Plot: Histogram

Selecting Plot: Histogram in the Graph Options dialog is one way to access the options described here. Use these options to customize your 2D histogram or to customize the classes on the x- and y-axes of your 3D bivariate histogram. These options are also used to include some useful statistics in the footnote and display the axes labels in a special format (representation, such as Scientific, Currency, etc.) in your graph.

Plot. Use the Plot drop-down box to select the plot to which you want to make changes. If you have several plots in one graph, assign names to each plot using the Name field in the Plot: General options pane.

Statistics. Click the Statistics button to display either the Statistics (2D Histogram Plots) or the Statistics (Bivariate Histogram Plots) dialog in which you can choose to include a variety of statistics in your graph.

Type. The Type group is only available for 2D Histograms. Use these options to control the display of your graph.

Histogram; show. Select the Histogram;show option button and then choose either Standard, Cumulative (the consecutive columns in the resulting histogram represents cumulative frequencies (i.e., the frequencies of all preceding categories) instead of the raw data), Hanging Bars, or Stacked from the corresponding drop-down box to control the display of your graph.

Pie chart. Select the Pie chart option button display your graph as a pie chart.

Pareto chart. Select this check box to sort the data in Pareto chart form. Note that this check box is available only when the Histogram; show option is set to Standard or when the Pie chart option is selected.

Intervals/X intervals/Y intervals. Use the options in the Intervals (for 2D Histograms), X intervals and Y intervals (for Bivariate Histograms) group boxes to change categorization of the X and Y variables. You can choose from the following methods of categorization (for more details of these methods, see Method of Categorization).

Integer mode. Select the Integer mode option button in order for each encountered value of the selected variable to be truncated to an integer value and create a category; several values can be truncated to the same integer value, thus falling into the same category.

Unique values. Select Unique values to create one category (or graph in the case of categorized graphs) for each unique value. The unique values can be either floating point or integral, and STATISTICA can handle up to 255 unique values.

Categories. Select the Categories option button to divide the entire range of values of the selected variable (from minimum to maximum) into the requested number of equal length intervals (Categories).

Boundaries. Select the Boundaries option button to create your own boundaries for the intervals. The Boundaries for the intervals (categories) are defined by the values you supply when you click the Specify Boundaries button; these intervals can not be of equal size.

Codes. Select the Codes option button to decide the categories according to the specified codes of the selected variable; the number of categories will be equal to the number of codes that you select. Note that to specify the codes, click the Specify Codes button.

Multiple subsets. Select the Multiple subsets option button and then click the Specify Subsets button to define the selection conditions that will create specific categories; more than one variable can be used to define a category. Note that if this categorization is used, then the current selection of the respective categorical variable will be ignored.

Change Variable/Specify Boundaries/Specify Codes/Specify Subsets. Note that the name of this button changes depending on the Interval option button that is selected. Click the Change Variable button to display a standard variable selection dialog, which is used to change the selection of the grouping variable to which the Intervals options apply. Click the Specify Boundaries/Codes/Subsets button to display various dialogs in which you enter the boundaries, codes, or subsets that are used to create user-defined categories.

Properties. The options in the Properties group are used to select a special format (representation) in which to display the X and Y labels in your graph. Most of the options available for this purpose have more than one representation, except the General and Percentage options (see Value Format). Some options also allow you to specify decimal information.

Bar width. This option is only available for 2D histograms. In the Bar width field, enter the percentage of bar width to be displayed in the graph.

Intervals value format. This option is only available for 2D histograms. Click the Intervals value format button to display the Value Format dialog; the display format chosen in that dialog will be used to label the x-axis of the 2D histogram.

Scale X. From the drop-down list, you can select Logical or Coordinate. Histograms and can be scaled in one of these two ways. Logical scaling is used when you want to compare the number of elements from different categories or ”bins,” and the values along the x-axis are usually textual rather than numeric. In this case, the location of each bar on the X-axis is arbitrary, and the bars are equally spaced and have uniform width. A histogram of MALE vs. FEMALE is a good example of logical scaling. The value ”MALE” or ”FEMALE” has no natural numeric value on the X-axis. The two bars are placed arbitrarily on the X-axis, and each bar has the same width. Coordinate scaling is used when the x-axis values do have some specific numeric meaning, such as using a histogram to display distribution of continuous data. Because the x-axis has some specific numeric meaning, you can calculate fits for coordinate scaling, but cannot use a fit for logical scaling.

The Coordinate setting is of most use when selecting the Boundaries option button from the Intervals group. When defining boundaries for histograms, you can specify a set of non-uniform ranges of data to include in each of the histogram bins. For instance, if you want to create a histogram with the boundaries 0-1, 1-100, 100-500, these boundaries will display three histogram bars to represent the three sets of boundaries. If logical scaling is used, then the three bars will be the same width, and the X-axis will not represent a continuous axis. However, if using Coordinate scaling, the X-axis has to be continuous, and each histogram bar is drawn so that its upper/lower boundaries are displayed on axes at their respective coordinate values.

X Intervals value format. This option is only available for bivariate histograms. Click the X Intervals value format button to display the Value Format dialog; the display format chosen in that dialog will be used to label the x-axis of the 3D bivariate histogram.

Y Intervals value format. This option is only available for bivariate histograms. Click the Y Intervals value format button to display the Value Format dialog; the display format chosen in that dialog will be used to label the y-axis of the 3D bivariate histogram.

Relative cumulative counts. This option is available when the histogram Type is set to Cumulative. When this check box is selected, the proportions or percentages of the observations are shown on the Y axis.

Percentage in category graphs relative to _ count. You can choose to display the Percentage in category graphs relative to the Total count or to the Category count.