Data Visualizations

Main areas

1: Column selector panel

A list of the columns in the active worksheet.

2: Source data panel

The available input fields and any display options associated with the selected graph.

3: Candidates list

The complete list of all available graphs.

4: Canvas

The area where Minitab displays the graph previews.

Tips

Continuous variables

Enter one or more numeric columns that you want to graph.

Group variable

Enter a variable that defines the groups. Group labels are shown in the graph legend.

Y-scale

Continuous variables

Enter one or more numeric columns that you want to graph.

Group variable

Enter a variable that defines the groups. Group labels are shown in the graph legend.

Distribution

Continuous variables

Enter one or more numeric columns that you want to graph.

Categorical variables (optional)

Enter up to five columns of categorical data that define the groups. The first variable is the outermost on the scale and the last variable is the innermost.

Whiskers and outliers

Custom percentiles

With large data sets, where outliers are common, you can display custom percentiles instead of outliers to gather more information about the data. Custom percentiles occur outside of the interquartile box and typically occur in the tails of the distribution. In addition, lines are placed at the minimum and maximum values. By default, these percentile values are 0.5, 2.5, 10, 90, 97.5, and 99.5, but you can add, delete, or change them.

Y-scale

Variable display order

Minitab uses the terms "innermost" and "outermost" to indicate the relative position of the scales for multiple levels of groups displayed on a graph. For a horizontal scale, outermost refers to the scale at the bottom of the graph, and innermost refers to the scale farthest from the bottom, closest to the horizontal axis. For a vertical scale, outermost refers to the scale to the far left, and innermost refers to the scale closest to the vertical axis.

Continuous variables

Enter one or more numeric columns that you want to graph.

Categorical variables (optional)

Enter up to five columns of categorical data that define the groups.

Jitter individual symbols

If you have identical data values on your graph, individual symbols could hide behind each other. Choose this option to move symbols slightly to reveal overlapping points.

Y-scale

Variable display order

Minitab uses the terms "innermost" and "outermost" to indicate the relative position of the scales for multiple levels of groups displayed on a graph. For a horizontal scale, outermost refers to the scale at the bottom of the graph, and innermost refers to the scale farthest from the bottom, closest to the horizontal axis. For a vertical scale, outermost refers to the scale to the far left, and innermost refers to the scale closest to the vertical axis.

Categorical variables

Enter one or more numeric columns that you want to graph.

Summarized variables (optional)

Enter a numeric or text column that defines the bars in the bar chart.

Function

From Function, select the function of the Summarized variables. For example, if you select Maximum, Minitab defines the color for the bar chart based on the maximum value of the summarized variables in each bar. If you enter a text column in Summarized variables, you can only select Percent equal to specified values, Number of nonmissing values, or Number of missing values.

Overlay

When you select multiple categorical variables and you choose to overlay the graphs, you can choose how to display the overlaid bars.

Overlaid with evenly spaced bars

By default, Minitab displays the bars evenly along the axis. Categorical groups are displayed at different levels along the axis in the order in which they appear in Categorical variables.

Overlaid with clustered bars

Minitab displays categorical groups as clustered bars and includes a legend. You can change the grouping by reordering the variables in Categorical variables, or you can choose a different variable from Cluster variable.

Overlaid with stacked bars

Minitab displays the categorical groups as stacked bars and includes a legend. You can change the levels at which the categories appear by reordering the variables in Categorical variables, or you can choose a different variable from Cluster variable.

Note

When you have more than one summarized variable and you choose to overlay the graphs, you have the option to cluster or stack the bars by the summarized variables.

Clustered by summarized variables

Stacked by summarized variables

Order bars by

Display counts as percent of total

When you do not enter Summarized variables, select this option to change the Y-scale type from count to percent.

Display function as percent of total

When you enter Summarized variables, select this option to change the Y-scale type to percent.

Same Y-scale

Make the Y-scale the same across multiple graphs.

Variable display order

Minitab uses the terms "innermost" and "outermost" to indicate the relative position of the scales for multiple levels of groups displayed on a graph. For a horizontal scale, outermost refers to the scale at the bottom of the graph, and innermost refers to the scale farthest from the bottom, closest to the horizontal axis. For a vertical scale, outermost refers to the scale to the far left, and innermost refers to the scale closest to the vertical axis.

Categorical variables

Enter one or more columns of categorical data that you want to graph.

Summarized variables (optional)

Enter one or more columns of summary values that you want to graph.

Display Type

Order slices by

Combine slices of this percent or less

Enter a minimum percentage for separate slices. Categories less than this percentage are grouped into a slice named Other.

Variables

You can graph the x and y-variables as individual pairs or you can graph every combination of the x-y variables. The y-variable is the variable that you want to explain or predict. The x-variable is a corresponding variable that might explain or predict changes in the y-variable. All columns must be numeric, and each x-y variable pair must have the same number of rows.

First, choose one of the following options.

Each Y versus each X

Select to display a separate graph for each possible combination of x and y-variables that you enter.

XY Pairs

Select to display a separate graph for each pair of x and y-variables that you enter. Each x-y variable pair must have the same number of rows.

Then, enter the variables.

Y variables

Select the variables that you want to explain or predict.

X variables

Select the variables that might explain or predict the changes in the Y variables.

Group variable

Enter a variable that defines the groups. Group labels are shown in the graph legend.

Log transformation

Variables

The y-variable is the variable that you want to explain or predict. The x-variable is a corresponding variable that might explain or predict changes in the y-variable. All columns must be numeric, and each x-y variable pair must have the same number of rows.

Y variables

Select the variables that you want to explain or predict.

X variables

Select the variables that might explain or predict the changes in the Y variables.

Gradient defined by mean

Select to define the gradient scale by the value of a third variable.

Gradient Type

Select the color scale for the bins.

Diverging

Bins with high values are red, and bins with low values are blue. In Gradient symmetric around value, enter a value to center the gradient scale at a specific value rather than the center of the frequency of the binned data.

Sequential from low to high

Bins with high values are dark blue, and bins with low values are light blue and light gray. You can use this option to highlight bins with more productivity or to maximize revenue.

Sequential from high to low

Bins with low values are dark blue, and bins with high values are light blue and light gray. You can use this option to highlight bins with low defect rates or to minimize cost.

Log transformation

Variables

The y-variable is the variable that you want to explain or predict. The x-variable is a corresponding variable that might explain or predict changes in the y-variable. All columns must be numeric, and each x-y variable pair must have the same number of rows.

Y variables

Select the variables that you want to explain or predict.

X variables

Select the variables that might explain or predict the changes in the Y variables.

Bubble Size

Enter a column that determines the size (area) of the bubbles.

Layout

Choose a layout option.

Separate graphs for each XY pair

Creates a separate scatterplot for each XY pair.

Overlay XY pairs

Each XY pair is overlaid on a single scatterplot.

Group variable

Enter a variable that defines the groups. Group labels are shown in the graph legend.

Log transformation

Continuous variables

Enter the columns that contain the data. You must enter at least two columns of numeric data. Each column must have the same number of rows.

Gradient Type

Select the color scale for the bins.

Automatic

When you select Automatic, the gradient type depends on the calculation of the Pearson correlation coefficients.

• If the coefficients are positive and negative, the Gradient Type is Diverging symmetric around 0.
• If the coefficients are all positive, the Gradient Type is Sequential from low to high.
• If the coefficients are all negative, the Gradient Type is Sequential from high to low.

Diverging

Bins with high values are red, and bins with low values are blue. In Gradient symmetric around value, enter a value to center the gradient scale at a specific value rather than the center of the frequency of the binned data.

Sequential from low to high

Bins with high values are dark blue, and bins with low values are light blue and light gray. You can use this option to highlight bins with more productivity or to maximize revenue.

Sequential from high to low

Bins with low values are dark blue, and bins with high values are light blue and light gray. You can use this option to highlight bins with low defect rates or to minimize cost.

Display correlation values

Select to display the value of the Pearson correlation coefficient within each rectangle of the graph.

Variables

Enter at least two columns of numeric data.

Row labels

Enter a column that contains labels for each series. Minitab uses this column to label a series on the plot when you hover over it with your cursor. If you don't enter a Group variable Minitab uses the label column to create a legend for the parallel plot.

Layout

Choose one of the following layout options.

Individual series

Creates a parallel plot that displays a series for every row in the worksheet. The worksheet must include at least two columns of numeric data.

Individual series by group

Creates a parallel plot that displays a series for every row in the worksheet and uses the same color for each level of the grouping variable. The worksheet must include at least two columns of numeric data and one column that contains the groups.

Summarized groups

Creates a parallel plot that displays the mean of each series for every group in the worksheet, not every row. The worksheet must include at least two columns of numeric data and one column that contains the groups.

Group variable

Enter a variable that defines the groups. Group labels are shown in the graph legend. You must enter a column when you select Individual series by group or Summarized groups.

Y-scale

Select how you want to display the y-scale.

Percent of range

Select to plot each variable with a unique y-scale. A series that contains all of the minimum or maximum values for each variable will be a horizontal

line.

Standard units

Select to plot each variable with a unique y-scale. The minimum and maximum values for each scale are the overall minimum and maximum z-score values from all the data that you entered converted to each variable's scale.

For example, the first variable's maximum value has a z-score of 2, while the other two variable's maximum value has a z-score of 1. The maximum value for each variable's y-scale is the value that corresponds to a z-score of 2.

Original data

Select to use a single y-scale that is repeated for each variable. The minimum and maximum values for the scale are the overall minimum and maximum values from all the data that you entered.

Sort variables from greatest to least variation in lines

If you select Standard units or Original data, you can sort the variables based on the variation. This can be useful when you have many variables and you want to see which one separates the series the most. If you do not select this option, Minitab sorts the columns in the same order as you enter them in the Variables dialog box.

Unsorted

Sorted

Categorical row variables

Enter up to five columns where the categories are represented as rows on the heatmap.

Categorical column variables

Enter up to five columns where the categories are represented as columns on the heatmap.

Summarized variables (optional)

Enter a numeric or text column that defines the color gradient of the rectangles in the heatmap. If you enter multiple columns, Minitab generates a separate heatmap for each variable that you enter.

Function

From Function, select the function of the Summarized variables. For example, if you select Maximum, Minitab defines the color gradient for the heatmap based on the maximum value of the summarized variables in each rectangle. If you enter a text column in Summarized variables, you can only select Percent equal to specified values, Number of nonmissing values, or Number of missing values.

Gradient Type

Continuous variables

Enter one or more columns of time-ordered numeric data that you want to graph.

Time scale labels (optional)

Label the x-axis with values from a column that contain date/time, numeric, or text values for the scale. For example, on the following time series plot, the column specifies the shift and day.

C1-T
Shift and Day
S1D1
S2D1
S3D1
...

Layout

Set the following layout options.

Panel continuous variables

Columns in the Continuous variables input field appear on a single time series plot where all variables share a single X-axis and each continuous variable appears in its own panel.

Overlay continuous variables

Columns in the Continuous variables input field are overlaid on a single time series plot.

Y-scale

Select how you want to display the y-scale.

Original data

Select to use a single y-scale that is repeated for each variable. The minimum and maximum values for the scale are the overall minimum and maximum values from all the data that you entered.

Percent of range

Select to plot each variable with a unique y-scale. A series that contains all of the minimum or maximum values for each variable will be a horizontal line.

Log scale

A logarithmic scale linearizes logarithmic relationships by changing the axis, so that the same distance represents different changes in value across the scale. This option is available only for positive data.