ENGI 4421 - First MINITAB Tutorial

^®

Place data in a Minitab worksheet,
sort the raw data into ascending order,
produce descriptive statistics,
export Minitab output,
produce a bar chart,
produce a histogram,
produce a frequency table,
produce a pie chart of categories and
produce a bar chart of categories.

From the course text, (Navidi, second edition), table 1.2 p. 20:

The data set below consists of observations on particulate matter emissions x (in g/gal) for 62 vehicles driven at high altitude.

7.59	6.28	6.07	5.23	5.54	3.46	2.44	3.01	13.63	13.02
23.38	9.24	3.22	2.06	4.04	17.11	12.26	19.91	8.50	7.81
7.18	6.95	18.64	7.10	6.04	5.66	8.86	4.40	3.57	4.35
3.84	2.37	3.81	5.32	5.84	2.89	4.68	1.85	9.14	8.67
9.52	2.68	10.14	9.20	7.31	2.09	6.32	6.53	6.32	2.01
5.91	5.60	5.61	1.50	6.46	5.29	5.64	2.07	1.11	3.32
1.83	7.56

[The simplest method to import these data into Minitab is to copy and paste from this link.]

Produce summary statistics of the data.
Construct a bar chart of the data, using class intervals of equal width, with the first interval having lower limit 1.0 (inclusive) and upper limit 3.0 (exclusive).
Modify the histogram of the data, combining the class intervals for [11, 15) into one class and the class intervals for [15, 25) into one class
Construct a tally chart for the following qualitative descriptions of emissions:
low: [0, 5);
moderate: [5, 10);
high: [10, 15);
very high: >= 15.

Open Minitab. Double click the Minitab icon on the desktop:

OR click on the search box next to the "Start" button in the lower left corner, then type in Minitab.
You should find Minitab 21 Statistical Software.
After first use, there may be a quick link to Minitab 21 on your Start Menu.

The upper large window is the output pane,
while you can scroll through your data in the data pane below it.
You can resize these panes by dragging on the edge of a pane.
To the left of these panes is the navigator pane.
Across the top are the menus and toolbars.

There is an excellent introductory tutorial to Minitab, at quickstart.minitab.com/.

Data Entry

Click on the grey cell just below 'C1' in the Data window.
Name that column by typing an appropriate label there, ending with the 'Enter' key. 'PM Emissions' is now an alias for 'C1'

From the link above open the plain text file "NavidiT1-2.txt" containing the data, using a simple program such as 'Notepad'. Right click anywhere in the main window and click on 'Select All' (or click on the menu item 'Edit' and then click on 'Select All' or press 'Ctrl A').

Right click anywhere in the main window and click on 'Copy'
(or click on the menu item 'Edit' and then click on 'Copy',
or press 'Ctrl C', or press 'Ctrl Insert').

Back in Minitab, click on the top white cell, just below the column label,
then right-click to bring up the pop-up menu and
click on 'Paste Cells'
(or press 'Ctrl V' or 'Shift Insert')

The data are now in column C1

Sorting Data

By itself, this column of raw data is not very helpful
as we try to grasp the overall picture of PM emissions.

One way to improve visibility is simply to rearrange these data into ascending order.

On the main menu bar near the top left corner of the Minitab window,
click on 'Data', then on 'Sort...'

A dialog box appears.
In the left pane all of the columns
of numeric data are displayed
- in this case, column 'C1 - PM Emissions' only.

Click on 'C1' in the box on the left,
then click the 'Select' button.
'PM Emissions' now appears in the
top right pane ('Columns to sort by:') in level 1.

[Alternatives are to double click 'C1'
in the left pane or
to type 'C1' manually in Level 1
of the top right pane.]

Keep the "Increasing" item as is and
accept the other defaults:

Columns to Sort: All columns
and
Storage location for the sorted columns:
At the end of the current worksheet.

Click the 'OK' button at the foot of this window.

The data, sorted into ascending order, now
appear in column 2 of the data pane,
alongside the raw non-sorted data.

You can edit the alias for column 2 if you wish.

Take a moment to scroll down through the
data pane, to gain some appreciation for the data.

Summary Statistics

On the main menu bar near the top of the Minitab window, click 'Stat' then,
on the successive pop-up menus, hover on 'Basic Statistics' then click on 'Display Descriptive Statistics...'.

In the new dialog box, select column 'C1'
(either click on 'C1' in the left pane then click on 'Select'
or double click on 'C1' in the left pane)

This will place 'C1' in the top right pane
('Variables:').

Now click on the 'Statistics...' button.

By default, the summary statistics that Minitab displays are those indicated by the checked boxes.

Uncheck 'SE Mean' and 'N missing'

Then click on the 'OK' button to return to the previous dialog box.

Click on the 'Graphs' button to bring this dialog box up.

Check the second and fourth boxes, as shown.

Then click on the 'OK' button to return to the previous dialog box.

Click on the 'OK' button of the main dialog box.

A table and two graphs appear in the output pane.
You will need to scroll down in the output pane to see the boxplot.

The summary statistics that we chose to display for the particulate matter emissions of the sample of 62 cars are now displayed.

Now click on the bar chart. Scroll down in the output pane if necessary to bring the full bar chart into view.

What Minitab calls a “histogram” is not a histogram! It is a bar chart.

The probability curve of a normal distribution that has the same mean and variance as the data is superimposed on this diagram. Clearly our data set was not drawn from any normal distribution. A substantial portion of the area under the normal curve extends into negative values of 'PM Emissions', which are impossible.

Now scroll down to the boxplot graph.

The line inside the box is the median

The upper and lower sides of the box are at
the upper and lower quartiles respectively.

The whiskers extend to the last observation
before the inner fences.
Minitab does not plot the locations of the fences.

This sample has four outliers, all off the upper end.
By default, Minitab plots mild and extreme outliers using the same symbol (asterisk).

We can tidy this display up.

The gridlines are present
but are very faint.

We would also like a different
colour for the box.

Right click anywhere in the boxplot
or
click on the boxplot then on the down
arrow at the top right corner of the plot.

On the pop-up menu,
click on 'Edit Graph...'

In the new 'Edit Graph' box,
there is a toolbar row at the top.

Click on the down arrow to the right of
the 'Select Item' box.

From the pull-down menu,
scroll down to Y Gridlines and click.

Now click on the 'Edit" button
just to the right.

In the new 'Edit Gridlines' dialog box,
in the group 'Major Gridlines' on the left,
click the radio button 'Custom'.

Next click the right arrow beside 'Type'
to pull down a set of choices.

Click on the second option (dashed line).

Next, change the colour
from 'Automatic' to 'Medium Gray'.

This should be sufficient, but
we can make the gridlines even more obvious.

In the 'Size' pull-down menu,
select '2'.

then click 'OK'.

Back on the 'Edit Graph' window,
click anywhere in the blue box of the boxplot.

At the top, the default edit option changes
to 'Interquartile Range Box'.

Again click on the edit button
just to the right.

In the 'Edit Interquartile Range Box'
dialog box,
click the radio button 'Custom'.

Pull down the menu beside 'Background color:'
and change the box background to
a colour of your choice.

If you wish, you can also add hatching
from the 'Type' pull-down menu.

Then click 'OK'.

By default Minitab does not display the mean on a boxplot.

In order to force the display of the mean,
on the toolbar at the top of the 'Edit Graph' window,

pull down the menu for 'Add Item'

and select 'Data Display'.

In the 'Add Data Display' dialog box,
check the box for 'Mean symbol'.

Click 'OK' to return to the
'Edit Graph' dialog box.

Back in the 'Edit Graph' dialog box,
by clicking on other elements of the plot
then on the 'Edit' button at the top,
you can customize other aspects of the boxplot
(such as the title and the axis label).

Click 'OK' to exit from the
'Edit Graph' dialog box and to
return to the output pane.

The output pane displays the
edited graph.

Leave the cursor resting on any part of the box.

A pop-up tool tip provides some numerical information about the boxplot.

Exporting Minitab Output

Previous versions of Minitab had a “report pad” to which output could be exported to an associated .rtf file that can be edited in Word.

In version 21, click on the output pane to ensure that it is the active pane.

From the file menu, select "Save Current Output As..."

Near the foot of the "Save As" dialog box,
use the pull-down menu to change the
file type from html to rtf.

You can edit your saved file in Word.
You may have to do some re-sizing,
as the default fonts are very small.

In Word, save your edited file
as a standard .docx file.

Bar Charts

On the main menu bar,
Click on 'Graph', then
click on 'Histogram...'.

In the dialog box,
accept the default
(click on the 'OK' button).

In the new dialog box,

select 'C1 PM Emissions'
(double click on it)

and click the 'Scale...' button.

In the new dialog box,

Click on the 'Gridline' tab,

Ensure that the first box
'Y major ticks'
is checked,

then click the 'OK' button

Click the 'Labels...' button.

In the new dialog box,

in the first tab
'Titles/Footnotes'
enter some appropriate text.

Click the 'OK' button.

Back in the main dialog box,
click the 'OK' button.

The following bar chart appears in a new tab in the output pane.
As noted earlier, this is not a histogram!

These default intervals (bins) are centred on 0.0, 2.5, 5.0, ... , 20.0 and 22.5.
The question requires intervals with boundaries at 1, 3, 5, ... , 23 and 25.

Right click anywhere on the plot.

In the pop-up menu,

click on 'Edit Graph...'.

In the new 'Edit Graph' window, click on one of the number labels.
The text in the menu pane at the top of the window changes
to 'X Scale'.
Click on the edit button to the right of that pane (highlighted here).

In the 'Edit Scale' dialog box,

Click on the third tab 'Binning',

In the upper group 'Interval Type'
click on the radio button 'Cutpoint'

In the lower group 'Interval Definition'
click on the radio button
'Midpoint/Cutpoint positions

Click in the pane and
manually type the class boundaries.

Click the 'OK' button.

The bar chart has been modified to a more reasonable form:

Now click on any bar.

The menu box now shows 'Bars'.

Click on the 'Edit' button.

In the 'Edit Bars' dialog box,

In the 'Attributes' tab,

in the left group
'Fill Pattern',
you may adjust the colour
and hatching of the bars
as you wish.

When done, click the fourth tab
'Binning'.

Now lets us combine some low frequency bins in the long right tail together.

In the lower group
'Interval Definition',

In the pane delete the numbers
13, 17, 19, 21 and 23

Then click the 'OK' button.

In this bar chart

the frequencies are displayed by the heights of the bars.
This gives a very misleading impression that there are more observations in the right tail than there really are.

Histograms

We need to replace this bar chart by a histogram, in which the frequencies are proportional to the areas of the bars, not their heights. The vertical axis needs to be changed from frequency to density.

Again we can use the 'Edit Graph' > 'Edit' feature.

There is also a nice shortcut:
In the 'Edit Graph' window, position the mouse cursor over any number label on the vertical axis and double click.

In the 'Edit Scale' dialog box,

Click on the third tab 'Type',

Click on the radio button 'Density',

click on the 'OK' button

and click on the next 'OK' button
to exit from 'Edit Graph'.

The histogram is now correct. [I also changed the bar colour, hatching and gridlines]

As we did with the descriptive statistics, we can save this tab of the output pane to a .rtf file for later editing in Word.

Of course, we should save our work.

In the main 'File' menu,
click on 'Save Project As...'.

Choose an appropriate directory and file name.

Tally (Frequency) Table

On the main menu bar near the
top left corner of the Minitab window,

click on 'Data',

then on 'Recode',

then on 'To Text...'

In the left pane, double click on 'C1 PM Emissions' to bring it into the top right pane.
In the mid-right pane 'Method:', click on the down arrow at the right and select 'Recode ranges of values' from the drop down menu.

This opens a new table. Enter the values and labels as shown.
The tab key can be used to move from cell to cell during this data entry.

Endpoints cannot be left blank. Choose boundaries that encompass all the data.
Accept the other defaults for 'Endpoints to include', 'Storage location ...' and 'Show summary table'.
Click 'OK'

The first few rows of the data pane
should now look like this,

while the output pane should now
have a new tab 'Recode'.

This provides a simple
frequency table for the four
categories of PM emissions.

On the main menu bar near the top left corner of the Minitab window,
click on 'Stat', then on 'Tables', then on 'Tally Individual Variables...'

In the resulting dialog box,

double click on
'C3 Recoded PM Emissions'
to select it into the
right pane 'Variables',

check all four boxes

and click 'OK'.

An extended tally table appears as a new tab in the output pane.

Unlike the case in many earlier versions of Minitab, the levels are listed in the correct numeric order.

'Count' = frequency

'Percent' = relative frequency

'CumCnt' = cumulative frequency

'CumPct' = cumulative relative frequency

One can then proceed to generate graphical summaries in various forms.

Pie Charts

To create a pie chart
based on our tally table,

click on 'Graph',
then 'Pie Chart...'

In the resulting dialog box,

click the radio button for
'Chart counts of unique values'
and

into the pane 'Categorical variables'
select from the left pane
'C3 Recoded PM Emissions',
then

click the button 'Labels...'

In the new dialog box:

In the first tab 'Titles/Footnotes",
you can add such labels as you wish.

Select the tab 'Slice Labels'
and check whichever boxes you wish.

Then click OK on each of the two dialog boxes.

A pie chart appears in a new tab of the output pane.

By clicking on various parts of the chart, one can edit its appearance as needed.

The font in the labels may need enlarging. Here is the pie chart after some further editing.

Bar Charts

To create a bar chart
based on our tally table,

click on 'Graph',
then 'Bar Chart...'

In the resulting 'Bar Charts' dialog box,

from the drop-down menu 'Bars represent',
accept the default
'Counts of unique values',

accept the default button 'Simple',
and
click the 'OK' button.

In the left pane of
the new dialog box,
double click on
'Recoded PM Emissions'

This choice appears in the right
pane 'Categorical variables'.

The six buttons allow for some finer control.

Just click 'OK'.

A bar chart (not a histogram) is produced.

Experiment with various settings to improve its appearance.

Don’t forget to save your work (if you want to keep it).

From the main 'File' menu, click on 'Exit' to terminate the Minitab program.

Return to the index of demonstration files

To the next tutorial

Return to your previous page

Dr. G.H. George