6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

How-To menu Minitab Express™ Support

One-Way ANOVA
Search Minitab Express Support

Before you start

Overview

Data considerations
Interpret all statistics and graphs f
Example One-Way ANOVA
Learn more about Minitab
Perform the analysis

Enter your data Find definitions and interpretation guidance for every statistic and graph that is pro
ANOVA.
Select the group comparisons
In This Topic
Select the graphs Adj MS Normal probability plot of the residuals
Adjusted p-value Null hypothesis and Alternative hypoth
Interpret the results Adj SS Pooled StDev
Boxplot P-value – Factor
Confidence Interval for group means (95% Residuals versus fits
Key results
CI) Residuals versus order
DF Residuals versus the variables
All statistics and graphs Difference of Means R-sq
Equal variances R-sq (adj)
Methods and formulas Fisher Individual Tests for Differences of R-sq (pred)
Means S
Methods and formulas
F-value SE of Difference
Grouping Significance level
Histogram of residuals Simultaneous CI of the difference
Individual value plot Standard Deviation (StDev)
Interval plot T-value
Interval plot for differences of means
Mean
N

Adj MS
Adjusted mean squares measure how much variation a term or a model explains, a
other terms are in the model, regardless of the order they were entered. Unlike the
squares, the adjusted mean squares consider the degrees of freedom.

The adjusted mean square of the error (also called MSE or s2) is the variance around

Interpretation
Minitab uses the adjusted mean squares to calculate the p-value for a term. Minitab
adjusted mean squares to calculate the adjusted R2 statistic. Usually, you interpret t
the adjusted R2 statistic instead of the adjusted mean squares.

Adjusted p-value
By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 1/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

The adjusted p-value indicates which pairs within a family of comparisons are signifi
How-To The adjustment limits the family error rate to the alpha level that you specify. If you
value for multiple comparisons, the family error rate increases with each additional
One-Way ANOVA It is important to consider the family error rate when making multiple comparisons,
chances of committing a type I error for a series of comparisons is greater than the
Before you start one comparison alone.

Overview Interpretation
If the adjusted p-value is less than alpha, reject the null hypothesis and conclude th
Data considerations between a pair of group means is statistically significant. The adjusted p-value also r
smallest family error rate at which a particular null hypothesis is rejected.
Example

Perform the analysis Adj SS

Enter your data Adjusted sums of squares are measures of variation for different components of th
of the predictors in the model does not affect the calculation of the adjusted sum of
Select the group comparisons Analysis of Variance table, Minitab separates the sums of squares into different com
describe the variation due to different sources.
Select the graphs Adj SS Term
The adjusted sum of squares for a term is the increase in the regression sum of s
Interpret the results to a model with only the other terms. It quantifies the amount of variation in the
is explained by each term in the model.
Key results Adj SS Error
The error sum of squares is the sum of the squared residuals. It quantifies the va
All statistics and graphs that the predictors do not explain.

Adj SS Total
Methods and formulas
The total sum of squares is the sum of the term sum of squares and the error sum
quantifies the total variation in the data.
Methods and formulas

Interpretation
Minitab uses the adjusted sums of squares to calculate the p-value for a term. Minit
sums of squares to calculate the R2 statistic. Usually, you interpret the p-values and
instead of the sums of squares.

Boxplot
A boxplot provides a graphical summary of the distribution of each sample. The box
to compare the shape, the central tendency, and the variability of the samples.

Interpretation
Use a boxplot to examine the spread of the data and to identify any potential outlie
best when the sample size is greater than 20.
Skewed data
Examine the spread of your data to determine whether your data appear to be sk
are skewed, the majority of the data are located on the high or low side of the gra
indicates that the data might not be normally distributed. Often, skewness is easi
an individual value plot, a histogram, or a boxplot.

By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 2/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

How-To

One-Way ANOVA

Right-skewed Left-skewed
Before you start
The boxplot with right-skewed data shows average wait times. Most of the wait times are relat
Overview only a few of the wait times are longer. The boxplot with left-skewed data shows failure rate d
fail immediately, and many more items fail later.

Data considerations
Data that are severely skewed can affect the validity of the p-value if your sample
Example
values). If your data are severely skewed and you have a small sample, consider in
sample size.

Perform the analysis Outliers

Outliers, which are data values that are far away from other data values, can stro
Enter your data results. Often, outliers are easiest to identify on a boxplot.

Select the group comparisons

Select the graphs

Interpret the results

On a boxplot, asterisks (*) denote outliers.

Key results
Try to identify the cause of any outliers. Correct any data-entry errors or measure
All statistics and graphs
Consider removing data values for abnormal, one-time events (special causes). Th
analysis.

Methods and formulas

Methods and formulas

Confidence Interval for group means (95% CI)
These confidence intervals (CI) are ranges of values that are likely to contain the tru
population. The confidence intervals are calculated using the pooled standard devia

Because samples are random, two samples from a population are unlikely to yield i
intervals. But, if you repeat your sample many times, a certain percentage of the res
intervals contain the unknown population parameter. The percentage of these confi
that contain the parameter is the confidence level of the interval.

The confidence interval is composed of the following two parts:

Point estimate
The point estimate is the parameter that is calculated from the sample data. The
is centered around this value.
Margin of error
The margin of error defines the width of the confidence interval and is determine
variability in the sample, the sample size, and the confidence level. To calculate th
confidence interval, the error margin is added to the point estimate. To calculate
the confidence interval, the error margin is subtracted from the point estimate.

Interpretation
Use the confidence interval to assess the estimate of the population mean for each

For example, with a 95% confidence level, you can be 95% confident that the confid
contains the group mean. The confidence interval helps you assess the practical sig
By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK
results. Use your specialized knowledge to determine whether the confidence interv

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 3/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

that have practical significance for your situation. If the interval is too wide to be use
How-To increasing your sample size.

One-Way ANOVA
Means
Before you start Paint N Mean StDev 95% CI
Blend 1 6 14.733 3.363 (11.369, 18.097)
Overview Blend 2 6 8.567 5.500 (5.203, 11.931)
Blend 3 6 12.983 3.730 (9.619, 16.347)
Data considerations Blend 4 6 18.067 2.636 (14.703, 21.431)

Pooled StDev = 3.95012

Example

Perform the analysis In these results, each blend has a confidence interval for its mean hardness. The multiple compa
these data show that Blend 4 is significantly harder than Blend 2. That Blend 4 is harder than Ble
show that Blend 4 is hard enough for the intended use of the paint. The confidence interval for t
Enter your data
is better for judging whether Blend 4 is hard enough.

Select the group comparisons

Select the graphs DF

Interpret the results The total degrees of freedom (DF) are the amount of information in your data. The a
information to estimate the values of unknown population parameters. The total DF
Key results
the number of observations in your sample. The DF for a term show how much info
uses. Increasing your sample size provides more information about the population,
the total DF. Increasing the number of terms in your model uses more information,
All statistics and graphs
the DF available to estimate the variability of the parameter estimates.

Methods and formulas If two conditions are met, then Minitab partitions the DF for error. The first conditio
be terms you can fit with the data that are not included in the current model. For ex
a continuous predictor with 3 or more distinct values, you can estimate a quadratic
Methods and formulas
predictor. If the model does not include the quadratic term, then a term that the da
included in the model and this condition is met.

The second condition is that the data contain replicates. Replicates are observations
predictor has the same value. For example, if you have 3 observations where pressu
temperature is 25, then those 3 observations are replicates.

If the two conditions are met, then the two parts of the DF for error are lack-of-fit an
DF for lack-of-fit allow a test of whether the model form is adequate. The lack-of-fit t
degrees of freedom for lack-of-fit. The more DF for pure error, the greater the powe
test.

Difference of Means
This value is the difference between the sample means of two groups.

Interpretation
The differences between the sample means of the groups are estimates of the diffe
populations of these groups.

Because each mean difference is based on data from a sample and not from the en
cannot be certain that it equals the population difference. To better understand the
between population means, use the confidence intervals.
By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 4/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

How-To
Equal variances
One-Way ANOVA
Minitab assumes that the population standard deviations for all groups are equal.

Before you start

Interpretation
Overview Look in the standard deviation (StDev) column of the one-way ANOVA output to det
standard deviations are approximately equal.
Data considerations If you cannot assume equal variances, use Welch's ANOVA, which is an option for on
is available in Minitab Statistical Software.
Example

Perform the analysis Fisher Individual Tests for Differences of Means

Enter your data Use the individual confidence intervals to identify statistically significant differences
means, to determine likely ranges for the differences, and to determine whether the
Select the group comparisons practically significant. Fisher's individual tests table displays a set of confidence inte
difference between pairs of means.
Select the graphs The individual confidence level is the percentage of times that a single confidence in
true difference between one pair of group means, if you repeat the study. Individua
Interpret the results intervals are available only for Fisher's method. All of the other comparison method
simultaneous confidence intervals.
Key results Controlling the individual confidence level is uncommon because it does not contro
confidence level, which often increases to unacceptable levels. If you do not control
All statistics and graphs confidence level, the chance that at least one confidence interval does not contain t
increases with the number of comparisons.
Methods and formulas
The confidence interval of the difference is composed of the following two parts:

Methods and formulas Point estimate

The point estimate is the difference between a pair of means and is calculated fro
The confidence interval is centered on this value.

Error margin
The error margin defines the width of the confidence interval and is determined b
variability in the sample and the confidence level. To calculate the upper limit of t
interval, the error margin is added to the point estimate. To calculate the lower lim
confidence interval, the error margin is subtracted from the point estimate.

Interpretation
Use the confidence intervals to assess the differences between group means.

Fisher Individual Tests for Differences of Means

Difference of Difference of SE of T- A
Levels Means Difference 95% CI Value
-6.167 2.281 (-10.924, -2.70
Blend 2-Blend 1
-1.409)
Blend 3-Blend 1 -1.750 2.281 (-6.507, 3.007) -0.77
Blend 4-Blend 1 3.333 2.281 (-1.424, 8.091) 1.46
Blend 3-Blend 2 4.417 2.281 (-0.341, 9.174) 1.94
Blend 4-Blend 2 9.500 2.281 (4.743, 14.257) 4.17
Blend 4-Blend 3 5.083 2.281 (0.326, 9.841) 2.23
By using this site you agree to the use of cookies for analytics and personalized content. Read our policy
Simultaneous confidence level = 80.83%
OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 5/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

How-To The confidence intervals indicate the following:

The confidence interval for the difference between the means of Blend 4 and 2 extends fr
14.257. This range does not include zero, which indicates that the difference between thes
One-Way ANOVA statistically significant.
The confidence interval for the difference between the means of Blend 2 and 1 extends fr
Before you start -1.409. This range does not include zero, which indicates that the difference between thes
statistically significant.
Overview The confidence interval for the difference between the means of Blend 4 and 3 extends fr
9.841. This range does not include zero, which indicates that the difference between these
statistically significant.
Data considerations
The confidence intervals for all the remaining pairs of means include zero, which indicates
differences are not statistically significant.
Example
The 95% individual confidence level indicates that you can be 95% confident that each con
contains the true difference for that specific comparison. However, the simultaneous confi
Perform the analysis indicates that you can be only 80.83% confident that all the intervals contain the true diffe

Enter your data

Select the group comparisons

F-value
An F-value appears for each term in the Analysis of Variance table:
Select the graphs
F-value for the model or the terms
Interpret the results The F-value is the test statistic used to determine whether the term is associated
F-value for the lack-of-fit test
Key results The F-value is the test statistic used to determine whether the model is missing h
that include the predictors in the current model.
All statistics and graphs
Interpretation
Methods and formulas
Minitab uses the F-value to calculate the p-value, which you use to make a decision
significance of the terms and model. The p-value is a probability that measures the
Methods and formulas the null hypothesis. Lower probabilities provide stronger evidence against the null h

A sufficiently large F-value indicates that the term or model is significant.

If you want to use the F-value to determine whether to reject the null hypothesis, co
to your critical value. You can calculate the critical value in Minitab or find the critica
distribution table in most statistics books. For more information on using Minitab to
critical value, go to Using the inverse cumulative distribution function (ICDF) and clic
calculate critical values".

Grouping
Use the grouping information table to quickly determine whether the mean differen
pair of groups is statistically significant.

The grouping column of the Grouping Information table contains columns of letters
factor levels. Groups that do not share a letter have a mean difference that is statist

If the grouping table identifies differences that are statistically significant, use the co
of the differences to determine whether the differences are practically significant.

Interpretation

Grouping
By using this site you agree to the use of cookies for analytics and Information
personalized Using
content. Read ourthe Tukey Method and 95% Confidence
policy OK

Paint N Mean Grouping

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 6/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express
Blend 4 6 18.067 A
How-To Blend 1 6 14.733 A B
Blend 3 6 12.983 A B
Blend 2 6 8.567 B
One-Way ANOVA
Means that do not share a letter are significantly different.
Before you start

Overview In these results, the table shows that group A contains Blends 1, 3, and 4, and group B contains B
3. Blends 1 and 3 are in both groups. Differences between means that share a letter are not stat
significant. Blends 2 and 4 do not share a letter, which indicates that Blend 4 has a significantly h
Data considerations than Blend 2.

Example

Perform the analysis Histogram of residuals

Enter your data The histogram of the residuals shows the distribution of the residuals for all observ

Select the group comparisons Interpretation

Use the histogram of the residuals to determine whether the data are skewed or inc
Select the graphs patterns in the following table may indicate that the model does not meet the mode

Interpret the results Pattern What the pattern may indicate

A long tail in one direction Skewness

Key results
A bar that is far away from the other bars An outlier
All statistics and graphs
Because the appearance of a histogram depends on the number of intervals used to
Methods and formulas don't use a histogram to assess the normality of the residuals. Instead, use a norma

A histogram is most effective when you have approximately 20 or more data points
Methods and formulas too small, then each bar on the histogram does not contain enough data points to r
skewness or outliers.

Individual value plot

An individual value plot displays the individual values in each sample. The individua
easy to compare the samples. Each circle represents one observation. An individual
especially useful when your sample size is small.

Interpretation
Use an individual value plot to examine the spread of the data and to identify any p
Individual value plots are best when the sample size is less than 50.

Skewed data
Examine the spread of your data to determine whether your data appear to be sk
are skewed, the majority of the data are located on the high or low side of the gra
indicate that the data might not be normally distributed. Often, skewness is easie
individual value plot, a histogram, or a boxplot.

By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

Right-skewed Left-skewed

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 7/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

The individual value plot with right-skewed data shows wait times. Most of the wait times are
How-To
and only a few wait times are longer. The individual value plot with left-skewed data shows fai
A few items fail immediately, and many more items fail later.
One-Way ANOVA
Outliers
Before you start Outliers, which are data values that are far away from other data values, can stro
results. Often, outliers are easy to identify on an individual value plot.
Overview

Data considerations

Example

On an individual value plot, unusually low or high data values indicate potential outliers.
Perform the analysis

Enter your data Try to identify the cause of any outliers. Correct any data-entry errors or measure
Consider removing data values for abnormal, one-time events (special causes). Th
analysis.
Select the group comparisons

Select the graphs

Interval plot
Interpret the results
Use the interval plot to display the mean and confidence interval for each group.

Key results The interval plots show the following:

Each dot represents a sample mean.
All statistics and graphs
Each interval is a 95% individual confidence interval for the mean of a group. Y
confident that the group mean is within the group's confidence interval.
Methods and formulas

IMPORTANT
Methods and formulas
Interpret these intervals carefully because your rate of type I error increases whe
multiple comparisons. That is, the more comparisons you make, the higher the p
least one comparison will incorrectly conclude that one of the observed differen
different.

Interpretation
In these results, Blend 2 has the lowest mean and Blend 4 has the highest. You cann
this graph whether any differences are statistically significant. To determine statistic
assess the confidence intervals for the differences of means.

Interval plot for differences of means

By using this site you agree to the use of cookies for analytics
Use and
the personalized
confidence content.
intervalsRead our policy likely ranges for the differences
to determine OK and to as
significance of the differences. The graph displays a set of confidence intervals for t

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 8/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

between pairs of means. Confidence intervals that do not contain zero indicate a m
How-To is statistically significant.

Depending on the comparison method you chose, the plot compares different pairs
One-Way ANOVA displays one of the following types of confidence intervals.

Before you start Individual confidence level

The percentage of times that a single confidence interval would include the tru
Overview between one pair of group means if the study were repeated multiple times.
Simultaneous confidence level
Data considerations The percentage of times that a set of confidence intervals would include the tr
all group comparisons if the study were repeated multiple times.
Example
Controlling the simultaneous confidence level is particularly important when y
multiple comparisons. If you do not control the simultaneous confidence leve
Perform the analysis least one confidence interval does not contain the true difference increases w
comparisons.
Enter your data
Interpretation
Select the group comparisons
In these results, the confidence intervals indicate the following:
Select the graphs The confidence interval for the difference between the means of Blend 4 and
from 3.114 to 15.886. This range does not include zero, which indicates that th
between these means is statistically significant.
Interpret the results
The confidence intervals for the remaining pairs of means all include zero, wh
the differences are not statistically significant.
Key results
The 95% simultaneous confidence level indicates that we can be 95% confiden
All statistics and graphs
confidence intervals contain the true differences.
Each individual confidence interval has a confidence level of 98.89%. This resu
Methods and formulas can be 98.89% confident that each individual interval contains the true differe
specific pair of group means. The individual confidence level for each compari
95% simultaneous confidence level for all six comparisons.
Methods and formulas

Tukey Simultaneous Tests for Differences of Means

Difference of Levels Difference of Means SE of Difference 95% CI T-Value A

Blend 2-Blend 1 -6.167 2.281 (-12.553, 0.219) -2.70
Blend 3-Blend 1 -1.750 2.281 (-8.136, 4.636) -0.77
Blend 4-Blend 1 3.333 2.281 (-3.053, 9.719) 1.46
Blend 3-Blend 2 4.417 2.281 (-1.969, 10.803) 1.94
Blend 4-Blend 2 9.500 2.281 (3.114, 15.886) 4.17
Blend 4-Blend 3 5.083 2.281 (-1.303, 11.469) 2.23
By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK
Individual confidence level = 98.89%

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statisti… 9/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

How-To

Mean
One-Way ANOVA
The mean of the observations within each group. The mean describes each group w
Before you start identifying the center of the data. It is the sum of all the observations with a group d
number of observations in that group.
Overview
Interpretation
Data considerations The mean of each sample provides an estimate of each population mean. The differ
sample means are the estimates of the difference between the population means.
Example
Because the difference between the group means are based on data from a sample
population, you cannot be certain it equals the population difference. To get a bette
Perform the analysis
population difference, you can use the confidence interval.

Enter your data

Select the group comparisons

N
The sample size (N) is the total number of observations in each group.
Select the graphs

Interpretation
Interpret the results
The sample size affects the confidence interval and the power of the test.
Key results Usually, a larger sample yields a narrower confidence interval. A larger sample size a
more power to detect a difference. For more information, go to What is power?.
All statistics and graphs

Methods and formulas Normal probability plot of the residuals

Methods and formulas The normal plot of the residuals displays the residuals versus their expected values
distribution is normal.

Interpretation
Use the normal probability plot of residuals to verify the assumption that the residu
distributed. The normal probability plot of the residuals should approximately follow

NOTE
If your one-way ANOVA design meets the guidelines for sample size, the results a
substantially affected by departures from normality.

The following patterns violate the assumption that the residuals are normally distrib

S-curve implies a distribution with long tails.

By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 10/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express
Inverted S-curve implies a distribution with short tails.
How-To

One-Way ANOVA

Before you start

Downward curve implies a right-skewed distribution.

Overview

Data considerations

Example

Perform the analysis

A few points lying away from the line implies a distribution with outliers.

Enter your data

If you see a nonnormal pattern, use the other residual plots to check for other prob
model, such as missing terms or a time order effect. If the residuals do not follow a
Select the group comparisons
and the data do not meet the sample size guidelines, the confidence intervals and p
inaccurate.
Select the graphs

Interpret the results

Null hypothesis and Alternative hypothesis
Key results One-way ANOVA is a hypothesis test that evaluates two mutually exclusive stateme
more population means. These two statements are called the null hypothesis and th
All statistics and graphs hypotheses. A hypothesis test uses sample data to determine whether to reject the

For one-way ANOVA, the hypotheses for the test are the following:
Methods and formulas
The null hypothesis (H0) is that the group means are all equal.
Methods and formulas The alternative hypothesis (HA) is that not all group means are equal.

Interpretation
Compare the p-value to the significance level to determine whether to reject the nu

Pooled StDev
The pooled standard deviation is an estimate of the common standard deviation for
pooled standard deviation is the standard deviation of all data points around their g
around the overall mean). Larger groups have a proportionally greater influence on
estimate of the pooled standard deviation.

Interpretation
A higher standard deviation value indicates greater spread in the data. A higher valu
precise (wider) confidence intervals and low statistical power.

Minitab uses the pooled standard deviation to create the confidence intervals for bo
means and the differences between group means.

Example of a pooled standard deviation

Suppose your study has four groups, as shown in the following table.

By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK
Group Mean Standard Deviation N

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 11/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

Group Mean Standard Deviation N

How-To
1 9.7 2.5 50
One-Way ANOVA
2 12.1 2.9 50

Before you start 3 14.5 3.2 50

4 17.3 6.8 200

Overview

The first three groups are equal in size (n=50) with standard deviations around 3. Th
Data considerations
much larger (n=200) and has a higher standard deviation (6.8). Because the pooled
uses a weighted average, its value (5.488) is closer to the standard deviation of the l
Example

Perform the analysis

P-value – Factor
Enter your data The p-value is a probability that measures the evidence against the null hypothesis.
provide stronger evidence against the null hypothesis.
Select the group comparisons
Interpretation
Select the graphs
Use the p-value in the ANOVA output to determine whether the differences betwee
means are statistically significant.
Interpret the results
To determine whether any of the differences between the means are statistically sig
the p-value to your significance level to assess the null hypothesis. The null hypothe
Key results
population means are all equal. Usually, a significance level (denoted as α or alpha)
A significance level of 0.05 indicates a 5% risk of concluding that a difference exists w
All statistics and graphs actual difference.

Methods and formulas

P-value ≤ α: The differences between some of the means are statistically significa
If the p-value is less than or equal to the significance level, you reject the null hyp
conclude that not all of population means are equal. Use your specialized knowle
Methods and formulas
whether the differences are practically significant. For more information, go to Sta
practical significance.

P-value > α: The differences between the means are not statistically significant
If the p-value is greater than the significance level, you do not have enough evide
null hypothesis that the population means are all equal. Verify that your test has
detect a difference that is practically significant. For more information, go to Incre
hypothesis test.

Residuals versus fits

The residuals versus fits graph plots the residuals on the y-axis and the fitted values

Interpretation
Use the residuals versus fits plot to verify the assumption that the residuals are ran
and have constant variance. Ideally, the points should fall randomly on both sides o
recognizable patterns in the points.

The patterns in the following table may indicate that the model does not meet the m

Pattern What the patter

Fanning or uneven spreading of residuals across fitted values Nonconstant va

By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK
A point that is far away from zero An outlier

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 12/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

The following graphs show an outlier and a violation of the assumption that the res
How-To

One-Way ANOVA

Before you start

Overview Plot with outlier

One of the points is much larger than all of the other points. Therefore, the point is an outlier. If
many outliers, the model may not be acceptable. You should try to identify the cause of any outl
Data considerations
data entry or measurement errors. Consider removing data values that are associated with abno
events (special causes). Then, repeat the analysis.
Example

Perform the analysis

Enter your data

Select the group comparisons Plot with nonconstant variance

The variance of the residuals increases with the fitted values. Notice that, as the value of the fits
Select the graphs scatter among the residuals widens. This pattern indicates that the variances of the residuals are
(nonconstant).

Interpret the results

If you identify any patterns or outliers in your residual versus fits plot, consider the
Key results
Issue Possible solution

All statistics and graphs Nonconstant Consider using Stat > ANOVA > One-Way in Minitab Statistic
variance Under Options, uncheck Assume equal variances.
Methods and formulas
An outlier or 1. Verify that the observation is not a measurement error or
influential point error.
Methods and formulas
2. Consider performing the analysis without this observatio
how it impacts your results.

Residuals versus order

The residual versus order plot displays the residuals in the order that the data were

Interpretation
Use the residuals versus order plot to verify the assumption that the residuals are in
one another. Independent residuals show no trends or patterns when displayed in t
in the points may indicate that residuals near each other may be correlated, and thu
independent. Ideally, the residuals on the plot should fall randomly around the cent

If you see a pattern, investigate the cause. The following types of patterns may indic
residuals are dependent.
By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 13/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

How-To

One-Way ANOVA

Before you start Trend

Overview

Data considerations

Example
Shift

Perform the analysis

Enter your data

Select the group comparisons

Cycle
Select the graphs

Interpret the results

Residuals versus the variables
Key results
The residual versus variables plot displays the residuals versus another variable. Th
All statistics and graphs already be included in your model. Or, the variable may not be in the model, but yo
the response.
Methods and formulas
Interpretation
Methods and formulas If you see a non-random pattern in the residuals, it indicates that the variable affect
systematic way. Consider including this variable in an analysis.

R-sq
R2 is the percentage of variation in the response that is explained by the model. It is
minus the ratio of the error sum of squares (which is the variation that is not explai
the total sum of squares (which is the total variation in the model).

Interpretation
Use R2 to determine how well the model fits your data. The higher the R2 value, the
fits your data. R2 is always between 0% and 100%.

You can use a fitted line plot to graphically illustrate different R2 values. The first plo
simple regression model that explains 85.5% of the variation in the response. The se
illustrates a model that explains 22.6% of the variation in the response. The more va
explained by the model, the closer the data points fall to the fitted regression line. T
model could explain 100% of the variation, the fitted values would always equal the
and all of the data points would fall on the fitted line.

By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 14/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

How-To

One-Way ANOVA

Before you start

Consider the following issues when interpreting the R2 value:
Overview R2 always increases when you add additional predictors to a model. For exam
predictor model will always have an R2 that is at least as high the best four-pre
Data considerations Therefore, R2 is most useful when you compare models of the same size.
Small samples do not provide a precise estimate of the strength of the relation
Example response and predictors. If you need R2 to be more precise, you should use a
(typically, 40 or more).
Perform the analysis R2 is just one measure of how well the model fits the data. Even when a mode
should check the residual plots to verify that the model meets the model assu
Enter your data

Select the group comparisons R-sq (adj)

Select the graphs Adjusted R2 is the percentage of the variation in the response that is explained by th
for the number of predictors in the model relative to the number of observations. A
Interpret the results calculated as 1 minus the ratio of the mean square error (MSE) to the mean square

Key results Interpretation

Use adjusted R2 when you want to compare models that have different numbers of
All statistics and graphs always increases when you add a predictor to the model, even when there is no rea
the model. The adjusted R2 value incorporates the number of predictors in the mod
Methods and formulas
choose the correct model.

For example, you work for a potato chip company that examines the factors that aff
Methods and formulas of crumbled potato chips per container. You receive the following results as you add
forward stepwise approach:

Step % Potato Cooling rate Cooking temp R2 Adjusted R2

1 X 52% 51%

2 X X 63% 62%

3 X X X 65% 62%

The first step yields a statistically significant regression model. The second step add
model. Adjusted R2 increases, which indicates that cooling rate improves the model
which adds cooking temperature to the model, increases the R2 but not the adjuste
indicate that cooking temperature does not improve the model. Based on these res
removing cooking temperature from the model.

R-sq (pred)
Predicted R2 is calculated with a formula that is equivalent to systematically removin
from the data set, estimating the regression equation, and determining how well th
the removed observation. The value of predicted R2 ranges between 0% and 100%.
calculations for predicted R2 can produce negative values, Minitab displays zero for

Interpretation
By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 15/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

Use predicted R2 to determine how well your model predicts the response for new o
How-To Models that have larger predicted R2 values have better predictive ability.

A predicted R2 that is substantially less than R2 may indicate that the model is over-fi
One-Way ANOVA model occurs when you add terms for effects that are not important in the populati
may appear important in the sample data. The model becomes tailored to the samp
Before you start therefore, may not be useful for making predictions about the population.

Predicted R2 can also be more useful than adjusted R2 for comparing models becau
Overview
with observations that are not included in the model calculation.

Data considerations For example, an analyst at a financial consulting company develops a model to pred
conditions. The model looks promising because it has an R2 of 87%. However, the p
Example
to 52%, which indicates that the model may be over-fit.

Perform the analysis

S
Enter your data
S represents how far the data values fall from the fitted values. S is measured in the
response.
Select the group comparisons

Select the graphs

Interpretation
Use S to assess how well the model describes the response. S is measured in the un
Interpret the results variable and represents the how far the data values fall from the fitted values. The l
the better the model describes the response. However, a low S value by itself does n
model meets the model assumptions. You should check the residual plots to verify t
Key results
For example, you work for a potato chip company that examines the factors that aff
All statistics and graphs of crumbled potato chips per container. You reduce the model to the significant pre
calculated as 1.79. This result indicates that the standard deviation of the data point
Methods and formulas
values is 1.79. If you are comparing models, values that are lower than 1.79 indicate
higher values indicate a worse fit.

Methods and formulas

SE of Difference
The standard error of the difference between means (SE of Difference) estimates th
difference between sample means that you would obtain if you took repeated samp
populations.

Interpretation
Use the standard error of the difference between means to determine how precisel
between the sample means estimate the differences between the population mean
error value indicates a more precise estimate.

Minitab uses the standard error of the difference to calculate the confidence interva
differences between means, which is a range of values that is likely to include the po
differences.

Significance level
The significance level (denoted by alpha or α) is the maximum acceptable level of ris
null hypothesis when the null hypothesis is true (type I error).

Interpretation
By using this site you agree to the use of cookies for analytics
Use and
the personalized
significancecontent.
level toRead ourwhether
decide policy to reject or fail to reject
OK the null hypoth
p-value is less than the significance level, the usual interpretation is that the results

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 16/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express

significant, and you reject H0.

How-To
For one-way ANOVA, you reject the null hypothesis when there is sufficient evidence
not all of the means are equal.
One-Way ANOVA

Before you start

Simultaneous CI of the difference
Overview
Use the simultaneous confidence intervals to identify mean differences that are stat
to determine likely ranges for the differences, and to assess the practical significanc
Data considerations differences. The table displays a set of confidence intervals for the difference betwe
Confidence intervals that do not contain zero indicate a mean difference that is stat
Example
The simultaneous confidence level is the percentage of times that a set of confidenc
the true differences for all group comparisons if the study were repeated multiple t
Perform the analysis
Controlling the simultaneous confidence level is particularly important when you pe
Enter your data comparisons. If you do not control the simultaneous confidence level, the chance th
confidence interval does not contain the true difference increases with the number
Select the group comparisons The confidence interval of the difference is composed of the following two parts:

Point estimate
Select the graphs
The point estimate is the difference between a pair of means and is calculated fro
The confidence interval is centered around this value.
Interpret the results
Margin of error
Key results The margin of error defines the width of the confidence interval and is determine
variability in the sample, the sample size, and the confidence level. To calculate th
confidence interval, the error margin is added to the point estimate. To calculate
All statistics and graphs
the confidence interval, the error margin is subtracted from the point estimate. T
as the number of comparisons increases in order to maintain the simultaneous le
Methods and formulas that all intervals contain the true population differences.

Methods and formulas

Interpretation
Use the confidence intervals to assess the differences between group means.

Tukey Simultaneous Tests for Differences of Means

Difference of Difference of SE of T- A
Levels Means Difference 95% CI Value
-6.167 2.281 (-12.553, -2.70
Blend 2-Blend 1
0.219)
Blend 3-Blend 1 -1.750 2.281 (-8.136, 4.636) -0.77
Blend 4-Blend 1 3.333 2.281 (-3.053, 9.719) 1.46
4.417 2.281 (-1.969, 1.94
Blend 3-Blend 2
10.803)
Blend 4-Blend 2 9.500 2.281 (3.114, 15.886) 4.17
5.083 2.281 (-1.303, 2.23
Blend 4-Blend 3
11.469)

Individual confidence level = 98.89%

The confidence intervals indicate the following:

The confidence interval for the difference between the means of Blend 2 and Blend 4 exte
to 15.886. This range does not include zero, which indicates that the difference is statistica
The confidence intervals for the remaining pairs of means all include zero, which indicates
differences
By using this site you agree to the use of cookies for analytics and are not
personalized statistically
content. Read oursignificant.
policy OK
The 95% simultaneous confidence level indicates that we can be 95% confident that all of
confidence intervals contain the true differences.
https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 17/18
6/24/22, 1:22 PM Interpret all statistics and graphs for One-Way ANOVA - Minitab Express
The 98.89% individual confidence level indicates that we can be 98.89% confident that eac
How-To interval contains the true difference for that specific comparison.

One-Way ANOVA

Standard Deviation (StDev)

Before you start
The standard deviation is the most common measure of dispersion, or how spread
Overview around the mean. The symbol σ (sigma) is often used to represent the standard dev
population. The symbol s is used to represent the standard deviation of a sample.
Data considerations
Interpretation
Example
The standard deviation uses the same units as the variable. A higher standard devia
greater spread in the data. A good rule of thumb for a normal distribution is as follo
Perform the analysis
Approximately 68% of the values fall within one standard deviation of the mea
Enter your data 95% of the values fall within two standard deviations.
99.7% of the values fall within three standard deviations.
Select the group comparisons
The sample standard deviation of a group is an estimate of the population standard
group. The standard deviations are used to calculate the confidence intervals and th
Select the graphs
sample standard deviations result in less precise (wider) confidence intervals and lo
power.
Interpret the results
Analysis of variance assumes that the population standard deviations for all levels a
cannot assume equal variances, use Welch's ANOVA, which is an option for One-Wa
Key results
available in Minitab Statistical Software.

All statistics and graphs

Methods and formulas

T-value
The t-value in one-way ANOVA is a test statistic that measures the ratio between the
Methods and formulas
means and the standard error of the difference.

Interpretation
You can use the t-value to determine whether to reject the null hypothesis, which st
difference in means is 0. However, the p-value is used more often because it is easie
more information on using the critical value, go to Using the t-value to determine w
null hypothesis.

Minitab uses the t-value to calculate the p-value.

Minitab.com ● License Portal ● Store ● Blog ● Contact Us

Copyright © 2022 Minitab, LLC. All rights Reserved.

By using this site you agree to the use of cookies for analytics and personalized content. Read our policy OK

https://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/anova/how-to/one-way-anova/interpret-the-results/all-statis… 18/18