The SAS System

  Year
1968 1969 1970 1971 1972
Mortality Std 0.99 1.40 0.78 1.30 0.95
Mean 5.99 5.77 4.91 4.80 4.72



The SAS System

The GLM Procedure

Class Level Information
Class Levels Values
Year 5 1968 1969 1970 1971 1972

Number of Observations Read 50
Number of Observations Used 50



The SAS System

The GLM Procedure
 
Dependent Variable: Mortality

Source DF Sum of Squares Mean Square F Value Pr > F
Model 4 14.16280000 3.54070000 2.88 0.0331
Error 45 55.31500000 1.22922222    
Corrected Total 49 69.47780000      

R-Square Coeff Var Root MSE Mortality Mean
0.203846 21.16653 1.108703 5.238000

Source DF Type I SS Mean Square F Value Pr > F
Year 4 14.16280000 3.54070000 2.88 0.0331

Source DF Type III SS Mean Square F Value Pr > F
Year 4 14.16280000 3.54070000 2.88 0.0331



The SAS System

The GLM Procedure
 
t Tests (LSD) for Mortality


Note: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.

Alpha 0.05
Error Degrees of Freedom 45
Error Mean Square 1.229222
Critical Value of t 2.01410
Least Significant Difference 0.9986

Means with the same letter
are not significantly different.
t Grouping Mean N Year
  A 5.9900 10 1968
  A      
B A 5.7700 10 1969
B        
B C 4.9100 10 1970
B C      
B C 4.8000 10 1971
  C      
  C 4.7200 10 1972



The SAS System

The GLM Procedure
 
Duncan's Multiple Range Test for Mortality


Note: This test controls the Type I comparisonwise error rate, not the experimentwise error rate.

Alpha 0.05
Error Degrees of Freedom 45
Error Mean Square 1.229222

Number of Means 2 3 4 5
Critical Range 0.999 1.050 1.084 1.108

Means with the same letter
are not significantly different.
Duncan Grouping Mean N Year
  A 5.9900 10 1968
  A      
B A 5.7700 10 1969
B        
B   4.9100 10 1970
B        
B   4.8000 10 1971
B        
B   4.7200 10 1972



The SAS System

The GLM Procedure
 

Bartlett's Test for Homogeneity
of Mortality Variance
Source DF Chi-Square Pr > ChiSq
Year 4 3.8867 0.4216



The SAS System

The GLM Procedure
 

Level of
Year
N Mortality
Mean Std Dev
1968 10 5.99000000 0.98933199
1969 10 5.77000000 1.39845152
1970 10 4.91000000 0.77952977
1971 10 4.80000000 1.30384048
1972 10 4.72000000 0.95078915



The SAS System

The GLM Procedure
 

Levene's Test for Homogeneity of Mortality Variance
ANOVA of Squared Deviations from Group Means
Source DF Sum of Squares Mean Square F Value Pr > F
Year 4 10.5638 2.6410 0.97 0.4329
Error 45 122.4 2.7206    



The SAS System

The GLM Procedure
 

Level of
Year
N Mortality
Mean Std Dev
1968 10 5.99000000 0.98933199
1969 10 5.77000000 1.39845152
1970 10 4.91000000 0.77952977
1971 10 4.80000000 1.30384048
1972 10 4.72000000 0.95078915



The SAS System

The GLM Procedure
 

O'Brien's Test for Homogeneity of Mortality Variance
ANOVA of O'Brien's Spread Variable, W = 0.5
Source DF Sum of Squares Mean Square F Value Pr > F
Year 4 13.0418 3.2604 0.86 0.4953
Error 45 170.6 3.7918    



The SAS System

The GLM Procedure
 

Level of
Year
N Mortality
Mean Std Dev
1968 10 5.99000000 0.98933199
1969 10 5.77000000 1.39845152
1970 10 4.91000000 0.77952977
1971 10 4.80000000 1.30384048
1972 10 4.72000000 0.95078915



The SAS System

The GLM Procedure
 

Brown and Forsythe's Test for Homogeneity of Mortality Variance
ANOVA of Absolute Deviations from Group Medians
Source DF Sum of Squares Mean Square F Value Pr > F
Year 4 1.2788 0.3197 0.60 0.6627
Error 45 23.8750 0.5306    



The SAS System

The GLM Procedure
 

Level of
Year
N Mortality
Mean Std Dev
1968 10 5.99000000 0.98933199
1969 10 5.77000000 1.39845152
1970 10 4.91000000 0.77952977
1971 10 4.80000000 1.30384048
1972 10 4.72000000 0.95078915



The SAS System

The GLM Procedure
 
Dependent Variable: Mortality

Contrast DF Contrast SS Mean Square F Value Pr > F
Linear 1 12.32010000 12.32010000 10.02 0.0028
Quadratic 1 0.75778571 0.75778571 0.62 0.4365



The SAS System
Results from PROC RSREG

The RSREG Procedure

Coding Coefficients for the Independent
Variables
Factor Subtracted off Divided by
Year 1970.000000 2.000000

Response Surface for Variable Mortality
Response Mean 5.238000
Root MSE 1.095444
R-Square 0.1882
Coefficient of Variation 20.9134

Regression DF Type I Sum of Squares R-Square F Value Pr > F
Linear 1 12.320100 0.1773 10.27 0.0024
Quadratic 1 0.757786 0.0109 0.63 0.4308
Crossproduct 0 0 0.0000 . .
Total Model 2 13.077886 0.1882 5.45 0.0074

Residual DF Sum of Squares Mean Square F Value Pr > F
Lack of Fit 2 1.084914 0.542457 0.44 0.6460
Pure Error 45 55.315000 1.229222    
Total Error 47 56.399914 1.199998    

Parameter DF Estimate Standard Error t Value Pr > |t| Parameter Estimate
from Coded Data
Intercept 1 286220 359301 0.80 0.4297 5.090857
Year 1 -290.222429 364.772859 -0.80 0.4303 -0.702000
Year*Year 1 0.073571 0.092582 0.79 0.4308 0.294286

Factor DF Sum of Squares Mean Square F Value Pr > F
Year 2 13.077886 6.538943 5.45 0.0074



The SAS System
Results from PROC RSREG

The UNIVARIATE Procedure
Variable: Mortality

Year=1968

Moments
N 10 Sum Weights 10
Mean 5.99 Sum Observations 59.9
Std Deviation 0.98933199 Variance 0.97877778
Skewness 0.08853675 Kurtosis -1.8535056
Uncorrected SS 367.61 Corrected SS 8.809
Coeff Variation 16.5163937 Std Error Mean 0.31285424

Basic Statistical Measures
Location Variability
Mean 5.990000 Std Deviation 0.98933
Median 6.100000 Variance 0.97878
Mode 4.900000 Range 2.40000
    Interquartile Range 2.00000

Note: The mode displayed is the smallest of 2 modes with a count of 2.


Tests for Location: Mu0=0
Test Statistic p Value
Student's t t 19.1463 Pr > |t| <.0001
Sign M 5 Pr >= |M| 0.0020
Signed Rank S 27.5 Pr >= |S| 0.0020

Tests for Normality
Test Statistic p Value
Shapiro-Wilk W 0.858614 Pr < W 0.0735
Kolmogorov-Smirnov D 0.241508 Pr > D 0.0953
Cramer-von Mises W-Sq 0.079161 Pr > W-Sq 0.1964
Anderson-Darling A-Sq 0.557215 Pr > A-Sq 0.1139

Quantiles (Definition 5)
Quantile Estimate
100% Max 7.30
99% 7.30
95% 7.30
90% 7.25
75% Q3 7.00
50% Median 6.10
25% Q1 5.00
10% 4.90
5% 4.90
1% 4.90
0% Min 4.90

Extreme Observations
Lowest Highest
Value Obs Value Obs
4.9 7 6.3 9
4.9 2 6.4 8
5.0 4 7.0 6
5.0 1 7.2 5
5.9 3 7.3 10



The UNIVARIATE Procedure
Variable: Mortality

Q-Q plot for Mortality



The SAS System
Results from PROC RSREG

The UNIVARIATE Procedure
Variable: Mortality

Year=1969

Moments
N 10 Sum Weights 10
Mean 5.77 Sum Observations 57.7
Std Deviation 1.39845152 Variance 1.95566667
Skewness -0.1466961 Kurtosis -0.5750493
Uncorrected SS 350.53 Corrected SS 17.601
Coeff Variation 24.2365949 Std Error Mean 0.4422292

Basic Statistical Measures
Location Variability
Mean 5.770000 Std Deviation 1.39845
Median 5.450000 Variance 1.95567
Mode 7.600000 Range 4.30000
    Interquartile Range 2.10000

Tests for Location: Mu0=0
Test Statistic p Value
Student's t t 13.04753 Pr > |t| <.0001
Sign M 5 Pr >= |M| 0.0020
Signed Rank S 27.5 Pr >= |S| 0.0020

Tests for Normality
Test Statistic p Value
Shapiro-Wilk W 0.943907 Pr < W 0.5972
Kolmogorov-Smirnov D 0.176548 Pr > D >0.1500
Cramer-von Mises W-Sq 0.042347 Pr > W-Sq >0.2500
Anderson-Darling A-Sq 0.27439 Pr > A-Sq >0.2500

Quantiles (Definition 5)
Quantile Estimate
100% Max 7.60
99% 7.60
95% 7.60
90% 7.60
75% Q3 7.10
50% Median 5.45
25% Q1 5.00
10% 3.90
5% 3.30
1% 3.30
0% Min 3.30

Extreme Observations
Lowest Highest
Value Obs Value Obs
3.3 12 5.5 14
4.5 13 6.4 18
5.0 15 7.1 20
5.3 19 7.6 16
5.4 11 7.6 17



The UNIVARIATE Procedure
Variable: Mortality

Q-Q plot for Mortality



The SAS System
Results from PROC RSREG

The UNIVARIATE Procedure
Variable: Mortality

Year=1970

Moments
N 10 Sum Weights 10
Mean 4.91 Sum Observations 49.1
Std Deviation 0.77952977 Variance 0.60766667
Skewness 0.95670008 Kurtosis 0.0277596
Uncorrected SS 246.55 Corrected SS 5.469
Coeff Variation 15.8763701 Std Error Mean 0.24650896

Basic Statistical Measures
Location Variability
Mean 4.910000 Std Deviation 0.77953
Median 4.700000 Variance 0.60767
Mode 4.300000 Range 2.40000
    Interquartile Range 0.90000

Tests for Location: Mu0=0
Test Statistic p Value
Student's t t 19.91814 Pr > |t| <.0001
Sign M 5 Pr >= |M| 0.0020
Signed Rank S 27.5 Pr >= |S| 0.0020

Tests for Normality
Test Statistic p Value
Shapiro-Wilk W 0.90363 Pr < W 0.2400
Kolmogorov-Smirnov D 0.156109 Pr > D >0.1500
Cramer-von Mises W-Sq 0.069106 Pr > W-Sq >0.2500
Anderson-Darling A-Sq 0.435813 Pr > A-Sq 0.2409

Quantiles (Definition 5)
Quantile Estimate
100% Max 6.40
99% 6.40
95% 6.40
90% 6.20
75% Q3 5.20
50% Median 4.70
25% Q1 4.30
10% 4.15
5% 4.00
1% 4.00
0% Min 4.00

Extreme Observations
Lowest Highest
Value Obs Value Obs
4.0 23 4.8 27
4.3 30 5.1 22
4.3 21 5.2 29
4.4 25 6.0 24
4.6 26 6.4 28



The UNIVARIATE Procedure
Variable: Mortality

Q-Q plot for Mortality



The SAS System
Results from PROC RSREG

The UNIVARIATE Procedure
Variable: Mortality

Year=1971

Moments
N 10 Sum Weights 10
Mean 4.8 Sum Observations 48
Std Deviation 1.30384048 Variance 1.7
Skewness 1.30609626 Kurtosis 2.73318504
Uncorrected SS 245.7 Corrected SS 15.3
Coeff Variation 27.1633434 Std Error Mean 0.41231056

Basic Statistical Measures
Location Variability
Mean 4.800000 Std Deviation 1.30384
Median 4.450000 Variance 1.70000
Mode 4.400000 Range 4.80000
    Interquartile Range 1.20000

Tests for Location: Mu0=0
Test Statistic p Value
Student's t t 11.64171 Pr > |t| <.0001
Sign M 5 Pr >= |M| 0.0020
Signed Rank S 27.5 Pr >= |S| 0.0020

Tests for Normality
Test Statistic p Value
Shapiro-Wilk W 0.901446 Pr < W 0.2272
Kolmogorov-Smirnov D 0.190989 Pr > D >0.1500
Cramer-von Mises W-Sq 0.075969 Pr > W-Sq 0.2161
Anderson-Darling A-Sq 0.473071 Pr > A-Sq 0.1959

Quantiles (Definition 5)
Quantile Estimate
100% Max 7.80
99% 7.80
95% 7.80
90% 6.80
75% Q3 5.20
50% Median 4.45
25% Q1 4.00
10% 3.45
5% 3.00
1% 3.00
0% Min 3.00

Extreme Observations
Lowest Highest
Value Obs Value Obs
3.0 33 4.5 37
3.9 40 5.0 34
4.0 36 5.2 31
4.4 39 5.8 35
4.4 38 7.8 32



The UNIVARIATE Procedure
Variable: Mortality

Q-Q plot for Mortality



The SAS System
Results from PROC RSREG

The UNIVARIATE Procedure
Variable: Mortality

Year=1972

Moments
N 10 Sum Weights 10
Mean 4.72 Sum Observations 47.2
Std Deviation 0.95078915 Variance 0.904
Skewness 0.89290833 Kurtosis 0.00685603
Uncorrected SS 230.92 Corrected SS 8.136
Coeff Variation 20.1438378 Std Error Mean 0.30066593

Basic Statistical Measures
Location Variability
Mean 4.720000 Std Deviation 0.95079
Median 4.500000 Variance 0.90400
Mode 3.800000 Range 2.80000
    Interquartile Range 1.50000

Tests for Location: Mu0=0
Test Statistic p Value
Student's t t 15.69849 Pr > |t| <.0001
Sign M 5 Pr >= |M| 0.0020
Signed Rank S 27.5 Pr >= |S| 0.0020

Tests for Normality
Test Statistic p Value
Shapiro-Wilk W 0.893614 Pr < W 0.1862
Kolmogorov-Smirnov D 0.191493 Pr > D >0.1500
Cramer-von Mises W-Sq 0.062287 Pr > W-Sq >0.2500
Anderson-Darling A-Sq 0.415674 Pr > A-Sq >0.2500

Quantiles (Definition 5)
Quantile Estimate
100% Max 6.6
99% 6.6
95% 6.6
90% 6.2
75% Q3 5.3
50% Median 4.5
25% Q1 3.8
10% 3.8
5% 3.8
1% 3.8
0% Min 3.8

Extreme Observations
Lowest Highest
Value Obs Value Obs
3.8 50 4.5 48
3.8 47 5.0 41
3.8 45 5.3 49
4.1 44 5.8 42
4.5 48 6.6 43



The UNIVARIATE Procedure
Variable: Mortality

Q-Q plot for Mortality