Assignment 2 Outliers and Normality
Assignment 2 Outliers and Normality
INDIVIDUAL ASSIGNMENT 2
TITLE
PREPARED BY
Cases
Descriptives
Median 23.00
Variance 45.827
Minimum 5
Maximum 35
Range 30
Interquartile Range 9
Extreme Values
2 183 35
3 195 35
4 212 35
5 227 35a
Lowest 1 382 5
2 344 5
3 61 5
4 7 5
5 2 5
a. Only a partial list of cases with the value 35 are shown in the table of
upper extremes.
1
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
2
3
Results
There are no any extreme values. The outlier’s score is genuine.
4
1.2 Normality Test - Tlifesat
Median 23.00
Variance 42.570
Minimum 5
Maximum 35
Range 30
Interquartile Range 9
Skewness -.293 .179
Median 23.00
Variance 47.762
Minimum 5
Maximum 35
Range 30
Interquartile Range 10
Males
Skewness z-values = -0.293/ 0.179 = -1.64 (between -1.96 and +1.96)
Kurtosis z-values = -0.283/ 0.436 = -0.65 (between -1.96 and +1.96)
Females
Skewness z-values = -0.374/ 0.154 = -2.43 (below -1.96)
Kurtosis z-values = -0.519/ 0.306 = -1.70 (between -1.96 and +1.96)
All the z-values between -1.96 and +1.96 except skewness z-values for females which is
below -1.96.
5
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
total life satisfaction MALES .094 185 .000 .984 185 .035
The null hypothesis for this test of normality, is that the data are normally distributed.
The hypothesis is rejected if the p-value is below 0.05.
Both p-values are below 0.05. We reject the null hypothesis. In terms of the Shapiro-
Wilk test, we can assume that our data are not normally distributed.
6
7
8
9
Results
Sample Characteristics
A Shapiro-Wilk’s test (p<0.05) (Shapiro & Wilk, 1965; Razali & Wah, 2011) and a
visual inspection of their histograms, normal Q-Q plots and box plots showed that the total
life satisfaction scores were not normally distributed for both males and females, with a
skewness of -0.293 (SE = 0.179) and a kurtosis of -0.283 (SE = 0.436) for the males and a
skewness of -0.374 (SE = 0.154) and a kurtosis of -0.519 (SE = 0.306) for the females
(Cramer & Howitt, 2004; Doane & Seward, 2011).
Cases
Descriptives
Median 26.00
Variance 34.194
Minimum 12
Maximum 46
Range 34
Interquartile Range 8
10
Extreme Values
2 262 157 44
3 216 61 43
4 190 6 42
2 189 5 12
3 247 127 13
4 244 119 13
5 98 301 13
a. Only a partial list of cases with the value 42 are shown in the table of upper extremes.
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
11
12
Results
There are no extreme values, but there are two outliers: ID numbers 24 and 157.
13
2.2 Checking for Outliers – Tpstress with outlier removed
Cases
Descriptives
Median 26.00
Variance 32.788
Minimum 12
Maximum 43
Range 31
Interquartile Range 8
Extreme Values
2 190 6 42
3 257 144 42
4 339 330 42
5 228 85 41a
2 189 5 12
3 247 127 13
4 244 119 13
5 98 301 13
a. Only a partial list of cases with the value 41 are shown in the table of upper extremes.
14
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
15
16
Results
The outlier’s score is genuine.
17
1.2 Normality Test - Tpstress
Case Processing Summary
Cases
Descriptives
Median 25.00
Variance 27.220
Minimum 13
Maximum 39
Range 26
Interquartile Range 8
Median 27.00
Variance 35.825
Minimum 12
Maximum 43
Range 31
Interquartile Range 7
18
Females
Skewness z-values = -0.128/ 0.155 = -0.83 (between -1.96 and +1.96)
Kurtosis z-values = -0.018/ 0.308 = -0.06 (between -1.96 and +1.96)
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
total perceived stress MALES .071 183 .026 .990 183 .245
The null hypothesis for this test of normality, is that the data are normally distributed.
The hypothesis is rejected if the p-value is below 0.05.
Both p-values are above 0.05. We accept the null hypothesis. In terms of the Shapiro-
Wilk test, we can assume that our data are approximately normally distributed.
19
20
21
22
Results
Sample Characteristics
A Shapiro-Wilk’s test (p>0.05) (Shapiro & Wilk, 1965; Razali & Wah, 2011) and a
visual inspection of their histograms, normal Q-Q plots and box plots showed that the total
life satisfaction scores were normally distributed for both males and females, with a skewness
of -0.046 (SE = 0.180) and a kurtosis of -0.312 (SE = 0.357) for the males and a skewness of
-0.128 (SE = 0.155) and a kurtosis of -0.018 (SE = 0.308) for the females (Cramer & Howitt,
2004; Doane & Seward, 2011).
23