Describing Data

Two numbers summarize most data sets: where the values cluster and how far they spread.

You specified 4,000 psi concrete, and two suppliers each send you five cylinder breaks. Both sets average exactly 4,000 psi, so by the average it's a tie. But Supplier A's breaks all land within 150 psi of that number, and Supplier B's miss it by as much as 700. You know which one you'd rather pour a column with. This Lesson gives you the numbers that tell them apart.

3,200 3,600 4,000 4,400 4,800 Cylinder strength (psi) x̄ = 4,000 psi Supplier A Supplier B
Five breaks from each supplier, on one strength scale

The mean is a balance point

The mean is the average you already know: add the values and divide by how many there are. The Handbook's "Dispersion, Mean, Median, and Mode Values" section gives it as

xˉ=1n∑i=1nxi\bar{x} = \frac{1}{n}\sum_{i=1}^{n} x_i

Here xˉ\bar{x} ("x-bar") is the mean of a sample of nn values, and Σ\Sigma means "add them all up." The Handbook capitalizes the symbols, as Xˉ\bar{X}. A whole population of size NN has its mean written μ\mu.

Here's a better picture than "add and divide." Put a one-pound weight at each value along a stiff, weightless ruler, and the mean is where the ruler balances. To see why, find each value's deviation, its signed distance from the mean. Supplier B's breaks of 3,400, 3,600, 4,000, 4,300, and 4,700 psi have deviations of −600, −400, 0, +300, and +700 psi. Those below the mean total −1,000, and those above total +1,000. Deviations from the mean cancel exactly, for any data set. Each one is that weight's turning effect about the balance point, so the ruler sits level.

3,200 3,600 4,000 4,400 4,800 Cylinder strength (psi) x̄ = 4,000 psi −600 −400 0 +300 +700 total −1,000 psi total +1,000 psi
Supplier B's deviations from its mean

Spread: the typical distance from the mean

You want the typical size of a deviation, and averaging them gives zero every time. So get rid of the signs by squaring:

(−600)2+(−400)2+02+3002+7002=1,100,000 psi2(-600)^2 + (-400)^2 + 0^2 + 300^2 + 700^2 = 1{,}100{,}000\ \text{psi}^2

Divide that total by the count, more or less, and you have the variance. Its units are square psi, which mean nothing to anyone, so take the square root to get back to psi. That's the standard deviation: square, average, root.

The "more or less" is the catch. The Handbook prints one version for a whole population and one for a sample, and σ\sigma and ss are their square roots:

σ2=1N∑i=1N(xi−μ)2\sigma^2 = \frac{1}{N}\sum_{i=1}^{N} (x_i - \mu)^2
s2=1n−1∑i=1n(xi−xˉ)2s^2 = \frac{1}{n - 1}\sum_{i=1}^{n} (x_i - \bar{x})^2

A whole population gives you the true mean μ\mu to measure from. A sample usually stands in for something bigger: those five cylinders stand in for all the concrete in the pour, and you don't know its true mean, so you measure from xˉ\bar{x} instead. But xˉ\bar{x} is a flattering choice, because it's the one point that makes these five squared deviations as small as possible. Measure from 4,100 psi instead of 4,000, and the total grows to 1,150,000 psi². So deviations from a sample's own mean run slightly small, and dividing by n−1n - 1 instead of nn corrects for it.

s=1,100,0005−1=524 psis = \sqrt{\frac{1{,}100{,}000}{5 - 1}} = 524\ \text{psi}

Supplier A works out to s=94s = 94 psi. There's what the average couldn't show you: the same mean, with over five times the scatter at Supplier B. Any data set deserves both questions: where do the values center, and how far do they spread?

Worked Example Weekly rainfall totals
1 of 2

A rain gauge records weekly totals of 1.2, 0.4, 2.1, 0.8, and 1.5 in. Treating these five weeks as a sample, find the mean and the sample standard deviation.

Whole population or sample?

So before you reach for a formula, ask one question: is this data set everything you care about, or a sample of something bigger? "The entire population," "all," or "every" point to NN. "A random sample" or "test results" point to n−1n - 1, and so does silence, since measurements are nearly always a sample of something bigger.

Treated as an entire population, Supplier B's breaks give σ=469\sigma = 469 psi instead of 524. The NN form always comes out smaller, and the gap fades as the data set grows.

Your calculator won't choose for you. One-variable statistics reports both standard deviations, usually labeled σx\sigma_x for the NN form and sxs_x for the n−1n - 1 form. Reading the right line is the whole job, and for a variance, square that line.

Checkpoint
1 of 2

A county's five bridges are the entire population a study cares about. Their deck ages are 12, 20, 25, 31, and 37 years. What is the population variance?

years²

When you only get the totals

Sometimes a lab report gives you only nn, Σx\Sigma x, and Σx2\Sigma x^2. The mean is immediate: xˉ=Σx/n\bar{x} = \Sigma x / n. For the spread, the sum of squared deviations can be rewritten using only those totals:

s2=∑x2−(∑x)2nn−1s^2 = \frac{\sum x^2 - \dfrac{(\sum x)^2}{n}}{n - 1}

Read the numerator as a correction. Σx2\Sigma x^2 measures every value from zero, so most of it just reflects the data sitting far from zero. The term (Σx)2/n(\Sigma x)^2/n is exactly that part. Subtract it, and the spread around xˉ\bar{x} is what's left. Supplier B has Σx=20,000\Sigma x = 20{,}000 and Σx2=81,100,000\Sigma x^2 = 81{,}100{,}000, so

81,100,000−(20,000)25=1,100,000 psi281{,}100{,}000 - \frac{(20{,}000)^2}{5} = 1{,}100{,}000\ \text{psi}^2

That's the same total as before. Those two big numbers nearly cancel, so keep every digit until after you subtract. The Handbook doesn't print this as a variance, so a question that needs it should state it, as the next one does.

Checkpoint

A lab reports only summary values for ten cylinder tests: n=10n = 10, Σx=320\Sigma x = 320 MPa, and Σx2=10,400\Sigma x^2 = 10{,}400 MPa². Using s2=[Σx2−(Σx)2/n]/(n−1)s^2 = [\Sigma x^2 - (\Sigma x)^2/n]/(n - 1), what is the sample standard deviation?

A.
4.00 MPa
B.
4.22 MPa
C.
17.8 MPa
D.
33.9 MPa

The middle value, the commonest value, and the width

The median is the middle value once the data are sorted. The Handbook's rule: with nn odd, it's item (n+1)/2(n + 1)/2, and with nn even, it's the average of items n/2n/2 and n/2+1n/2 + 1. Sort first, because the middle of an unsorted list is just whatever got written there. The mode is the value that occurs most often, and the range RR is the largest value minus the smallest.

Data often arrive rolled up in a frequency table, and that's where people stumble. Here are twenty inspected precast panels:

Defects found on twenty precast panels
Defects per panelPanelsCumulative count
077
1815
2318
3119
4120

The listed values run from 0 to 4, and the middle of that list is 2. That's the trap, because the median is the middle observation, not the middle row. A frequency table is a sorted list written compactly, so unroll it.

0 1 2 3 4 Defects per panel 1st 10th 11th 20th middle two observations
The same twenty panels, unrolled into a sorted line

With n=20n = 20, the middle observations are the 10th and 11th. The cumulative counts, a running total of the Panels column, say the first 7 are 0s and 8 through 15 are 1s, so both are 1, and the median is 1. The Handbook says nothing about frequency tables, so the cumulative count is yours to bring.

Worked Example Cracks per pavement section
1 of 3

A survey counts the cracks in 19 pavement sections. Find the median, mode, and range.

Cracks counted in 19 pavement sections
Cracks per sectionSections
29
35
43
51
61

Spread relative to size

A standard deviation of 300 psi is ordinary on 4,000 psi concrete and alarming on a flexural test that averages 600 psi. The Handbook's sample coefficient of variation compares spread with size:

CV=sxˉCV = \frac{s}{\bar{x}}

The units cancel, so you can compare entirely different tests. The Handbook's statistics chapter prints CVCV as a plain ratio, while its environmental section multiplies by 100 to make a percent, so give the form the question asks for. Our two suppliers come in at 2.3% and 13%.

Checkpoint
1 of 2

A mix's cylinder breaks have a mean of 3,500 psi and a standard deviation of 280 psi. What is the coefficient of variation, as a percent?

%

When the plain mean is the wrong average

A plain mean gives every value one equal vote. The Handbook prints three alternatives for when that's wrong.

Weighted mean

Say 4 ft of soil at 110 lb/ft³ lies over 6 ft at 120 lb/ft³. The average isn't 115, because there's more of the heavier soil. So each value gets votes according to how much of it there is:

xˉw=∑wixi∑wi=(4)(110)+(6)(120)4+6=116 lb/ft3\bar{x}_w = \frac{\sum w_i x_i}{\sum w_i} = \frac{(4)(110) + (6)(120)}{4 + 6} = 116\ \text{lb/ft}^3

A frequency table is the same idea with the counts as the weights, though the Handbook leaves you to notice that.

Geometric mean

Some quantities multiply instead of add. If traffic grows by a factor of 1.10 one year and 1.50 the next, the total factor is 1.65, and the single factor that gives the same total when applied twice is 1.65=1.28\sqrt{1.65} = 1.28, not the plain average of 1.30. The Handbook's sample geometric mean is the nnth root of the product:

geometric mean=x1x2x3⋯xnn\text{geometric mean} = \sqrt[n]{x_1 x_2 x_3 \cdots x_n}

Root-mean-square

Read the name backward for the recipe: square each value, take the mean, then take the root. That's the Handbook's sample root-mean-square value:

RMS=1n∑xi2\text{RMS} = \sqrt{\frac{1}{n}\sum x_i^2}

Squaring gives big values extra say, which suits anything whose effect grows with the square of its size, like the energy in a water wave. In fact, σ\sigma is the RMS of the deviations.

As a check, for positive values that aren't all equal, the geometric mean falls below the plain mean, and the RMS falls above it.

Checkpoint
1 of 4

A haul truck drives for 2 h at 60 mph and then for 1 h at 30 mph. What is its average speed, weighting each speed by its time?

mph

What moves when the data change

An outlier

Take 2, 3, 4, 6, and 10, with a mean of 5 and a median of 4, and change the 10 to 30. The mean jumps to 9. The median stays at 4, because it depends only on order: the top value is still on top, however far up it went. A mode, if the data have one, doesn't move either, unless the outlier lands on a repeated value. Spread reacts hardest of all, with the range going from 8 to 28 and ss from 3.16 to 11.8.

0 4 8 12 16 20 24 28 32 Original mean 5 median 4 The 10 changed to 30 mean 9 median 4
One value runs away, and only the mean follows

A shift or a scale

Add a constant to every value, such as a new datum, and the data set slides along the line without changing shape. Multiply every value by a constant, such as a unit conversion, and it stretches. Anything that says where the data are follows both. Anything that measures width ignores the slide and stretches with the stretch, and variance, built from squared distances, stretches by the square of the factor.

0 4 8 12 16 20 Original: x̄ = 5, s = 3.16 R = 8 Add 8: x̄ = 13, s = 3.16 R = 8 Multiply by 2: x̄ = 10, s = 6.32 R = 16
The same five values, shifted or scaled
What happens to each measure, for a constant cc and a positive factor kk
MeasureAdd cc to allMultiply all by kk
Mean, median, modeAdd ccMultiply by kk
Range, standard deviationNo changeMultiply by kk
VarianceNo changeMultiply by k2k^2

Skew

Weekly rainfall has lots of ordinary weeks and the occasional huge storm, so its distribution has a long tail on the high side. That's right-skewed. The mode sits at the peak. The median splits the area in half, and the tail puts extra area on the right, so the median lands right of the peak. The mean is the balance point, and a long tail has a lot of leverage, so it gets pulled farther still.

mode median mean long tail Value How often
A right-skewed distribution

A left-skewed distribution is the mirror image. Either way, reading from the tail inward gives mean, median, mode, which is conveniently alphabetical. The Handbook prints no rules for skew, shifts, or scaling, so reason these out from the pictures.

Checkpoint
1 of 3

Nine monthly rainfall totals all fall between 2 and 4 in. A tenth month brings a tropical storm and 18 in. Which measure of center moves the most?

A.
The mean
B.
The median
C.
The mode
D.
All three move by about the same amount

Recap

  • The mean xˉ\bar{x} is the balance point: deviations from it cancel.
  • Standard deviation is square, average, root. Divide by NN for a whole population (σ\sigma) and by n−1n - 1 for a sample (ss). The calculator shows both, and you choose.
  • The median is the middle observation after sorting, and in a frequency table the cumulative counts find it.
  • CV=s/xˉCV = s/\bar{x} is spread relative to size. Weighted means give unequal votes, geometric means suit things that multiply, and RMS is square, mean, root.
  • An outlier drags the mean and leaves the median alone. A shift moves only the centers. A factor kk multiplies the centers and ss by kk, and the variance by k2k^2. From a long tail inward: mean, median, mode.