Frequency Table Generation

Frequency Table Generation

A 2014 desktop project implementing frequency and cumulative-frequency tables, class intervals and optional Sturges-rule bin estimation with the algorithmic assumptions made explicit.

  • Project Features
  • Detailed guidance
  • Optional calculation of the class count with Sturges’ formula
  • Optional rounding of values
  • Sample dataset
  • Project Details
  • The “Round” checkbox determines whether values are rounded.
  • The “Find k with Sturges’ formula” checkbox controls the corresponding calculation.
  • The “S” button runs the calculation for a sample dataset.
  • The “Calculate” button performs the required calculations.
  • The “?” button opens the About window.

Statistical Basis of the Frequency Table

Although this project was originally implemented as a desktop application, the problem it solves is statistical summarization. A frequency table counts observations for values or class intervals; cumulative frequency maintains the running total across ordered classes.

For continuous or wide-range data, the class count can be selected explicitly. The application could also use Sturges' rule as an optional estimate:

k = 1 + 3.322 log10(n)

Here n is the number of observations and k is the approximate number of classes. It should be treated as a starting heuristic rather than a guarantee of the best binning for every distribution.

Algorithmic Flow

  1. Determine the minimum and maximum.
  2. Obtain the class count from the user or Sturges' rule.
  3. Derive the class width from range and class count.
  4. Assign each observation to the corresponding frequency counter.
  5. Build cumulative frequencies.
  6. Optionally round interval/display values.

The original Round, Find k with Sturges' formula, sample-data and Calculate controls exposed these algorithmic choices through the UI.

For broader theory, see Probability and Statistics: Distributions, Sampling and Regression.

What a Frequency Table Represents

A frequency table organizes how often observed values or class intervals occur in a dataset. Absolute frequency is the observation count, relative frequency is its proportion of the total, and cumulative frequency sums observations up to a given class. For continuous or wide-range numeric data, grouping values into intervals gives the tabular basis of a histogram.

This project used Sturges' rule as an optional estimate of class count. The rule does not guarantee an optimal binning for every distribution; sample size and distribution shape can make other bin-width methods more appropriate.

QR code for this page