การกระจาย: ความแปรปรวนและส่วนเบี่ยงเบนมาตรฐาน

Two columns can share the same mean yet look nothing alike. To tell them apart you measure their spread, how widely the values scatter. 📏

Spread starts with each point's distance from the average, its deviation. Small deviations mean tight data, large ones mean it sprawls.

Positive and negative deviations would cancel to zero. So we square each one first, turning every gap into a positive contribution.

The variance is the average of those squared deviations. A bigger variance means values stray further from the mean on average.

Squaring inflates the units too. Variance of prices in dollars comes out in dollars squared, which is impossible to read at a glance.

Take the square root of variance and you get the standard deviation. It lives in the same units as your data, so it finally makes sense.

Standard deviation is roughly the typical distance of a point from the mean. A small std means consistent values, a large one means volatility.

For roughly bell-shaped data, about 68 percent of values land within one standard deviation of the mean. It is a fast sanity check on range.

pandas divides by n minus one by default, the sample version. That small tweak corrects bias when your data is just a sample of a bigger group.

Standard deviation lets you compare consistency fairly. The product with the smaller std in delivery time is the more reliable one.

You do not pick between them by hand. The describe output already includes std for every numeric column you have.

You want a spread measure that uses the same units as the original data.

You learned that variance averages squared distances and standard deviation brings that back to real units. Together they reveal how much your data moves. 🌟