THE STATISTICS PLAYGROUND
Every dataset
has a shape.
Make yours visible. Explore the patterns, find the center,
and discover what your numbers are telling you.
02
See the distribution
Mean Median
03
Explore percentiles
A percentile locates a value in the sorted data.
—
Calculated by linear interpolation at rank (n − 1) × p / 100. Different software may use different conventions.
Frequency table
Same bins as the chart| Value interval | Count | Share | Cumulative |
|---|
Intervals include the left edge and exclude the right. The final interval includes both edges.
A FIELD GUIDE TO THE SHAPES
Look beyond the average.
The same average can hide very different stories. Watch the shape, center, and spread together.
The familiar bell
Values cluster near the center, with fewer toward either end. A normal distribution is symmetric; in an ideal normal model, about 68% of values fall within one standard deviation of the mean.
An even playing field
Equal-width intervals are equally likely in a uniform model. A finite random sample may look uneven. Our sample is deliberately balanced to make the flat shape easy to see.
A tail tells a story
A right-skewed distribution has a longer tail toward larger values. High values pull the mean upward; the median is usually less affected. Income and waiting times often show this pattern.
Two peaks, two groups?
Two clusters can create two peaks. That may suggest a mixture of groups or processes, but the shape alone does not prove why. Bin choices can reveal or obscure those peaks.
Try this: Load the skewed sample and add 200. Compare the mean and median. Then change the bin count—your data stays the same, but its visual story changes.