2D Categorized Histograms - Advanced Tab

Select the Advanced tab of the 2D Categorized Histograms dialog to access the options described here.

Layout. Choose either a Separate plot layout (in which each subset of cases is displayed in a separate small graph) or an Overlaid plot layout (in which all subsets are overlaid in one graph and identified only by patterns and colors) for the categorized histograms.

Fit. You can choose to superimpose a distribution over your histogram. Select the type of distribution to fit from this list. For more details on fit types, see Fitted Functions for Histograms.

Variables. Click the Variables button to display the standard variable selection dialog, in which you can select the X and (optional) Y grouping variables and the dependent variable(s) to be displayed in the graph. If more than one dependent variable is selected, then a sequence of graphs (one for each dependent variable) will be produced using the same set of grouping variables. The selection that you make will then be displayed in the area of the dialog below the Variables button.

Note that the selected grouping variables do not have to be categorical variables (e.g., contain codes); you can use one of the methods of categorization in the X Categories or Y Categories group boxes to categorize continuous variables. The selection of grouping variables is not necessary if the categories are defined via the Multiple Subsets method.

Statistics. Include a variety of statistics as footnotes in the graph by selecting one or more of the statistics listed in this box.

Shapiro-Wilk test. Select this check box to include the Shapiro-Wilk W statistics for normality and its associated p-value.

Total count. Select this check box to include the total count or number of cases used in creating the histogram.

Descriptive statistics. Select this check box to include a variety of descriptive statistics including the total count, mean, and standard deviation as well as the minimum and maximum values.

Kolmogorov-Smirnov test. Select this check box to include the Kolmogorov-Smirnov d statistics (for normality), its associated p-value, and the Lilliefors statistic.

X-Categories / Y-Categories / Intervals. Select Integer mode, Unique values, or Categories to specify that method of categorization for each of the variables selected via the Change Variable button, or use the Boundaries, Codes, or Multiple subsets options. For more information about each of these methods of categorization, click on the links below:

Histogram options. The options in this group box determine the overall appearance of the histograms.

Cumulative counts. The consecutive columns in the resulting histogram will represent cumulative frequencies (i.e., the frequencies of all preceding categories) instead of the raw data when this check box is selected.

Breaks between columns. Select this check box if you want the histogram columns in the plot to be separated from each other. According to common convention, histograms of continuous variables should be displayed with no breaks (the breaks signify discrete values).

Y axis. Select different types of labels or scales for the vertical (y) axes in the histogram in this box. These options are not available when the Cumulative counts check box is selected. The available options are N, %, % & N, N & %.

If you select N, the histogram will show only the respective frequency counts for the left y-axis labels. Note that you can use the Show percentages check box to display relative frequencies (percentages of total n) at the top of each column in the histogram.

If you select %, the histogram will show the respective relative frequency counts (percentages) for the left y-axis labels. The percentages are computed as the relative frequencies with respect to the total n, and not the n in the respective group or category. Note that the values that will be displayed in the Graph Data Editor window (and that are actually plotted in the histogram) will still be frequency counts; however, appropriate text labels (percentage values) are used as Scale Values for the left y-axis. Therefore, you can easily later change to frequency counts for the y-axis labels. Also note that in order to change the step sizes (increments) for the y-axis, you need to change the respective Step value (to adjust the horizontal grid lines) as well as Skip values to display the respective percentage labels at the desired intervals.

If you select % & N, the histogram will show the relative frequency counts (percentages of total n) as labels for the left y-axis, and the frequency counts as labels for the right y-axis. Refer to the description of the previous selection for a description of percentage labels for the y-axis in histograms.

If you select N & %, the histogram will show the frequency counts as labels for the left y-axis, and the relative frequency counts (percentages of total n) as labels for the right y-axis. This is the reverse labeling as described in the previous paragraph.

Show percentages. Select this check box to display percentages (relative frequencies) at the top of each column in the histogram. This check box is disabled when the Cumulative counts check box is selected. The percentages are computed as the relative frequencies with respect to the total n, and not the n in the respective group or category.