母集団の分散 電卓

データ入力

値はカンマ、スペース、タブ、または改行で区切ります
0 入力された値の数

データ値を入力すると結果が表示されます

What Is Population Variance?

Population variance, written σ², describes the spread of a group when you have measured every member of it. There is nothing to estimate, so you take the average of the squared deviations from the mean exactly as it is: divide by N, the number of values. Typical examples are the marks of every student in one class, the weight of every parcel in a single shipment you weighed in full, or one company's monthly sales for a completed year. The key question is not how big your data set is, but whether it is the whole group you want to describe.

Worked Example: A Complete Tutorial Group

A tutor has exactly six students and wants to describe how spread out their test scores are. These six scores are the entire group, so population variance applies.

Data (marks):70, 85, 75, 90, 65, 95

  1. Mean. x̄ = 480 / 6 = 80
  2. Deviations and squares. Subtract the mean from each value and square the result:
xᵢxᵢ − x̄(xᵢ − x̄)²
70−10100
85525
75−525
9010100
65−15225
9515225
Sum0700
  1. Sum of squares. Σ(xᵢ − x̄)² = 700
  2. Population variance. σ² = 700 / 6 = 116.6667; population standard deviation σ = √116.6667 = 10.8012

Population variance σ² = 700 / 6 ≈ 116.67, so the population standard deviation is σ ≈ 10.80 marks. If the same six scores were treated as a sample, s² = 700 / 5 = 140; the gap between the two answers is largest for small groups like this one.

Formula

Population variance uses the population mean μ and divides by N:

Population Variance

σ² = Σ(xᵢ − μ)² / N
  • σ²Population variance
  • xᵢEach value in the population
  • μPopulation mean
  • NNumber of values in the population

Population Standard Deviation

σ = √σ²
  • σPopulation standard deviation, in the original units

How to Calculate Population Variance Step by Step

  1. Confirm you have the entire population.Every member of the group you are describing must be in the data.
  2. Find the population mean (μ).Add all values and divide by N.
  3. Compute deviations.Subtract μ from each value.
  4. Square each deviation.Calculate (xᵢ − μ)².
  5. Add the squared deviations.Σ(xᵢ − μ)² is the sum of squares.
  6. Divide by N.σ² = Σ(xᵢ − μ)² / N.

More Population Variance Examples

1All Students in a Class

Data:78, 82, 91, 67, 85, 73, 88, 79
Solution:Mean μ = 80.375. Sum of squared deviations = 435.875.
Answer:Population variance (σ²) = 435.875 / 8 ≈ 54.48

2Every Product in a Small Batch

Data:10.2, 10.0, 10.1, 10.3, 10.0
Solution:Mean μ = 10.12. Sum of squared deviations = 0.068.
Answer:Population variance (σ²) = 0.068 / 5 = 0.0136

3Complete Inventory Count

Data:150, 200, 175, 180, 195
Solution:Mean μ = 180. Squared deviations: 900, 400, 25, 0, 225. Sum of squares = 1550.
Answer:Population variance (σ²) = 1550 / 5 = 310

4All Employees' Years of Experience

Data:2, 5, 8, 3, 12, 1, 7
Solution:Mean μ ≈ 5.43. Sum of squared deviations ≈ 89.71.
Answer:Population variance (σ²) ≈ 89.71 / 7 ≈ 12.82

How to Use This Population Variance Calculator

Switch the calculator to Population, then:

  1. Enter every value in the population, separated by commas, spaces or line breaks.
  2. Check that the result card says Population Variance (σ²).
  3. Read σ², the population standard deviation σ and the mean μ.
  4. Open the calculation steps to see the division by N written out with your numbers.

What the Calculator Shows You

  • The exact population variance and standard deviation, with no correction applied.
  • The sample figure for the same data, so you can see the difference if you are unsure which applies.
  • Every deviation and squared deviation.
  • Calculations run in your browser; your numbers are not sent anywhere.

When to Use Population Variance

A single class or team

A teacher describing the spread of marks for the class they actually taught.

Census data

Statistics about a population that has been counted in full.

Inspected-in-full batches

When every unit in a lot has been measured, not just a few.

Completed periods

Describing last year's twelve monthly figures as they are, without predicting anything.

Probability distributions

The variance of a die roll or another fully specified distribution is a population variance.

Population or Sample? How to Decide

Size does not decide it. What matters is which group your conclusions are about:

It depends on your question

If you are describing "all students in Section A", Section A's marks are the population. If you are describing "all students at the university", Section A is a sample.

Describing versus predicting

Last year's monthly sales are a population if you only want to describe last year. If you are using them to say something about future months, treat them as a sample.

Same data, two answers

For the tutorial group above, σ² ≈ 116.67 but s² = 140. With 600 values instead of 6 the two would be almost identical.

When in doubt, use sample variance

Most real data is a sample of something larger, and the n − 1 version will not understate the spread.

よくある質問: 母集団の分散 電卓

分散について知っておくべきことのすべて。

母集団分散とは何ですか?

母集団分散 (σ²) は、母集団平均 (μ) からのデータ ポイントの正確な広がりを測定します。式はσ² = Σ(xᵢ − μ)² / N で、N は母集団内のデータ ポイントの総数です。標本分散とは異なり、すべてのデータが存在するため、推定や補正は必要なく、N (N−1 ではなく) で除算されます。

母集団の分散をどのような場合に使用する必要がありますか?

データセットに調査対象の母集団のすべてのメンバーが含まれている場合は、母集団の分散を使用します。例: クラス内のすべての生徒、バッチ内のすべての製品、完全な国勢調査データ、一定期間内のすべてのトランザクション。より大きな母集団からのサンプル (サブセット) がある場合は、代わりにサンプル分散を使用します。重要な質問は、データが完全なグループを表しているかどうかです。

2、4、4、4、5、5、7、9 の母集団分散はいくらですか?

2、4、4、4、5、5、7、9 の場合: 平均 (μ) = 40/8 = 5。偏差: −3、−1、−1、−1、0、0、2、4。2 乗: 9、1、1、1、0、0、4、16。2 乗和 = 32。母集団分散(σ²) = 32/8 = 4となります。母集団標準偏差 (σ) = √4 = 2。

σ²とs²の違いは何ですか?

σ² (母分散) はN で割って母平均μを使用します。 s² (標本分散) はn−1 で除算し、標本平均 x̄ を使用します。サンプルの分散は、同じデータセットの母集団の分散よりも常にわずかに大きくなります。この差 (ベッセル補正) は、小さなデータセットで最も重要であり、大きな N では無視できる程度になります。

母集団の分散をゼロにすることはできるのでしょうか?

はい、すべての値が同一の場合、母集団の分散はゼロになります。母集団内のすべてのデータ ポイントが同じ値を持つ場合 (例: {7, 7, 7, 7})、平均からの偏差はすべてゼロとなり、σ² = 0 となります。これは、ばらつきがゼロ、つまり広がりがまったくないことを示します。値間に差異があると、正の分散が生成されます。

Excel で母集団の分散を見つけるにはどうすればよいですか?

Excel では、母集団分散に =VAR.P(range) を使用します。例: =VAR.P(A1:A20)。母集団の標準偏差には、=STDEV.P(range) を使用します。 Google スプレッドシートでも同じ機能が機能します。 Python の場合: np.var(data) (NumPy のデフォルトは母分散)。 R では、var(x) * (n-1)/n となります。これは、R の var() のデフォルトがサンプル分散であるためです。