Minitab produces two general types of graphs: the standard
graphing commands produce **High resolution graphs** which
use graphics drawing methods and appear in special graph windows
(but are not attached to the session file - so they have to be
saved and printed separately). You can also get somewhat clunky
graphs drawn in the Session window (so they print with the session
window) using the **Character graphs** commands (There is a
**Character graphs** submenu at the bottom of the **Graph**
menu). For information on saving the results, see VI.
Saving and printing your work

**Histogram:** Use **Graph>Histogram**. Select the
variable(s) by double-clicking in the list. The **Annotation**
button will let you write a title for your histogram. The **Options**
button will let you choose 1. the **type** of histogram (frequency,
relative frequency, cumulative, etc.) 2. Whether the bars are
labeled by the class **midpoints** or **cutpoints** (class
limits) 3. The **number** of classes or the values for either
the midpoints or the cutpoints.[**Automatic** is the default
- the program decides]. Any options you set will **remain in
place** for your session (until you Exit Minitab) unless you
change them.

**Boxplots:** Use **Graph>Boxplot**. The variable
to be plotted goes in Y - so the boxplot will be drawn vertically..
If several side-by-side boxplots (for categories) are desired,
the values must be stacked in one column, with a second column
containing an index identifying the category from which each value
comes.

**Normal probability plots:** Use **Graph>Probability
Plot... **Select the variable, leave the distribution choice
as "Normal".The graph will print with "confidence
bands" around the 45-degree line. The data values are on
the horizontal axis, the percentile labels (rather than z-values)
are on the vertical axis (the spacing is arranged to match equal
spacing of z-values).The "percentile" table in the Session
window shows where the percentiles **would** be if the data
**were** normally distributed (with the same mean and standard
deviation)

**Scatter plot:**Use **Graph>Plot**. You must specify
the variables for the vertical (response - y) and horizontal (predictor/explanatory
- x) axes by moving the cursor to the correct box (with mouse
or tab key) and either typing them in or double-clicking on your
choices. With the **Annotation** button you can give a title
to the graph and label the axes, with the **Frame** button
you can set the minimum and maximum x and y values for the axes.
The **Data display** options are used for having different
**groups of points **(data for males & females, perhaps)**
plotted with different symbols**: You must have a column that
indicates which group each row comes from (might be column full
of M's and F's) .Under "Data display" (in the window)
use the popup menus to select Display **symbol** for each **group**
and enter the name of the column with the group variable (the
column of M's & F's - or whatever) under "Group Variables"**.
**To select the symbols to use, click on the "Edit Attributes"
button - you can choose a symbol (circle, plus, star, etc.) for
each group in the popup menu of the window that opens.

**Stemplot: **Use **Graph>Stem-and-Leaf **. Select
the variable(s) by double-clicking in the list at the left of
the window. You can set the **increment** at 5 to get a split
stemplot. In the display, an extra column at the left of the display
shows the "depth" of the class - the distance (number
of items) from the **nearer **end (large or small) of the data.
The row with the median is marked in parentheses.This is really
a character graph - it shows up in the Session window - it's on
the main graph menu because people use it a lot.

**Dotplot**: Use **Graph>Dotplot**. Choose the variable(s)
you want by double-clicking on them in the window. If you are
plotting several variables and want them all on the same scale
(so you can compare them) choose the "Each column constitutes
a group" option. Use the "Title" box to title the
graph.

Choose the graph using the command as shown (the "Character graphs" option is at the bottom of the "Graph" menu list). The Stem-and-leaf and Dotplot commands always produce character graphs.

**(Grouped) Frequency distribution**: Use **Graph>Character
Graphs>Histogram**. Select the variable(s) by double-clicking
in the list. The **Interval** is the class width (class interval),
the **First midpoint** is the class midpoint of the first class.
The resulting printout does not show class limits but does show
class midpoints and frequencies.

**Ungrouped frequency distribution** (counts for different
values): Use **Stat>Tables>Tally** Choose the columns
you want by double-clicking on them. You can choose counts, percentages,
etc. by clicking on the boxes next to the options you want.

**Scatter plot**: Use **Graph>Character Graphs>Scatter
Plot**. You must specify the variables for the vertical (y)
and horizontal axes by moving the cursor to the correct box (with
mouse or tab key) and either typing them in or double-clicking
on your choices. With the " scale " option you can set
the graph scale, with the " annotate " you can label
the graph and axes.

Last modified 8/10/2000