1. What would be the appropriate statistical procedure to test the following hypothesis: "Determine the relationship between the year in school (freshman, sophomore, junior and senior) and having a car on campus."
2. What is (are) the function(s) of parametric statistical procedures?
3. What is Type 1 error?
4. What are the assumptions underlying the use of parametric statistical procedures?
5. If a critical value is greater than the test statistic, would you accept or reject the null hypothesis?
6. Under what circumstance(s) is it appropriate to use a 2-tailed test of significance?
7. What is the appropriate statistical procedure to use when you interest is in detecting a bivariate, curvilinear association?
8. For a study comparing outcomes under alternative treatment conditions, when the null hypothesis is rejected, the researcher concludes that a difference among groups exists.
True
False
9. A researcher, for reasons passing understanding, wishes to assess the association between gender and total cholesterol values. What would be the appropriate statistical procedure?
10. An HIV educator wishes to determine whether the method of delivering teaching influences adherence to antiretroviral therapy. She decided to measure adherence as viral load (a ration measure). She teaches on group using lecture-discussion techniques. She adapts the information for access on the internet and give another group this information using this medium. For yet another group, she decided to give a CD Rom for home study and then meets with individuals to answer any questions. She obtains viral loads for all clients for comparison. What procedure will determine the significance of any differences?
Items 11-15 related to the following study results:
Study A Study B Study C
X2 F=7.357 r=.83
df=4 df=4/203 df=98
p> .05 p < .05 p < .01
11. What statistical procedure was used to analyze data in Study C?
12. How many groups were compared in study B?
13. How many subjects were enrolled in Study C?
14. Which demonstrated the greatest level of statistical significance?
15. In which study is the likelihood of Type I error greatest?
Items 16, 17, and 18 relate to the following:
In a regression analysis, a nurse researcher found a correlation of ,80 between pain relief scores and satisfaction with nursing care. She also calculated the following for her regression analysis:
Pain relief (x): Mean = 58 sd = 3.9
Satisfaction (y): Mean =42 sd = 5.4
Slope = 1.56
Y intercept = -3.53
16. What will be the predicted satisfaction scores (expressed as a point estimate) for a patient with pain relief score of 60?
17. What is the standard error of estimate when predicting satisfaction from knowledge of pain relief score
18. What would be the interval estimate for satisfaction for the patient in problem 16?
Items 19-21: A nurse researcher is investigating the effect of timing of standard pain control interventions on severity of pain in adolescents with sickle-cell disease. She establishes three treatment protocols: 1) initiation of pain control immediately upon the presence of prodromal sign (an "aura" signaling the imminent onset of pain); 2) initiation of pain control one hour after the onset of pain; and 3) initiation of pain control only at the points where non-steroidal anti-inflammatories and guided imagery are no longer effective in keeping pain bearable. She conducted a one-way ANOVA to analyze her data and the following table summarizes her findings:
Source df SS MSS F p___
Among 2 75536.2 37768.1 5.159 < .05
Within 27 197660.3 7320.8______________________
Total 29 273196.5
On the basis of these data alone, she drew the following conclusions. For each conclusion, indicate whether you feel the conclusion is justified or unjustified.
19. Severity of pain is influenced by the timing of pain interventions in sickle cell crises.
20. Immediate intervention is better than either slightly delayed intervention or initiation at crisis stage.
21. She has more than 90% confidence in her conclusion that severity of pain is influenced by timing.
Items 22-23 relate to the following study results:
Study A Study B
r = .64 r=.77
df = 18 df = 21
p< .05 p< .01
22. In using the data from Study A to make predictions, what percent of time would you expect predictions to be exactly correct?
23. Which study would have the smallest margin of error in predicting one variable from knowledge of the other?
Items 24 and 25 relate to the following SPSS output. A researcher, interested in characteristics of HIV+ and HIV- adolescents, interviewed 166 young adults about their experiences during adolescence. He wished to know, among other things, if there were significant difference in the ages at which HIV+ and HIV- young adults became sexually active. The following is the printout of this analysis:
HIV Status N Mean sd Stnd. Error
Age at Positive 57 13.2 2.96576 .39282
first sexual
experience Negative 109 15.1 2.57286 .24644
______________________________________________________________________________
Independent Samples Test
______________________________________________________________________________
Levene's Test for
Equality of Variances
__________________________________________F Sig
Age at Equal variance assumed 1.313 .254
first sexual
experience Equal variance not assumed
t test for Equality of Means
t df sig. mean difference
______________________________________________________________________________
Age at Equal variance assumed -2.870 164 .005 -1.9
first sexual Equal variance not assumed -2.745 99.66 .007 -1.9
experience ____________________________________________________________
24. Were there significant differences between the groups? Give the relevant statistical data to support your answer.
25. What is the confidence interval associated with your answer to Questions #24?
Each of the questions on the following pages is a calculation problem worth 10points. Partial credit will be awarded if I can follow your procedures and determine that errors are arithmetic rather than conceptual. If would be wise, therefore, to clearly indicate your worksteps on each problem.
26. For the following data set, calculate the one-way ANOVA and test for significance at the .05 level.
|
Group 1
|
Group 2
|
Group 3
|
Group 4
|
|
x
|
x2
|
x
|
x2
|
x
|
x2
|
x
|
x2
|
|
6
|
36
|
10
|
100
|
12
|
144
|
16
|
256
|
|
8
|
64
|
12
|
144
|
14
|
196
|
15
|
225
|
|
7
|
49
|
13
|
169
|
13
|
169
|
18
|
324
|
|
9
|
81
|
13
|
169
|
11
|
121
|
17
|
289
|
|
8
|
64
|
14
|
196
|
13
|
169
|
20
|
400
|
|
11
|
121
|
15
|
225
|
15
|
225
|
21
|
441
|
|
9
|
81
|
14
|
196
|
13
|
169
|
22
|
484
|
|
58
|
496
|
91
|
1,199
|
91
|
1,193
|
129
|
2,419
|
27. Calculate the x2for the following 3 x 2 table and test for significance at the .01 level.
Group 1 Group 2 Group 3
|
Positive outcome
|
9
|
12
|
9
|
30
|
|
Negative outcome
|
5
|
16
|
5
|
26
|
14 28 1456
28. For the following group data, calculate a t-test and test for significance at the .05 level, 2-tailed level of significance.
Treatment Group Control Group
Mean = 70.4 Mean = 51.2
sd = 5.6 sd = 6.6
n=42 n=48
29. For the following paired observations, calculate the Pearson product-moment correlation coefficient and test for significance at the .01 level.
|
X
|
x2
|
y
|
y2
|
xy
|
|
16
|
256
|
20
|
400
|
320
|
|
17
|
289
|
19
|
361
|
323
|
|
18
|
324
|
20
|
400
|
360
|
|
19
|
361
|
17
|
289
|
323
|
|
21
|
441
|
15
|
225
|
315
|
|
22
|
484
|
19
|
361
|
418
|
|
21
|
441
|
20
|
400
|
420
|
|
23
|
529
|
19
|
361
|
437
|
|
22
|
484
|
20
|
400
|
440
|
|
19
|
361
|
17
|
289
|
323
|
|
198
|
3,970
|
186
|
3,486
|
3,679
|
30. For the following data regarding paired rank orders for a sample, calculate the correlation coefficient and test for significance at the .05 level.
Subject # Rank 1 Rank 2
1 1.5 1
2 1.5 2.5
3 3 2.5
4 5 4
5 6 5
6 4 8
7 7 6
8 9.5 7
9 8 10
10 9.5 9