Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Client Age
Indicates number of subjects in the analysis (120 females and 301 males)
Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Difference Lower Upper .126 .328 3.973 3.772
F Client Age Equal variances assumed Equal variances not assumed 14.534
Sig. .000
t 2.094 2.343
df 419 283.332
Indicates the probability of error (p) Since the probability of error is < .05, the analyst must reject the null hypothesis of no difference and conclude that there is a significant difference between the mean ages of males and females.
N 8th Grade or Below 9th Grade Up (No Graduation) Graduated High School G.E.D. Certificate Some College Bachelor's Degree Total 28 146 112 47 79 11 423
Minimum 0 0 0 0 0 0 0
Maximum 12 6 5 6 9 5 12
Indicates number of subjects in the analysis (e.g. 112 subjects graduated high school)
Indicates individual group means (e.g. subjects possessing a G.E.D. had, on average, 2.45 prior convictions)
ANOVA Total Prior Convictions Sum of Squares 52.406 1105.084 1157.489 df 5 417 422 Mean Square 10.481 2.650 F 3.955 Sig. .002
Since the probability of error is < .05, the analyst must reject the null hypothesis of no difference and conclude that there are significant differences in mean prior convictions among education categories.
Multiple Comparisons Dependent Variable: Total Prior Convictions Tukey HSD Mean Difference (I-J) .448 1.027* .446 .994 1.529 -.448 .579 -.002 .546 1.082 -1.027* -.579 -.581 -.033 .502 -.446 .002 .581 .548 1.083 -.994 -.546 .033 -.548 .535 -1.529 -1.082 -.502 -1.083 -.535
G.E.D. Certificate
Some College
Bachelor's Degree
(J) Highest Year of Education Completed 9th Grade Up (No Graduation) Graduated High School G.E.D. Certificate Some College Bachelor's Degree 8th Grade or Below Graduated High School G.E.D. Certificate Some College Bachelor's Degree 8th Grade or Below 9th Grade Up (No Graduation) G.E.D. Certificate Some College Bachelor's Degree 8th Grade or Below 9th Grade Up (No Graduation) Graduated High School Some College Bachelor's Degree 8th Grade or Below 9th Grade Up (No Graduation) Graduated High School G.E.D. Certificate Bachelor's Degree 8th Grade or Below 9th Grade Up (No Graduation) Graduated High School G.E.D. Certificate Some College
Std. Error .336 .344 .389 .358 .579 .336 .204 .273 .227 .509 .344 .204 .283 .239 .514 .389 .273 .283 .300 .545 .358 .227 .239 .300 .524 .579 .509 .514 .545 .524
Sig. .767 .035 .861 .063 .090 .767 .054 1.000 .157 .276 .035 .054 .315 1.000 .925 .861 1.000 .315 .449 .352 .063 .157 1.000 .449 .911 .090 .276 .925 .352 .911
95% Confidence Interval Lower Bound Upper Bound -.51 .04 -.67 -.03 -.13 -1.41 -.01 -.78 -.10 -.38 -2.01 -1.16 -1.39 -.72 -.97 -1.56 -.78 -.23 -.31 -.48 -2.02 -1.20 -.65 -1.41 -.96 -3.19 -2.54 -1.98 -2.64 -2.04 1.41 2.01 1.56 2.02 3.19 .51 1.16 .78 1.20 2.54 -.04 .01 .23 .65 1.98 .67 .78 1.39 1.41 2.64 .03 .10 .72 .31 2.04 .13 .38 .97 .48 .96
Indicates a significant difference in prior convictions between the education categories 8th grade education or below and Graduated High School
Specifically, since there were significant differences ONLY between 8th Grade or Below and Graduated High Schoolthe analyst must modify the hypothesis analysis and conclude that significant differences in mean number of prior convictions exist between those subjects with an 8th grade education or below and subjects who graduated from high school.
Interpreting an
Correlations Total Prior Convictions 1 Total Violations .252** .000 427 406 .252** 1 .000 406 406
Total Violations
Probability of error (p) Since the probability of error is < .05, the analyst must reject the null hypothesis of no directional relationship (or correlation) between total prior convictions and total (program) violations and conclude that there is a significant correlation between total prior convictions and total (program) violations. Since there IS a significant directional relationship (correlation) between the two variables, the analyst must examine r (the correlation coefficient) for strength and direction. Direction is determined by ascertaining the sign of r (positive or negative). Strength is determined by the absolute value of r and its approximate distance to 1. In this case, r is positive indicating a POSITIVE correlation between total prior convictions and total (program) violations. This indicates that subjects with greater numbers of prior convictions will likely violate program guidelines at a higher rate. However, the absolute value of r is not very close to 1, indicating a somewhat weak correlation. This could lead the analyst to conclude that
ALTHOUGH THERE IS A SIGNIFICANT CORRELATION BETWEEN TOTAL PRIOR CONVICTIONS AND TOTAL (PROGRAM) VIOLATIONS, THE CORRELATION IS RELATIVELY WEAK.
Client Race * Client Sex Crosstabulation Client Sex Female Male 78 151 34.1% 65.9% 65.0% 50.5% 18.6% 36.0% 40 147 21.4% 78.6% 33.3% 49.2% 9.5% 35.1% 0 1 .0% 100.0% .0% .3% .0% .2% 1 0 100.0% .0% .8% .0% .2% .0% 1 0 100.0% .0% .8% .0% .2% .0% 120 299 28.6% 71.4% 100.0% 100.0% 28.6% 71.4%
Client Race
Caucasian
African-American
Hispanic
Native American
Other
Total
Count % within Client Race % within Client Sex % of Total Count % within Client Race % within Client Sex % of Total Count % within Client Race % within Client Sex % of Total Count % within Client Race % within Client Sex % of Total Count % within Client Race % within Client Sex % of Total Count % within Client Race % within Client Sex % of Total
If you get confused as to which percentage represents what, use RAW numbers to calculate the desired percentage: What percentage of CAUCASIANS are MALE? # of Male Caucasians = 151 # of Caucasians = 229 151 / 229 = .6593 OR 65.9%
Chi-Square Tests Value 13.487a 13.956 2.094 419 df 4 4 1 Asymp. Sig. (2-sided) .009 .007 .148
a. 6 cells (60.0%) have expected count less than 5. The minimum expected count is .29.
Since the probability of error is < .05, the analyst must reject the null hypothesis of no relationship and conclude that there is a significant relationship between race and gender.