Center of a Distribution vs Spread of a Distribution

Center is one of four things a description of a quantitative distribution has to cover, alongside shape, variability, and unusual features such as outliers, gaps, and clusters. Two statistics report it. The mean $\bar{x}$ (x-bar) is the balance point of the values, and the median is the middle value once they are sorted. Shape picks between them, because the mean follows a long tail and the median mostly ignores it.

Take 4, 5, 5, 6, 6, 30. The mean is $56/6 = 9.33$ and the median is 5.5. Five of the six values sit between 4 and 6, so 5.5 describes them and 9.33 describes none of them. The median is the honest center here, and the matching move is to report the IQR beside it rather than the standard deviation.

"The center is halfway between the smallest and largest value." That is the midrange, $(4 + 30)/2 = 17$, and no observation in the set sits closer than 11 units to it. The midrange reads two values and throws the rest away, so it moves whenever an extreme value does: push the 30 out to 60 and the midrange jumps from 17 to 32, while the mean shifts from 9.33 to 14.33 and the median does not move at all.

Both standard measures can fail together. In 2, 3, 3, 4, 18, 19, 19, 20 the mean is 11 and the median is also 11, and not one of the eight values sits closer than 7 to that number. Two clusters with a gap between them have no single typical value, so name the clusters and describe each one instead of reporting a center the data never produced.

Center is examined in topic 1.6 of the Fall 2026 course and again in topic 1.9 when you compare distributions. The framework states the pairing rule outright: the median and the IQR are resistant because outliers barely affect them, while the mean, the range, and the standard deviation are not.

Spread, also called variability, is how far apart the values in a distribution sit. The Fall 2026 course names three measures. The range is maximum minus minimum. The interquartile range is $\text{IQR} = Q_3 - Q_1$, the width of the middle half. The standard deviation $s$ is a typical distance of a value from the mean. Report the one that matches your center: the mean travels with $s$, the median travels with the IQR.

Two five-value sets, 0, 5, 5, 5, 10 and 0, 0, 5, 10, 10, share a mean of 5, a median of 5, and a range of 10, and they do not look alike. The first has $s = 3.54$ and $\text{IQR} = 5$. The second has $s = 5.00$ and $\text{IQR} = 10$, because everything in it has been pushed out to the ends.

"Both sets run from 0 to 10, so they are equally spread out." The range knows only the two extremes, so it cannot tell those sets apart, and it can never shrink when you add another observation. The IQR and the standard deviation read the values in between, which is exactly why they separate 3.54 from 5.00 where the range reports 10 twice.

One caution on the IQR. This site and the TI-84 locate quartiles by the median-excluded rule, dropping the overall median from both halves when $n$ is odd. Other software uses a different rule and can return a slightly different $Q_1$ and $Q_3$ from identical data, so a mismatch with a spreadsheet is usually a convention difference rather than an arithmetic error.

Spread appears in topic 1.6 of the Fall 2026 course as one of the features a description has to cover, and in topic 1.7 as a calculation, where all three measures are named. Topic 1.9 asks you to compare it across groups, and that comparison has to be written explicitly: say that one distribution is more variable than the other and quote both numbers, rather than describing each group in its own sentence and leaving the reader to do the subtraction.