Add Maple
← All help articles

Number Binning

Number binning allows you to group numeric values into ranges for easier analysis and visualization. Instead of analyzing every individual number, you can group them into meaningful ranges like age groups, income brackets, or satisfaction levels.

Accessing Number Binning Settings

You can configure number binning in two ways:

Method 1: Column Settings (Recommended)

  1. Click on the column header of a numeric column
  2. Select "Column Settings" from the dropdown menu
  3. Scroll to the "Binning" section

Method 2: Manage Columns

  1. More → Project Settings from the dashboard
  2. Select your numeric column
  3. Scroll to the "Binning" section

Binning Options

Auto (Recommended)

Uses the Freedman-Diaconis rule to create optimal bins based on your data distribution. You can optionally suggest a number of bins, and the system will approximate to it.

Features:

  • Optimal bin width: Calculated using statistical best practices
  • Adjustable bin count: Use the slider to suggest more or fewer bins
  • Outlier handling: Option to collapse extreme outliers into separate bins

Best for: Most use cases where you want statistically optimal grouping Example: Age data might be binned into 18-25, 26-35, 36-45, etc.

Equal Frequency

Creates bins where each bin contains approximately the same number of data points.

Features:

  • Balanced distribution: Each bin has similar counts
  • Adjustable bin count: Choose how many bins to create
  • Good for skewed data: Works well with uneven distributions

Best for: When you want equal representation from each range Example: Income data where you want equal numbers in each income bracket

Fixed Width

Creates bins of consistent width that you define.

Features:

  • Consistent intervals: All bins have the same width
  • Adjustable width: Set the exact width for each bin
  • Predictable ranges: Easy to understand and interpret

Best for: When you need consistent intervals for comparison Example: Test scores binned into 10-point ranges (0-10, 11-20, 21-30, etc.)

Custom

Define your own bin boundaries manually with full control over ranges and labels.

Features:

  • Custom ranges: Set exact start and end values for each bin
  • Custom labels: Give meaningful names to each bin
  • Flexible boundaries: Create bins of different sizes
  • Gap detection: System warns about gaps between bins

Best for: When you need specific business-defined ranges Example: Age groups: 18-24 (Young Adults), 25-34 (Early Career), 35-44 (Mid Career), etc.

Live Preview

The binning interface shows a live histogram preview so you can see exactly how your data will be grouped before applying changes. This helps you:

  • Verify the number of bins created
  • Check that ranges make sense for your data
  • See the distribution of values across bins
  • Adjust settings until you're satisfied

Examples

Example 1: Age Groups

Scenario: You have age data and want to create age groups for analysis Solution: Use Custom bins with ranges like 18-24, 25-34, 35-44, 45-54, 55-64, 65+ Result: Meaningful age groups instead of individual ages

Example 2: Income Brackets

Scenario: You have income data and want equal representation in each bracket Solution: Use Equal Frequency with 5 bins Result: Five income brackets with similar numbers of people in each

Example 3: Test Scores

Scenario: You have test scores and want consistent 10-point ranges Solution: Use Fixed Width with 10-unit width Result: Bins like 0-10, 11-20, 21-30, 31-40, etc.

Example 4: Likert Scale Analysis

Scenario: You have 1-5 Likert scale data and want to group responses Solution: Use Custom bins: 1-2 (Low), 3 (Medium), 4-5 (High) Result: Simplified three-category analysis

Tips for Number Binning

Choose the Right Method

  • Auto: For most cases where you want optimal statistical grouping
  • Equal Frequency: When you need balanced representation across ranges
  • Fixed Width: When you need consistent intervals for comparison
  • Custom: When you have specific business requirements or meaningful ranges

Consider Your Data Distribution

  • Normal distribution: Auto mode works well
  • Skewed data: Equal Frequency or Custom might be better
  • Sparse data: Custom bins can help group sparse values
  • Dense data: Auto or Fixed Width work well

Think About Your Audience

  • Business users: Custom bins with meaningful labels
  • Statistical analysis: Auto mode for optimal grouping
  • Comparisons: Fixed Width for consistent intervals
  • Reporting: Equal Frequency for balanced representation

Advanced Features

Outlier Handling (Auto Mode)

When using Auto mode, you can enable "Collapse extreme outliers" to:

  • Identify extreme values using statistical methods
  • Group outliers into separate bins
  • Adjust sensitivity with the outlier threshold (k-value)
  • Prevent outliers from distorting your main analysis

Custom Bin Management

When using Custom mode, you can:

  • Add bins: Click "Add Bin" to create new ranges
  • Remove bins: Click the minus icon to delete bins
  • Fix gaps: Automatically adjust bin boundaries to eliminate gaps
  • Sort bins: Arrange bins by their range values
  • Create equal bins: Automatically generate equal-width bins across your data range

Common Use Cases

Demographic Analysis

  • Age groups: 18-24, 25-34, 35-44, 45-54, 55-64, 65+
  • Income brackets: Custom ranges based on your market
  • Education levels: Grouped by years or categories

Survey Analysis

  • Likert scales: Group 1-5 scales into Low/Medium/High
  • NPS scores: Group 0-10 into Detractors/Passives/Promoters
  • Satisfaction ratings: Custom ranges based on your benchmarks

Performance Metrics

  • Test scores: Consistent point ranges (0-20, 21-40, 41-60, etc.)
  • Sales performance: Custom ranges based on your targets
  • Customer ratings: Grouped into meaningful categories

Troubleshooting

My bins are too granular/fine

  • Try Auto mode with fewer bins
  • Use Fixed Width with larger width
  • Combine adjacent bins in Custom mode

My bins are too broad

  • Try Auto mode with more bins
  • Use Fixed Width with smaller width
  • Split bins in Custom mode

Some bins are empty

  • This is normal for sparse data
  • Consider using Equal Frequency for better distribution
  • Check if your data range is appropriate for the binning method

My data looks wrong

  • Ensure your column is detected as Numeric type
  • Check for non-numeric values in your data
  • Use Manage Columns to change the column type if needed

Outliers are distorting my bins

  • Enable "Collapse extreme outliers" in Auto mode
  • Adjust the outlier threshold sensitivity
  • Use Custom mode to manually handle outliers