최솟값, 최댓값, 사분위수

Center and spread leave out the extremes. Knowing the smallest and largest values frames the range your data can possibly take. 🔭

The min is the lowest value and the max is the highest. They are your first guard against typos like a negative age or an impossible price.

Subtract min from max and you get the range, the total width of your data. It is simple but easily wrecked by a single extreme value.

Sort your values and cut them into four equal parts. The three cut points are your quartiles, and they describe the data far more robustly.

The first quartile Q1 is the 25th percentile. One quarter of your values fall below it, marking the bottom of the bulk.

The second quartile Q2 is the 50th percentile, which is exactly the median you already met. Half the data sits on each side of it.

The third quartile Q3 is the 75th percentile. Three quarters of values fall below it, so only the top slice lies above.

The quantile method gives you any cut point you want. Pass 0.25 for Q1 or 0.75 for Q3 directly on a column.

Subtract Q1 from Q3 and you have the IQR, the spread of the middle half. It ignores the extremes, so outliers cannot inflate it.

Min, Q1, median, Q3, and max form the five-number summary. Together they sketch the full shape of a distribution in just five values.

You do not compute quartiles by hand. The describe output lists the 25, 50, and 75 percent rows for every numeric column.

You want a spread measure that ignores extreme values entirely.

You can now read the edges with min and max and the inner structure with quartiles and IQR. Five numbers describe a whole distribution. 📊