Skip to content
LoveToCalc

Definition — what does that mean?

Sample variance

The same sum of squares, divided by n − 1.

In Swedish: Stickprovsvarians

Sample variance

The relation

s²=Σ(xi−x)²n−1

The variance estimated from a sample: the squared distances from the mean, summed and divided by one less than the count.

Separate the numbers with spaces, commas or semicolons, for example 2.5, 4, 7

s²≈4.5714

s² equals ((2 minus 5) squared plus (4 minus 5) squared plus (4 minus 5) squared plus (4 minus 5) squared plus (5 minus 5) squared plus (5 minus 5) squared plus (7 minus 5) squared plus (9 minus 5) squared) divided by (8 minus 1), equals ((−3) squared plus (−1) squared plus (−1) squared plus (−1) squared plus 0 squared plus 0 squared plus 2 squared plus 4 squared) divided by 7, equals (9 plus 1 plus 1 plus 1 plus 0 plus 0 plus 4 plus 16) divided by 7, equals 32 divided by 7, is approximately 4.5714

Working5 steps

What the formula says

s² = Σ(xᵢ − x̄)² / (n − 1)

The same sum of squares as in the variance, but divided by n − 1 instead of n. It is used when the values are a sample and you want to say something about the spread in the larger group they come from. The formula is a definition.

The reason for n − 1 is that the mean x̄ is worked out from the same sample. The values therefore lie closer, on average, to their own mean than to the population’s, and dividing by n would underestimate the spread on average. Dividing by a smaller number corrects for that.

Worked examples

Eight values. 2, 4, 4, 4, 5, 5, 7, 9 has a mean of 5 and a sum of squares of 32. The sample variance is 32 / 7 ≈ 4.571. The population variance of the same list is 32 / 8 = 4.

How large the difference is. The ratio between the two is n / (n − 1). For eight values it is 8/7 ≈ 1.14, so 14% larger. For a hundred values it is 100/99 ≈ 1.01. The difference matters mostly when the sample is small.

Common mistakes

Common questions

Why n − 1? Because the mean has already been worked out from the same values. The sum of squares is smallest around the sample’s own mean, so it is never larger than the sum of squares around the population’s true mean, and usually smaller. Dividing by n − 1 makes up for that. The usual explanation is that one degree of freedom has been spent on the mean.

Which one do I use in the exam? The one the question describes. If it says sample, or the answer should say something about a larger group, use n − 1. If the values are the whole group, use n. If the question does not say, ask which one the course means.

What does the calculator’s s button give? This one, with n − 1. The button marked σ or σₙ gives the population version. In a spreadsheet, VAR.S gives the sample variance and VAR.P the population variance.

Looking for another letter? The formula solved for each one
Solved for each letter
Solve forRearrangedWith the numbersComes to
s²Σ(xi−x)²n−1(2−5)²+(4−5)²+(4−5)²+(4−5)²+(5−5)²+(5−5)²+(7−5)²+(9−5)²8−14.5714

F-table

Table

F-table
ν₂α123456789101215203060120∞
10.05161.45199.50215.71224.58230.16233.99236.77238.88240.54241.88243.91245.95248.01250.10252.20253.25254.31
10.014,052.184,999.505,403.355,624.585,763.655,858.995,928.365,981.076,022.476,055.856,106.326,157.286,208.736,260.656,313.036,339.396,365.86
20.0518.5119.0019.1619.2519.3019.3319.3519.3719.3819.4019.4119.4319.4519.4619.4819.4919.50
20.0198.5099.0099.1799.2599.3099.3399.3699.3799.3999.4099.4299.4399.4599.4799.4899.4999.50
30.0510.139.559.289.129.018.948.898.858.818.798.748.708.668.628.578.558.53
30.0134.1230.8229.4628.7128.2427.9127.6727.4927.3527.2327.0526.8726.6926.5026.3226.2226.13
40.057.716.946.596.396.266.166.096.046.005.965.915.865.805.755.695.665.63
40.0121.2018.0016.6915.9815.5215.2114.9814.8014.6614.5514.3714.2014.0213.8413.6513.5613.46
50.056.615.795.415.195.054.954.884.824.774.744.684.624.564.504.434.404.36
50.0116.2613.2712.0611.3910.9710.6710.4610.2910.1610.059.899.729.559.389.209.119.02
60.055.995.144.764.534.394.284.214.154.104.064.003.943.873.813.743.703.67
60.0113.7510.929.789.158.758.478.268.107.987.877.727.567.407.237.066.976.88
70.055.594.744.354.123.973.873.793.733.683.643.573.513.443.383.303.273.23
70.0112.259.558.457.857.467.196.996.846.726.626.476.316.165.995.825.745.65
80.055.324.464.073.843.693.583.503.443.393.353.283.223.153.083.012.972.93
80.0111.268.657.597.016.636.376.186.035.915.815.675.525.365.205.034.954.86
90.055.124.263.863.633.483.373.293.233.183.143.073.012.942.862.792.752.71
90.0110.568.026.996.426.065.805.615.475.355.265.114.964.814.654.484.404.31
100.054.964.103.713.483.333.223.143.073.022.982.912.852.772.702.622.582.54
100.0110.047.566.555.995.645.395.205.064.944.854.714.564.414.254.084.003.91
120.054.753.893.493.263.113.002.912.852.802.752.692.622.542.472.382.342.30
120.019.336.935.955.415.064.824.644.504.394.304.164.013.863.703.543.453.36
150.054.543.683.293.062.902.792.712.642.592.542.482.402.332.252.162.112.07
150.018.686.365.424.894.564.324.144.003.893.803.673.523.373.213.052.962.87
200.054.353.493.102.872.712.602.512.452.392.352.282.202.122.041.951.901.84
200.018.105.854.944.434.103.873.703.563.463.373.233.092.942.782.612.522.42
300.054.173.322.922.692.532.422.332.272.212.162.092.011.931.841.741.681.62
300.017.565.394.514.023.703.473.303.173.072.982.842.702.552.392.212.112.01
600.054.003.152.762.532.372.252.172.102.041.991.921.841.751.651.531.471.39
600.017.084.984.133.653.343.122.952.822.722.632.502.352.202.031.841.731.60
1200.053.923.072.682.452.292.182.092.021.961.911.831.751.661.551.431.351.25
1200.016.854.793.953.483.172.962.792.662.562.472.342.192.031.861.661.531.38
∞0.053.843.002.602.372.212.102.011.941.881.831.751.671.571.461.321.221.00
∞0.016.634.613.783.323.022.802.642.512.412.322.182.041.881.701.471.321.00

Last reviewed October 6, 2026 ·How pages are reviewed

Statistics and probabilityAll formulas

Put this formula on your site

Copy the code into your page where the formula card should appear. It loads from lovetocalc.com, so it stays current on its own — all we ask is that the credit line comes with it.