samples
d. one-sample T test
IV - Cathy wants to test the hypothesis that drinking coffee containing caffeine improves memory among college students. She obtains a representative sample of 20 students and randomly assigns 10 to a control group and 10 to an experimental group. The control group drinks 1 cup of decaffeinated coffee. The experimental group drinks 1 cup of caffeinated coffee from the same manufacturer. Cathy waits 20 minutes for any drugs to take effect. Participants are read a list of 20 words and must remember as many as possible. She obtains the following data:
Number of words remembered:
|
No caffeine group
|
Low Dose Caffeine group (1 cup of regular coffee)
|
High Dose Caffeine group (2 cups of regular coffee)
|
|
9
|
9
|
12
|
|
8
|
10
|
12
|
|
9
|
15
|
10
|
|
10
|
12
|
14
|
|
5
|
11
|
12
|
|
4
|
18
|
12
|
|
8
|
10
|
12
|
|
9
|
10
|
15
|
|
4
|
13
|
10
|
|
5
|
10
|
12
|
41. The mean number of words remembered for the Low Dose caffeine (1 cup of caffeinated coffee) group is 7.10?
a. True
b. False
42. The two groups (No caffeine and Low Dose caffeine) differ significantly with respect to their variances?
a. True
b. False
43. The correct value for T comparing these two groups is -4.06?
a. True
b. False
44. The p value for the T above is significant?
a. True
b. False
45. Based upon this experiment, drinking 1 cup of coffee containing caffeine does appear to reliably influence memory function?
a. True
b. False
46. Add the data for the additional group and complete a one-way ANOVA comparing the 3 means. The values for test statistic F and the p value associated with it are
a. F = 2.87 p = .074
b. F = 2.00 p < .001
c. F = 15.02 p < .001
47. The ANOVA for this problem
a. Indicates a significant overall difference between the means.
b. does not provide enough information about the omnibus test.
c. indicates no significant overall difference between the means.
48. Using Tukey HSD as a post hoc test
a. there is a significant difference between the no caffeine group and the low dose group.
b. there is a significant difference between the low dose caffeine group and the high dose group.
c. you cannot tell using this test.
49. Using Dunnett C as a post hoc test
a. there is a significant difference between the high dose caffeine group and the low dose group.
b. there is a significant difference between the no caffeine group and the high dose group.
c. you cannot tell using this test.
50. The mean on the dependent variable for the low dose group is
a. 7.10
b. 11.80
c. 12.10