In this section: |
How to: |
Binning is a powerful tool for analyzing your data using ranges that you define.
Binning enables you to determine the frequency of values across the entire range of values. It is used for analyzing a frequency distribution. With binning, you can create discrete buckets of continuous data that control how groups of your data display. In addition, binning gives you the ability to review trends and spot outliers.
For example, you can review the range of expenses incurred by households. In the following example, these ranges are represented by bins that are grouped by $1000. In this case, you can see that the largest number of households had the smallest expense (16) while the trend declines as the bin size gets larger. With binning, you can see the frequency of how often values in a range appear across the different groupings, as shown in the following image.
In InfoAssist, bins are automatically created for Histograms. You can change the size of your bins to meet the requirements of your data. For example, if your data has very large values (for example, in billions) you might want to create larger bins. You can also create bins manually in Vertical bar charts and reports.
Note: When creating a bin, the format of the bin and the value set for the Width of Bins must be compatible. If the Width of Bins is a large decimal value, define the field format to match the format of the field being converted so that the bins can be successfully generated with the appropriate numeric precision.
Binning is available for the different output types, including reports, charts, and visualizations, depending on how you choose to display and analyze your data.
Note: If you are working with a report, chart, or visualization where the bin is specified and therefore displays in the Query pane, the position of the Edit Bins field in the shortcut menu appears in a different location.
You can create bins in reports, which lets you view the detail behind a chart or histogram. You can optionally switch between report and chart as you perform your analysis. This allows you to review the actual values that fall into each bin.
How to: |
Histograms graphically represent the distribution of numeric data. They facilitate the identification and discovery of the underlying frequency distribution within a set of continuous data. You can use histograms to identify trends and illustrate categorizations, or groupings, also known as bins. For more information, see Binning.
Histograms use bins to group data. Bins allow you to establish a range of values for your data. For example, you can review how the bins are designated when you consider the age of everyone in your company. In the first table, you can review the bins (using the default value of 10) and counts for each bin.
AGE |
COUNT |
---|---|
0-9 |
0 |
10-19 |
5 |
20-29 |
26 |
30-39 |
351 |
40-49 |
460 |
50-59 |
310 |
60-69 |
285 |
70-79 |
22 |
80-89 |
3 |
Total |
1462 |
In the second table, you can review the same data for the bins (with a bin width of 15) and counts for each bin. Notice that as the bin size gets larger, more employees fall into these different ranges.
AGE |
COUNT |
---|---|
0-14 |
0 |
15-29 |
31 |
30-44 |
611 |
45-59 |
510 |
60-74 |
296 |
75-89 |
14 |
Total |
1462 |
When you create a histogram in either Chart or Visualization mode, a bin is created automatically for the measure you select. When working with bins, you can change the width of the bin by editing it. In the Query or Data pane, right-click a bin field and click Edit Bins to change the value that dictates the width or format of the contents of the bin.
The bin value is designated as a dimension field, since it is a limited field with a discrete set of possible values that was created from a field with an unlimited, continuous range of values. The measure displays as a count (.CNT) field and the related bin is created in the Query pane. It is also placed in the Data Pane for future use.
Note: This process produces the same results as if you had created the histogram manually, however it automatically converts the measure to a count (.CNT) field. It also creates the bin for you, placing it on the x-axis, accordingly.
You can optionally edit the bin to change its width or format. For information on building a histogram manually, in Chart mode, see Binning.