Professional Documents
Culture Documents
Number of observations: 10
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_3. It constructs various tests and graphs to compare
the mean values of Col_3 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
100
80
60
Col_3
40
20
0
S1 S2 S3 S4 S5
Col_1
The StatAdvisor
This table shows various statistics for Col_3 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
Graphical ANOVA for Col_3
S2 S1 S3 S4 S5
Col_1
Residuals
-90 -60 -30 0 30 60 90
The StatAdvisor
The ANOVA table decomposes the variance of Col_3 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals , is a ratio of the between-group estimate to the within-group
estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference between the mean
Col_3 from one level of Col_1 to another at the 95.0% confidence level. To determine which means are significantly
different from which others, select Multiple Range Tests from the list of Tabular Options.
100
80
60
Col_3
40
20
0
S1 S2 S3 S4 S5
Col_1
Table of Means for Col_3 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
S1 2 22.0 0 22.0 22.0
S2 2 1.0 0 1.0 1.0
S3 2 59.0 0 59.0 59.0
S4 2 85.0 0 85.0 85.0
S5 2 100.0 0 100.0 100.0
Total 10 53.4
The StatAdvisor
This table shows the mean Col_3 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
S1
S2
Col_1
S3
S4
S5
0 20 40 60 80 100
Col_3
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 10 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence
level. At the top of the page, 5 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
0.1
0.06
0.02
residual
-0.02
-0.06
-0.1
S1 S2 S3 S4 S5
Col_1
100
UDL=53.40
CL=53.40
80 LDL=53.40
60
Mean
40
20
0
S1 S2 S3 S4 S5
Col_1
Median Plot
100
80
60
Col_3
40
20
0
S1 S2 S3 S4 S5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
S1 2 2 0 22.0
S2 2 2 0 1.0
S3 2 2 0 59.0
S4 2 0 2 85.0
S5 2 0 2 100.0
Test statistic = 10.0 P-Value = 0.0404277
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 59.0. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
S1
S2
0.8 S3
S4
S5
proportion
0.6
0.4
0.2
0
0 20 40 60 80 100
Col_3
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_2. It constructs various tests and graphs to compare
the mean values of Col_2 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
65
61
57
Col_2
53
49
45
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_2 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
L5 L4 L3 L2 L1
Col_1 P = 0.0000
Residuals
-23 -13 -3 7 17
The StatAdvisor
The ANOVA table decomposes the variance of Col_2 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 57.3882, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_2 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
Means and 95.0 Percent LSD Intervals
66
63
60
Col_2
57
54
51
48
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_2 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 62.982 0.705997 61.9407 64.0233
L2 5 60.768 0.705997 59.7267 61.8093
L3 5 59.898 0.705997 58.8567 60.9393
L4 5 57.694 0.705997 56.6527 58.7353
L5 5 49.152 0.705997 48.1107 50.1933
Total 25 58.0988
The StatAdvisor
This table shows the mean Col_2 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
45 49 53 57 61 65
Col_2
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 9 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 4 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
Residual Plot for Col_2
residual 2
-2
-4
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 0.16931 0.951521
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_2 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 0, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_2
With 95% Decision Limits
64
UDL=59.86
CL=58.10
61 LDL=56.34
58
Mean
55
52
49
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_2 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
64
61
58
Col_2
55
52
49
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 0 5 63.09
L2 5 1 4 61.5
L3 5 2 3 60.62
L4 5 5 0 58.31
L5 5 5 0 49.72
Test statistic = 16.9872 P-Value = 0.00194407
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 59.22. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
45 49 53 57 61 65
Col_2
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_3. It constructs various tests and graphs to compare
the mean values of Col_3 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
18
17
16
Col_3
15
14
13
12
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_3 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
WARNING: There is more than a 3 to 1 difference between the smallest standard deviation and the largest. This may
cause problems since the analysis of variance assumes that the standard deviations at all levels are equal. Select
Variance Check from the list of Tabular Options to run a formal statistical test for differences among the sigmas. You
may want to consider transforming the values of Col_3 to remove any dependence of the standard deviation on the
mean.
Graphical ANOVA for Col_3
L1 L2L3 L4 L5
Col_1 P = 0.0000
Residuals
-3.8 -1.8 0.2 2.2 4.2 6.2
The StatAdvisor
The ANOVA table decomposes the variance of Col_3 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 36.6032, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_3 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
18
17
16
Col_3
15
14
13
12
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_3 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 12.716 0.254609 12.3405 13.0915
L2 5 13.392 0.254609 13.0165 13.7675
L3 5 13.56 0.254609 13.1845 13.9355
L4 5 14.732 0.254609 14.3565 15.1075
L5 5 16.638 0.254609 16.2625 17.0135
Total 25 14.2076
The StatAdvisor
This table shows the mean Col_3 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
12 13 14 15 16 17 18
Col_3
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 8 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 4 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
1.2
0.8
0.4
residual
-0.4
-0.8
-1.2
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 1.79713 0.16905
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_3 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 2, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_3
With 95% Decision Limits
17
UDL=14.84
CL=14.21
16 LDL=13.57
15
Mean
14
13
12
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_3 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
17
16
15
Col_3
14
13
12
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 5 0 12.49
L2 5 3 2 13.06
L3 5 5 0 13.53
L4 5 0 5 14.65
L5 5 0 5 16.72
Test statistic = 20.1923 P-Value = 0.000457582
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 13.96. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
12 13 14 15 16 17 18
Col_3
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_4. It constructs various tests and graphs to compare
the mean values of Col_4 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
15
14.5
14
Col_4
13.5
13
12.5
12
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_4 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
WARNING: There is more than a 3 to 1 difference between the smallest standard deviation and the largest. This may
cause problems since the analysis of variance assumes that the standard deviations at all levels are equal. Select
Variance Check from the list of Tabular Options to run a formal statistical test for differences among the sigmas. You
may want to consider transforming the values of Col_4 to remove any dependence of the standard deviation on the
mean.
Graphical ANOVA for Col_4
L2 L4 L3
L1 L5
Col_1 P = 0.4104
Residuals
-1.2 -0.8 -0.4 0 0.4 0.8 1.2
The StatAdvisor
The ANOVA table decomposes the variance of Col_4 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 1.04202, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is greater than or equal to 0.05, there is not a statistically
significant difference between the mean Col_4 from one level of Col_1 to another at the 95.0% confidence level.
14.4
14
13.6
Col_4
13.2
12.8
12.4
12
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_4 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 13.8 0.223643 13.4701 14.1299
L2 5 13.314 0.223643 12.9841 13.6439
L3 5 13.788 0.223643 13.4581 14.1179
L4 5 13.714 0.223643 13.3841 14.0439
L5 5 13.904 0.223643 13.5741 14.2339
Total 25 13.704
The StatAdvisor
This table shows the mean Col_4 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. There are no
statistically significant differences between any pair of means at the 95.0% confidence level. At the top of the page,
one homogenous group is identified by a column of X's. Within each column, the levels containing X's form a group of
means within which there are no statistically significant differences. The method currently being used to discriminate
among the means is Fisher's least significant difference (LSD) procedure. With this method, there is a 5.0% risk of
calling each pair of means significantly different when the actual difference equals 0.
1.2
0.8
0.4
residual
-0.4
-0.8
-1.2
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 2.83243 0.0518835
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_4 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 3, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_4
With 95% Decision Limits
14.5
UDL=14.26
CL=13.70
14.2 LDL=13.15
13.9
Mean
13.6
13.3
13
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_4 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is greater than or equal to 0.05, there is not a statistically
significant difference amongst the medians at the 95.0% confidence level.
Median Plot
14.1
13.9
Col_4
13.7
13.5
13.3
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 13.71. Since the P-value for the chi-
square test is greater than or equal to 0.05, the medians of the samples are not significantly different at the 95.0%
confidence level. Also included (if available) are 95.0% confidence intervals for each median based on the order
statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
12 12.5 13 13.5 14 14.5 15
Col_4
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_2. It constructs various tests and graphs to compare
the mean values of Col_2 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
53
51
49
Col_2
47
45
43
41
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_2 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
Graphical ANOVA for Col_2
L5 L4 L3 L1 L2
Col_1 P = 0.0000
Residuals
-12 -8 -4 0 4 8
The StatAdvisor
The ANOVA table decomposes the variance of Col_2 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 45.0922, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_2 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
51
49
47
Col_2
45
43
41
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_2 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 49.534 0.491531 48.809 50.259
L2 5 50.264 0.491531 49.539 50.989
L3 5 48.98 0.491531 48.255 49.705
L4 5 45.212 0.491531 44.487 45.937
L5 5 42.544 0.491531 41.819 43.269
Total 25 47.3068
The StatAdvisor
This table shows the mean Col_2 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
41 43 45 47 49 51 53
Col_2
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 7 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 3 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
2.5
1.5
0.5
residual
-0.5
-1.5
-2.5
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 1.09501 0.385994
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_2 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 0, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_2
With 95% Decision Limits
52
UDL=48.53
CL=47.31
50 LDL=46.08
48
Mean
46
44
42
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_2 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
52
50
48
Col_2
46
44
42
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 1 4 49.4
L2 5 0 5 50.05
L3 5 2 3 48.79
L4 5 5 0 45.38
L5 5 5 0 42.43
Test statistic = 16.9872 P-Value = 0.00194407
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 48.65. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
41 43 45 47 49 51 53
Col_2
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_2. It constructs various tests and graphs to compare
the mean values of Col_2 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
53
51
49
Col_2
47
45
43
41
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_2 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
Graphical ANOVA for Col_2
L5 L4 L3 L1 L2
Col_1 P = 0.0000
Residuals
-12 -8 -4 0 4 8
The StatAdvisor
The ANOVA table decomposes the variance of Col_2 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 45.0922, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_2 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
51
49
47
Col_2
45
43
41
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_2 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 49.534 0.491531 48.809 50.259
L2 5 50.264 0.491531 49.539 50.989
L3 5 48.98 0.491531 48.255 49.705
L4 5 45.212 0.491531 44.487 45.937
L5 5 42.544 0.491531 41.819 43.269
Total 25 47.3068
The StatAdvisor
This table shows the mean Col_2 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
41 43 45 47 49 51 53
Col_2
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 7 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 3 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
2.5
1.5
0.5
residual
-0.5
-1.5
-2.5
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 1.09501 0.385994
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_2 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 0, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_2
With 95% Decision Limits
52
UDL=48.53
CL=47.31
50 LDL=46.08
48
Mean
46
44
42
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_2 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
52
50
48
Col_2
46
44
42
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 1 4 49.4
L2 5 0 5 50.05
L3 5 2 3 48.79
L4 5 5 0 45.38
L5 5 5 0 42.43
Test statistic = 16.9872 P-Value = 0.00194407
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 48.65. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
41 43 45 47 49 51 53
Col_2
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_3. It constructs various tests and graphs to compare
the mean values of Col_3 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
14
Col_3 13
12
11
10
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_3 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
Graphical ANOVA for Col_3
L2 L5 L3
L4 L1
Col_1 P = 0.0027
Residuals
-2.7 -1.7 -0.7 0.3 1.3 2.3 3.3
The StatAdvisor
The ANOVA table decomposes the variance of Col_3 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 5.88063, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_3 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
14
13.5
13
Col_3
12.5
12
11.5
11
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_3 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 13.422 0.291314 12.9923 13.8517
L2 5 11.45 0.291314 11.0203 11.8797
L3 5 12.59 0.291314 12.1603 13.0197
L4 5 12.63 0.291314 12.2003 13.0597
L5 5 12.362 0.291314 11.9323 12.7917
Total 25 12.4908
The StatAdvisor
This table shows the mean Col_3 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
10 11 12 13 14
Col_3
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 5 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 3 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
1.2
0.8
0.4
residual
-0.4
-0.8
-1.2
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 0.142945 0.964035
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_3 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 0, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_3
With 95% Decision Limits
13.5
UDL=13.22
CL=12.49
13 LDL=11.76
12.5
Mean
12
11.5
11
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_3 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
14
13.5
13
Col_3
12.5
12
11.5
11
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 0 5 13.72
L2 5 5 0 11.23
L3 5 3 2 12.68
L4 5 2 3 12.77
L5 5 3 2 12.63
Test statistic = 10.5769 P-Value = 0.0317537
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 12.68. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
10 11 12 13 14
Col_3
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_4. It constructs various tests and graphs to compare
the mean values of Col_4 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
15
14
13
Col_4
12
11
10
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_4 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
WARNING: There is more than a 3 to 1 difference between the smallest standard deviation and the largest. This may
cause problems since the analysis of variance assumes that the standard deviations at all levels are equal. Select
Variance Check from the list of Tabular Options to run a formal statistical test for differences among the sigmas. You
may want to consider transforming the values of Col_4 to remove any dependence of the standard deviation on the
mean.
Graphical ANOVA for Col_4
L5 L2 L4 L1 L3
Col_1 P = 0.0000
Residuals
-2.8 -1.8 -0.8 0.2 1.2 2.2 3.2
The StatAdvisor
The ANOVA table decomposes the variance of Col_4 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 12.7038, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_4 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
15
14
Col_4
13
12
11
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_4 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 13.248 0.223633 12.9181 13.5779
L2 5 12.426 0.223633 12.0961 12.7559
L3 5 13.74 0.223633 13.4101 14.0699
L4 5 12.698 0.223633 12.3681 13.0279
L5 5 11.652 0.223633 11.3221 11.9819
Total 25 12.7528
The StatAdvisor
This table shows the mean Col_4 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
10 11 12 13 14 15
Col_4
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 7 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 4 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
1.5
0.5
residual
-0.5
-1
-1.5
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 0.502908 0.734
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_4 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 1, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_4
With 95% Decision Limits
14
UDL=13.31
13.5 CL=12.75
LDL=12.19
13
Mean
12.5
12
11.5
11
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_4 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
13.8
13.5
13.2
Col_4
12.9
12.6
12.3
12
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 1 4 13.01
L2 5 4 1 12.41
L3 5 0 5 13.69
L4 5 3 2 12.71
L5 5 5 0 12.06
Test statistic = 13.7821 P-Value = 0.00802415
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 12.85. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
10 11 12 13 14 15
Col_4
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_2. It constructs various tests and graphs to compare
the mean values of Col_2 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
57
53
49
Col_2
45
41
37
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_2 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
WARNING: There is more than a 3 to 1 difference between the smallest standard deviation and the largest. This may
cause problems since the analysis of variance assumes that the standard deviations at all levels are equal. Select
Variance Check from the list of Tabular Options to run a formal statistical test for differences among the sigmas. You
may want to consider transforming the values of Col_2 to remove any dependence of the standard deviation on the
mean.
Graphical ANOVA for Col_2
L5 L4 L3 L2 L1
Col_1 P = 0.0000
Residuals
-14 -9 -4 1 6 11 16
The StatAdvisor
The ANOVA table decomposes the variance of Col_2 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 33.279, is a ratio of the between-group estimate to the within-
group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference between
the mean Col_2 from one level of Col_1 to another at the 95.0% confidence level. To determine which means are
significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
55
52
49
Col_2
46
43
40
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_2 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 52.884 0.765573 51.7548 54.0132
L2 5 48.134 0.765573 47.0048 49.2632
L3 5 47.428 0.765573 46.2988 48.5572
L4 5 43.9 0.765573 42.7708 45.0292
L5 5 41.282 0.765573 40.1528 42.4112
Total 25 46.7256
The StatAdvisor
This table shows the mean Col_2 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
37 41 45 49 53 57
Col_2
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 9 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 4 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
1
residual
-1
-3
-5
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 1.51662 0.235272
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_2 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 3, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_2
With 95% Decision Limits
53
UDL=48.64
51 CL=46.73
LDL=44.82
49
Mean
47
45
43
41
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_2 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
54
52
50
Col_2
48
46
44
42
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 0 5 53.07
L2 5 1 4 48.13
L3 5 2 3 47.74
L4 5 5 0 43.73
L5 5 5 0 42.53
Test statistic = 16.9872 P-Value = 0.00194407
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 47.02. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
37 41 45 49 53 57
Col_2
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_3. It constructs various tests and graphs to compare
the mean values of Col_3 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
14.6
13.6
12.6
Col_3
11.6
10.6
9.6
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_3 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
WARNING: There is more than a 3 to 1 difference between the smallest standard deviation and the largest. This may
cause problems since the analysis of variance assumes that the standard deviations at all levels are equal. Select
Variance Check from the list of Tabular Options to run a formal statistical test for differences among the sigmas. You
may want to consider transforming the values of Col_3 to remove any dependence of the standard deviation on the
mean.
Graphical ANOVA for Col_3
L1 L3 L2 L4 L5
Col_1 P = 0.0017
Residuals
-2.9 -1.9 -0.9 0.1 1.1 2.1
The StatAdvisor
The ANOVA table decomposes the variance of Col_3 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 6.43814, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_3 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
13
12.5
12
Col_3
11.5
11
10.5
10
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_3 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 10.594 0.310116 10.1366 11.0514
L2 5 12.108 0.310116 11.6506 12.5654
L3 5 11.264 0.310116 10.8066 11.7214
L4 5 12.298 0.310116 11.8406 12.7554
L5 5 12.44 0.310116 11.9826 12.8974
Total 25 11.7408
The StatAdvisor
This table shows the mean Col_3 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 5 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 3 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
1.5
0.5
residual
-0.5
-1
-1.5
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 2.2312 0.102057
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_3 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is greater than or equal to 0.05,
there is not a statistically significant difference amongst the standard deviations at the 95.0% confidence level.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 4, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_3
With 95% Decision Limits
13
UDL=12.51
12.5 CL=11.74
LDL=10.97
12
Mean
11.5
11
10.5
10
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_3 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
12.5
12
11.5
Col_3
11
10.5
10
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 5 0 10.77
L2 5 2 3 12.21
L3 5 4 1 11.1
L4 5 1 4 12.37
L5 5 2 3 12.43
Test statistic = 8.76623 P-Value = 0.0672156
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 11.99. Since the P-value for the chi-
square test is greater than or equal to 0.05, the medians of the samples are not significantly different at the 95.0%
confidence level. Also included (if available) are 95.0% confidence intervals for each median based on the order
statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
9.6 10.6 11.6 12.6 13.6 14.6
Col_3
Number of observations: 25
Number of levels: 5
The StatAdvisor
This procedure performs a one-way analysis of variance for Col_4. It constructs various tests and graphs to compare
the mean values of Col_4 for the 5 different levels of Col_1. The F-test in the ANOVA table will test whether there are
any significant differences amongst the means. If there are, the Multiple Range Tests will tell you which means are
significantly different from which others. If you are worried about the presence of outliers, choose the Kruskal-Wallis
Test which compares medians instead of means. The various plots will help you judge the practical significance of the
results, as well as allow you to look for possible violations of the assumptions underlying the analysis of variance.
Scatterplot by Level Code
16
15
14
Col_4
13
12
11
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
This table shows various statistics for Col_4 for each of the 5 levels of Col_1. The one-way analysis of variance is
primarily intended to compare the means of the different levels, listed here under the Average column. Select Means
Plot from the list of Graphical Options to display the means graphically.
WARNING: There is more than a 3 to 1 difference between the smallest standard deviation and the largest. This may
cause problems since the analysis of variance assumes that the standard deviations at all levels are equal. Select
Variance Check from the list of Tabular Options to run a formal statistical test for differences among the sigmas. You
may want to consider transforming the values of Col_4 to remove any dependence of the standard deviation on the
mean.
Graphical ANOVA for Col_4
L5 L3 L1 L2 L4
Col_1 P = 0.0092
Residuals
-2.4 -1.4 -0.4 0.6 1.6 2.6
The StatAdvisor
The ANOVA table decomposes the variance of Col_4 into two components: a between-group component and a within-
group component. The F-ratio, which in this case equals 4.51948, is a ratio of the between-group estimate to the
within-group estimate. Since the P-value of the F-test is less than 0.05, there is a statistically significant difference
between the mean Col_4 from one level of Col_1 to another at the 95.0% confidence level. To determine which means
are significantly different from which others, select Multiple Range Tests from the list of Tabular Options.
15
14.5
14
Col_4
13.5
13
12.5
12
L1 L2 L3 L4 L5
Col_1
Table of Means for Col_4 by Col_1 with 95.0 percent LSD intervals
Stnd. error
Col_1 Count Mean (pooled s) Lower limit Upper limit
L1 5 13.552 0.319796 13.0803 14.0237
L2 5 13.94 0.319796 13.4683 14.4117
L3 5 13.116 0.319796 12.6443 13.5877
L4 5 14.26 0.319796 13.7883 14.7317
L5 5 12.536 0.319796 12.0643 13.0077
Total 25 13.4808
The StatAdvisor
This table shows the mean Col_4 for each level of Col_1. It also shows the standard error of each mean, which is a
measure of its sampling variability. The standard error is formed by dividing the pooled standard deviation by the
square root of the number of observations at each level. The table also displays an interval around each mean. The
intervals currently displayed are based on Fisher's least significant difference (LSD) procedure. They are constructed in
such a way that if two means are the same, their intervals will overlap 95.0% of the time. You can display the intervals
graphically by selecting Means Plot from the list of Graphical Options. In the Multiple Range Tests, these intervals are
used to determine which means are significantly different from which others.
Box-and-Whisker Plot
L1
L2
Col_1
L3
L4
L5
11 12 13 14 15 16
Col_4
The StatAdvisor
This table applies a multiple comparison procedure to determine which means are significantly different from which
others. The bottom half of the output shows the estimated difference between each pair of means. An asterisk has been
placed next to 4 pairs, indicating that these pairs show statistically significant differences at the 95.0% confidence level.
At the top of the page, 3 homogenous groups are identified using columns of X's. Within each column, the levels
containing X's form a group of means within which there are no statistically significant differences. The method
currently being used to discriminate among the means is Fisher's least significant difference (LSD) procedure. With
this method, there is a 5.0% risk of calling each pair of means significantly different when the actual difference equals
0.
1
residual
-1
-2
L1 L2 L3 L4 L5
Col_1
Variance Check
Test P-Value
Levene's 3.23907 0.0334105
The StatAdvisor
The statistic displayed in this table tests the null hypothesis that the standard deviations of Col_4 within each of the 5
levels of Col_1 is the same. Of particular interest is the P-value. Since the the P-value is less than 0.05, there is a
statistically significant difference amongst the standard deviations at the 95.0% confidence level. This violates one of
the important assumptions underlying the analysis of variance and will invalidate most of the standard statistical tests.
The table also shows a comparison of the standard deviations for each pair of samples. P-Values below 0.05, of which
there are 3, indicate a statistically significant difference between the two sigmas at the 5% significance level.
Analysis of Means Plot for Col_4
With 95% Decision Limits
14.4
UDL=14.28
14 CL=13.48
LDL=12.68
13.6
Mean
13.2
12.8
12.4
12
L1 L2 L3 L4 L5
Col_1
The StatAdvisor
The Kruskal-Wallis test tests the null hypothesis that the medians of Col_4 within each of the 5 levels of Col_1 are the
same. The data from all the levels is first combined and ranked from smallest to largest. The average rank is then
computed for the data at each level. Since the P-value is less than 0.05, there is a statistically significant difference
amongst the medians at the 95.0% confidence level. To determine which medians are significantly different from
which others, select Box-and-Whisker Plot from the list of Graphical Options and select the median notch option.
Median Plot
14.4
14
13.6
Col_4
13.2
12.8
12.4
12
L1 L2 L3 L4 L5
Col_1
Col_1 Sample Size n<= n> Median 95.0% lower CL 95.0% upper CL
L1 5 3 2 13.51
L2 5 1 4 13.89
L3 5 5 0 13.21
L4 5 0 5 14.09
L5 5 4 1 12.53
Test statistic = 13.7821 P-Value = 0.00802415
The StatAdvisor
Mood's median test tests the hypothesis that the medians of all 5 samples are equal. It does so by counting the number
of observations in each sample on either side of the grand median, which equals 13.62. Since the P-value for the chi-
square test is less than 0.05, the medians of the samples are significantly different at the 95.0% confidence level. Also
included (if available) are 95.0% confidence intervals for each median based on the order statistics of each sample.
Quantile Plot
1 Col_1
L1
L2
0.8 L3
L4
L5
proportion
0.6
0.4
0.2
0
11 12 13 14 15 16
Col_4