CALIFORNIA STATE UNIVERSITY, BAKERSFIELD
SCHOOL OF BUSINESS AND PUBLIC ADMINISTRATION
Department of Public Policy and Administration
PPA 415 – Research Methods in Public Administration
Exercise 6
1. Do problem 9.4 in Healey (p. 222). Do calculations by hand.
Step 1: The problem involves independent random samples, interval-ratio measurement of change in crime rates, normally distributed populations, and equal population variances.
Step 2: H0: µ1 = µ2 = µ3
H1: at least one strategy has a significantly different outcome.
Step 3: Using the F distribution, an alpha of .05, degrees of freedom within of 15, and degrees of freedom between of 2, F(critical) is 3.68.
Step 4:
Neighborhood Watch / NW2 / Foot Patrol / FP2 / No Program / NP2-10 / 100 / -21 / 441 / 30 / 900
-20 / 400 / -15 / 225 / -10 / 100
10 / 100 / -80 / 6400 / 14 / 196
20 / 400 / -10 / 100 / 80 / 6400
70 / 4900 / -50 / 2500 / 50 / 2500
10 / 100 / -10 / 100 / -20 / 400
80 / 6000 / -186 / 9766 / 144 / 10496 / 26262 / 2.11
13.33333333 / -31 / 24
SST=ΣX2-N(Xmean)2= / 26181.78
SSB=ΣNk(Xmean k-Xmean)2= / 10208.44
SSW=SST-SSB= / 15973.33
Dfw=N-k= / 15
Dfb=k-1= / 2
Mean square between / 5104.22
Mean square within / 1064.89
F= / 4.79
Step 5: F(obtained) = 4.79 > F(critical) = 3.68. Therefore, we can reject the null hypothesis that the crime-reduction programs have the same effect on crime. Only one of the three programs shows a net reduction in crime over a one-year period, foot patrols. With 95 percent confidence, I believe that foot patrols have had the greatest effect in reducing crime in Shinbone, Kansas.
2. Do problem 9.7 in Healey (p. 223). Do problem in SPSS.
Step 1: The problem involves independent random samples, interval-ratio measurement of change in knowledge of the news, normally distributed populations, and equal population variances.
Step 2: H0: µ1 = µ2 = µ3 = µ4
H1: Older people lose interest in politics and public affairs..
Step 3: Using the F distribution, an alpha of .05, degrees of freedom within of 44, and degrees of freedom between of 3, F(critical) is 2.816465827.
Step 4:
ANOVA
Correct Answers on Headline Quiz
Sum of Squares / df / Mean Square / F / Sig.Between Groups / 124.063 / 3 / 41.354 / 5.958 / .002
Within Groups / 305.417 / 44 / 6.941
Total / 429.479 / 47
Step 5: F(obtained) = 5.96 > F(critical) = 2.82. Therefore, we can reject the null hypothesis that the age has no effect on interest in politics and current affairs. However, the results do not support the research hypothesis that older people lose interest. The mean scores on the recall of newspaper stories actually increase by age, reaching their highest level among the older population.
3. The second round of the CSUB Policy Delphi asked faculty, administrators, students, staff, and community members to rank ten learning outcomes (critical speaking, critical reading, ethical framework, working independently, critical writing, technology applications to problem solving, application of discipline to real-world, critical thinking, diversity and cultural understanding, basic understanding of a discipline [Speaking to Undertanding]) from first to tenth most important. The most important was given a score of ten and the least important a score of one. Does the mean ranking of each learning outcome vary significantly by the respondent’s relationship to the university? Use the .10 level of significance (alpha). HINT: Use the recoded relationship variable.
Step 1: The problem involves independent random samples, interval-ratio measurement of the importance rankings, normally distributed populations, and equal population variances.
Step 2: H0: µ1 = µ2 = µ3
H1: at least one type of respondent rates the importance of the learning outcome differently.
Step 3: Using the F distribution, an alpha of .05, degrees of freedom between of 2, and degrees of freedom within for each variable listed below, F(critical) equals the following values:
Critical thinking. 238 2.33
Critical reading. 235 2.33
Application of discipline to real-world. 234 2.33
Critical writing. 232 2.33
Ethical framework. 230 2.33
Critical speaking. 234 2.33
Basic understanding of a discipline. 237 2.33
Work Independently. 229 2.33
Diversity and cultural understanding. 234 2.33
Technology applications to problem solving. 234 2.33
Step 4:
Recoded Relationship
Learning Outcome / Faculty, staff, or administrator / Student / Alumni or community member / Total
Importance (10=highest, 1=lowest) / Mean / Mean / Mean / Mean
Critical thinking. / 8.76 / 8.23 / 8.22 / 8.36
Critical reading. / 6.81 / 6.18 / 6.57 / 6.43
Application of discipline to real-world. / 5.43 / 6.52 / 6.41 / 6.22
Critical writing. / 6.25 / 6.18 / 6.17 / 6.20
Ethical framework. / 6.38 / 5.32 / 5.89 / 5.73
Critical speaking. / 4.88 / 5.42 / 5.95 / 5.42
Basic understanding of a discipline. / 5.00 / 5.15 / 4.98 / 5.07
Work Independently / 4.42 / 5.01 / 4.27 / 4.67
Diversity and cultural understanding. / 3.71 / 3.97 / 3.00 / 3.66
Technology applications to problem solving. / 3.36 / 3.64 / 3.97 / 3.66
Source: CSUB Policy Delphi, Second Round, May 2006
Step 5: Making a Decision
F(obtained) > F(critical) for four of the ten learning outcomes: critical speaking, ethical framework, application of a discipline to the real world, and diversity and cultural understanding. Alumni, community members, and students rated critical speaking and the application of a discipline to the real world more highly than did faculty, staff, and administrators. By contrast, faculty, staff, and administrators rated an ethical framework higher than did students, alumni, and community members. Faculty, staff, administrators, and students rated diversity more important than did alumni and community members.
4. Several authors have suggested that some regions of the country are more likely to receive disaster declarations than other regions of the country. Does the probability of a major disaster declaration (if SBA declarations are treated as turndowns [ActionType2]) vary significantly by FEMA region?
Step 1: The problem involves independent random samples, interval-ratio measurement of the proportion of major disasters granted, normally distributed populations, and equal population variances.
Step 2: H0: µ1 = µ2 = µ3 = µ4 = µ5 = µ6 = µ7 = µ8 = µ9 = µ10
H1: at least one region has a significantly different probability of receiving a declaration.
Step 3: Using the F distribution, an alpha of .05, degrees of freedom between of 9, and degrees of freedom within of 1398, F(critical) equals 1.886563971.
Step 4: Calculating the test statistic.
Proportion of Major Disasters by FEMA RegionFEMA Region / Mean / N / Std. Deviation
1 (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont) / 0.80 / 79 / 0.37
10 (Alaska, Idaho, Oregon and Washington) / 0.75 / 87 / 0.43
3 (Delaware, District of Columbia, Maryland, Pennsylvania, Virginia and W. Virginia) / 0.74 / 102 / 0.42
9 (Arizona, California, Hawaii, Nevada, American Samoa, Guam, Northern Mariana Islands, Marshall Islands, Micronesia) / 0.71 / 138 / 0.44
6 (Arkansas, Louisiana, New Mexico, Oklahoma and Texas) / 0.69 / 244 / 0.46
2 (New Jersey, New York, Puerto Rico, and the Virgin Islands) / 0.69 / 70 / 0.43
5 (Illinois, Indiana, Michigan, Minnesota, Ohio and Wisconsin) / 0.66 / 191 / 0.46
8 (Colorado, Montana, N. Dakota, S. Dakota, Utah and Wyoming) / 0.65 / 83 / 0.47
7 (Iowa, Kansas, Missouri and Nebraska) / 0.65 / 127 / 0.47
4 (Alabama, Florida, Georgia, Kentucky, Mississippi, N. Carolina, S. Carolina and Tennessee) / 0.59 / 287 / 0.48
Total / 0.67 / 1408 / 0.45
Step 5: Making a decision.
F(obtained)=2.659 > F(critical)=1.887. Therefore, I can reject the null hypothesis that major disaster approval rates do not vary by region. FEMA regions do show significant variation in approval rates. The regions with the highest rates of approval (75 percent and over) are FEMA Regions 1 (New England) and 10 (Pacific Northwest). The FEMA Region with the lowest rate of approval (below 60 percent) is FEMA Region 4 (Deep South).
5. Using the Leadership Skills data set, determine if the type of organization in which a student works affects the way in which themselves and their supervisors on technical, human, and conceptual skills (use the full scales, not the ranked ones).
Step 1: The problem involves independent random samples, interval-ratio measurement of the technical, human, and conceptual skills rankings, normally distributed populations, and equal population variances.
Step 2: H0: µ1 = µ2 = µ3
H1: students from at least type of organization rate the skills levels differently from students of the other types of organization.
Step 3: Using the F distribution, an alpha of .05, degrees of freedom between of 2, and degrees of freedom within of 60, F(critical) equals 3.150411311.
Step 4: Calculating the test statistic.
Type of OrganizationPublic / Private / Nonprofit
Mean / Count / Mean / Count / Mean / Count
Technical Skills Scale / 26.07 / 25 / 26.19 / 32 / 25.33 / 6
Human Skills Scale / 24.69 / 25 / 24.94 / 32 / 26.17 / 6
Conceptual Skills Scale / 23.31 / 25 / 24.28 / 32 / 25.67 / 6
ANOVA
Sum of Squares / df / Mean Square / F / Sig.Technical Skills Scale / Between Groups / 3.694 / 2 / 1.847 / .144 / .867
Within Groups / 771.913 / 60 / 12.865
Total / 775.607 / 62
Human Skills Scale / Between Groups / 10.516 / 2 / 5.258 / .361 / .698
Within Groups / 873.220 / 60 / 14.554
Total / 883.736 / 62
Conceptual Skills Scale / Between Groups / 31.098 / 2 / 15.549 / 1.038 / .360
Within Groups / 898.679 / 60 / 14.978
Total / 929.777 / 62
Step 5: Making a Decision:
F(obtained) < F(critical) for all three leadership skill scales (technical, human, and conceptual). Therefore, I cannot reject the null hypothesis that students working in public, private, and nonprofit organizations do not assess their own and their supervisor’s leadership skills differently. Type of organization does not influence assessment of various types of leadership skills.
6