MINITAB Tip Sheet 10: Test of One or Two Proportions

Department of
Mathematics and Statistics
MINITAB Tip sheet 10

Test of one or two proportions
Use this tip sheet if you want to test one proportion against a given value, or two
proportions against each other.
Screen shots have been taken from Minitab version 15.
One proportion test

For this example the dataset used is population2.xlsx which can be accessed from the Tip
Sheets webpage.
In a certain population, is thought that 40% of individuals are male. You are given a sample
of 100 individuals, with their gender recorded. You want to test the hypothesis that the
population proportion of males is greater than 40%.
The null and alternative hypotheses are:

Null hypothesis, H0: P = 0.4
Alternative hypothesis, H1: P > 0.4
To perform the test select Stat, then Basic statistics, then 1 proportion. A new window
will then be opened in Minitab as follows:
Enter the variable you are testing

into the Samples in columns box.
This section is used if you are

have the exact values for the
counts in each group.
Tick the box Perform hypothesis

test, and then type the proportion
for the null hypothesis in
Hypothesized proportion.
For more Tip Sheets, and datasets used within them, visit http://www.reading.ac.uk/statistics/tipsheets/
This project was funded by the generosity of alumni donations to the Annual Fund.
©University of Reading 2011 Page 1

By clicking the Options button the following window will appear:
If you need a confidence interval then

you can change the level here: this is
fine as it is.
Change the Alternative with the correct

alternative hypothesis you are testing.
Here this greater than.
The output given by Minitab is as follows:
Test and CI for One Proportion: Gender
Test of p = 0.4 vs p > 0.4
Event = Male
95% Lower Exact

Variable X N Sample p Bound P-Value
Gender 50 100 0.500000 0.413622 0.027
Interpretation: From the results we can see the p-value is significant as it lies below 0.05,
meaning that we reject the null hypothesis that the proportion of males is equal to 40%, in
favour of the alternative hypothesis that the population proportion of males is greater than
40%.
Two proportions test

The hypothesis we will be testing here is that the difference in two population proportions
is equal to zero, versus they are unequal.
In this example we will use the data population3.xlsx which can be accessed from the Tip
Sheets webpage. This datasets contains the results of a survey of a group of males and
females about whether they were in favour of the university library being open longer. The
two proportions test is a test to see if there is a difference in response, i.e. if males and
females have a similar view.
The hypotheses we are testing are:

Null hypothesis, H0: P1 = P2 (Males and females have equal proportions saying yes)
Alternative hypothesis, H1: P1 ≠ P2 (males and females have different proportions saying yes)

To generate the proportions select Stat, Basic Statistics then 2 proportions. You will then
be presented with the following window:
This selection is used for stacked

data. Enter the response variable
in the Samples box and the group
factor in the Subscripts box.
For unstacked data add each of

the columns into the Samples in
different columns section.
For summary values, i.e. the

numbers of trials and events, use
the First row to enter data for the
first sample, in the form of the
number of Trials and Events, and
the Second row for the second.
When clicking the Options button the following window will appear:
This is for selecting the size of the

confidence interval you wish to produce.
This section is for the difference you are

testing with the null hypothesis, so zero
for equal proportions.
This is for the alternative hypothesis you

are testing. In this case it is not equal.
The output given by Minitab is shown below:
Test and CI for Two Proportions: Male, Female
Event = Y
Variable X N Sample p
Male 10 20 0.500000
Female 18 20 0.900000
Difference = p (Male) - p (Female)

Estimate for difference: -0.4
95% CI for difference: (-0.655548, -0.144452)
Test for difference = 0 (vs not = 0): Z = -3.07 P-Value = 0.002
Fisher's exact test: P-Value = 0.014
* NOTE * The normal approximation may be inaccurate for small samples.

Interpretation: The p-value given for ‘test for difference’ is based on a Normal assumption,
which is likely to be invalid for small data sets. If Minitab produces the ‘*NOTE*’ about being
inaccurate for small samples we should instead look at the results for Fishers exact test,
which does not rely on a distributional assumption.
In this example the calculated p-value for Fisher’s exact test is 0.014. This value is less than
0.05 so we reject the null hypothesis of equal proportions at the 5% level of significance, in
favour of the alternative that the difference in proportions is non-zero. Looking at the
estimated proportions we can see that a higher proportion of females answered yes,
compared with males.
Extra information on interpretation

 P-value: When interpreting the p-value for a test, if the value is less than 0.05 then
the test is significant at the 5% level, and we would usually say there is evidence to
reject the null hypothesis. If the p-value is less than 0.1 but greater than 0.05 then
there is weak evidence in favour of the alternative hypothesis. Finally if the p-value is
greater than 0.1 then we would usually say there is no evidence to reject the null
hypothesis. Note: NEVER ACCEPT the null hypothesis and conclude it to be true as
this will be incorrect; we always reject or do not reject the null.
 We should report p<0.001 if Minitab gives 0.000 in the output for a p-value.
And finally ...

If you are unsure about the correct statistical methods to apply in the analysis of your data,
book an appointment for our Statistical Advisory Service (www.reading.ac.uk/Stats-Advisory).
We will be pleased to advise you on aspects of design, analysis and interpretation.
Portions of the input and output contained in this publication are printed with permission of Minitab Inc. All
material remains the exclusive property and copyright of Minitab Inc. All rights reserved.

MINITAB Tip Sheet 10: Test of One or Two Proportions

Uploaded by

Copyright:

Available Formats

MINITAB Tip Sheet 10: Test of One or Two Proportions

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MINITAB Tip Sheet 10: Test of One or Two Proportions

Uploaded by

Copyright:

Available Formats

Department of

Mathematics and Statistics

MINITAB Tip sheet 10

Screen shots have been taken from Minitab version 15.

One proportion test

The null and alternative hypotheses are:

Enter the variable you are testing

This section is used if you are

Tick the box Perform hypothesis

©University of Reading 2011 Page 1

If you need a confidence interval then

Change the Alternative with the correct

The output given by Minitab is as follows:

Test and CI for One Proportion: Gender

Test of p = 0.4 vs p > 0.4

95% Lower Exact

Two proportions test

The hypotheses we are testing are:

©University of Reading 2011 Page 2

This selection is used for stacked

For unstacked data add each of

For summary values, i.e. the

This is for selecting the size of the

This section is for the difference you are

This is for the alternative hypothesis you

The output given by Minitab is shown below:

Test and CI for Two Proportions: Male, Female

Difference = p (Male) - p (Female)

Fisher's exact test: P-Value = 0.014

* NOTE * The normal approximation may be inaccurate for small samples.

©University of Reading 2011 Page 3

Extra information on interpretation

And finally ...

©University of Reading 2011 Page 4

You might also like