Descriptive Statistics Calculator

How to Use the Descriptive Statistics Calculator

This descriptive statistics calculator will help you summarize numerical, categorical, or paired data. Select the type of data you have, enter the observations, and the calculator will return the appropriate descriptive statistics.

The example values included in the calculator are summarized automatically. After entering or changing your own data, click Calculate to update the results.

For Numerical Data

Use Numerical Data when you have one quantitative variable, such as exam scores, ages, heights, weights, or measurements.

1. Select Numerical Data.
2. Enter the observations in the Data Values box. Separate values with commas, spaces, tabs, semicolons, or line breaks. Alternatively, you can also paste a column directly from Excel or Google Sheets.
3. Click Calculate after entering or editing the values.
4. Review the results for center, position, spread, variability, and distribution shape.

For Categorical Data

Choose Categorical Data when your observations are groups or labels rather than numerical measurements.

1. Select Categorical Data.
2. Enter one category per line or separate categories with commas, tabs, or semicolons. Category names can contain spaces, so Very Satisfied is treated as one response.
3. Click Calculate to summarize the data.
4. Review the frequencies, relative frequencies, percentages, mode, and number of categories.

For Paired Data

Choose Paired Data when every X observation has a corresponding Y observation.

1. Enter the first variable in the X Values box.
2. Enter the matching second variable in the Y Values box. Keep the observations in corresponding order and make sure X and Y contain the same number of values.
3. Click Calculate after entering or changing the data.
4. Review the calculation totals, separate X and Y summaries, covariance, Pearson correlation, and r².

Paired mode also provides values such as Σxy, SSₓₓ, SSᵧᵧ, and SSₓᵧ. These are useful if you need to calculate covariance, correlation, or simple linear regression manually.

What Is a Descriptive Statistics Calculator?

A descriptive statistics calculator takes raw observations and reduces them to statistics that are easier to interpret.

For numerical data, those statistics describe the center, position, spread, and shape of the distribution. Categorical observations are summarized using counts and proportions, while paired numerical data can also be described using covariance and correlation.

Descriptive statistics help answer questions such as:

• What is the typical value?
• How spread out are the observations?
• Which value or category occurs most often?
• Where does the middle 50% of the data fall?
• Is the numerical distribution skewed?
• How often does each category occur?
• Do two numerical variables tend to move together?

These calculations describe the data you entered. They do not, by themselves, test a statistical hypothesis or determine whether the findings can be generalized to a larger population.

Descriptive Statistics for Numerical Data

Select Numerical Data when you want a complete summary of one quantitative variable.

The results are organized into four groups.

Center and Totals

Position and Spread

The calculator also gives the five-number summary:

Minimum, Q1, Median, Q3, Maximum

Variability

Shape and Other Measures

Example 1: Descriptive Statistics for Exam Scores

A statistics instructor records the scores of seven students on an exam:

68, 72, 75, 80, 80, 85, 90

Find the descriptive statistics for the scores.

Enter the values under Numerical Data and click Calculate.

The calculator gives:

• Number of observations, n = 7
• Sum, Σx = 550
• Sum of squares, Σx² = 43,558
• Mean, x̄ ≈ 78.571429
• Median, Mdn = 80
• Mode = 80
• Minimum = 68
• First quartile, Q1 = 72
• Third quartile, Q3 = 85
• Maximum = 90
• Range, R = 22
• Interquartile range, IQR = 13
• Population variance, σ² ≈ 49.102041
• Population standard deviation, σ ≈ 7.007285
• Sample variance, s² ≈ 57.285714
• Sample standard deviation, s ≈ 7.568733
• Sample coefficient of variation, CV ≈ 9.63%
• Geometric mean, GM ≈ 78.259104
• Sample skewness, G₁ ≈ 0.136406
• Excess kurtosis, G₂ ≈ −0.665649

The five-number summary is:

68, 72, 80, 85, 90

From the results, the mean score is about 78.57 points, while the median is 80. The middle 50% of the scores extends from 72 to 85.

How Are Quartiles Calculated?

Quartile values can differ slightly between calculators because several accepted calculation methods exist.

StatCalc uses the median-of-halves method. When the dataset contains an odd number of observations, the overall median is excluded before finding Q1 and Q3.

Consider the ordered exam scores:

68, 72, 75, 80, 80, 85, 90

The middle observation is 80, so:

Median, Q2 = 80

After excluding that middle observation, the lower half is:

68, 72, 75

Its median is:

Q1 = 72

The upper half is:

80, 85, 90

Therefore:

Q3 = 85

The interquartile range is:

IQR = Q3 − Q1

= 85 − 72

= 13

If you only need quartiles and want to see the calculation in more detail, use our Quartile Calculator.

Sample vs. Population Statistics

Before using variance or standard deviation, decide whether the observations represent a sample or the entire population of interest.

For population data, use:

• Population variance, σ²
• Population standard deviation, σ

For sample data, use:

• Sample variance, s²
• Sample standard deviation, s

Suppose the seven exam scores represent every student in a particular study group. If that study group is the complete group you want to describe, the population statistics are appropriate.

Now suppose those seven students were selected from all students taking introductory statistics at a university. Because the observations represent only part of the larger population, the sample statistics would normally be used.

For a complete step-by-step calculation, see our Variance Calculator or Standard Deviation Calculator.

How to Interpret Mean, Median, and Mode

Mean, median, and mode describe the center of data in different ways.

The mean is calculated by adding all numerical observations and dividing by the number of values. Since every observation contributes to the calculation, unusually large or small values can influence it.

To find the median, first arrange the observations from smallest to largest and locate the middle. This measure is often useful when extreme values make the mean less representative.

Mode identifies the value that appears most often. Unlike the mean and median, the mode can also summarize categorical observations.

If you want to focus on one measure and see its full calculation, use our Mean Calculator, Median Calculator, or Mode Calculator.

How to Interpret Measures of Spread

Measures of spread show how much the observations differ from one another.

A quick measure is the range:

R = Maximum − Minimum

Because range depends only on the two extreme observations, it can change substantially when an outlier is present.

The interquartile range focuses on the middle half of the data:

IQR = Q3 − Q1

Variance and standard deviation take every numerical observation into account. Standard deviation is usually easier to interpret because it is expressed in the same units as the original values.

As a result, two datasets may have identical means but very different amounts of variability.

Skewness and Kurtosis

Skewness and kurtosis provide additional information about the shape of a numerical distribution.

Sample skewness, G₁, measures asymmetry.

• A value near 0 suggests a relatively symmetric distribution.
• A positive value indicates a longer right tail.
• A negative value indicates a longer left tail.

StatCalc reports the adjusted Fisher-Pearson sample skewness coefficient.

On the other hand, excess kurtosis, G₂, describes tail behavior relative to a normal distribution.

• A value near 0 indicates kurtosis similar to a normal distribution.
• A positive value indicates heavier tails.
• A negative value indicates lighter tails.

Neither statistic should be interpreted in isolation. Consider them together with the center, spread, sample size, and shape of the data.

Geometric Mean and Coefficient of Variation

The calculator also reports two measures that are useful in particular situations.

Geometric Mean

Geometric mean is commonly used for growth factors, ratios, and quantities that change multiplicatively.

For n positive values:

GM = (x₁ × x₂ × … × xₙ)^(1/n)

All observations must be greater than zero. If a dataset contains zero or negative values, the calculator does not report a geometric mean.

Coefficient of Variation

The sample coefficient of variation compares the standard deviation with the mean:

CV = (s / x̄) × 100%

Because CV is expressed as a percentage, it can be useful when comparing relative variability across datasets measured on different scales.

For a dedicated calculation, use our Coefficient of Variation Calculator.

Descriptive Statistics for Categorical Data

Categorical data consist of labels or groups rather than numerical measurements.

Examples include preferred transportation method, college major, political district, product category, and survey responses such as Satisfied or Dissatisfied.

For this type of data, the calculator reports:

Mean, variance, and standard deviation are not calculated for ordinary nominal categories because those measures require meaningful numerical values.

Example 2: Descriptive Statistics for Categorical Data

A U.S. high school asks 10 randomly selected students how they usually travel to school.

The responses are:

Bus, Car, Walk, Bus, Bus, Car, Bike, Bus, Walk, Car

Find the frequency distribution and mode.

Select Categorical Data, enter the responses, and click Calculate.

From the data:

• Number of observations, n = 10
• Number of categories, k = 4
• Mode = Bus
• Modal frequency, fₘₐₓ = 4

The frequency distribution is:

Therefore, Bus is the most common transportation method among the students surveyed, accounting for 40% of the responses.

Cumulative frequency is not shown for general categorical data because categories such as Bus, Car, Walk, and Bike do not have a natural numerical order.

Descriptive Statistics for Paired Data

Paired data contain two numerical observations that belong together.

Common examples include:

• study time and exam score for the same student;
• height and weight for the same participant;
• advertising expenditure and sales for the same month;
• temperature and electricity use on the same day.

Paired mode first provides the totals needed for further calculations. It then summarizes X and Y separately before reporting covariance and correlation.

Calculation Totals

The three deviation totals can be obtained from:

SSₓₓ = Σx² − (Σx)²/n

SSᵧᵧ = Σy² − (Σy)²/n

SSₓᵧ = Σxy − (ΣxΣy)/n

These values are particularly useful in covariance, Pearson correlation, and simple linear regression calculations.

X and Y Variable Summaries

Each variable also receives its own measures of spread and position.

For X, the calculator reports:

• Median, Mdnₓ
• Minimum, xₘᵢₙ
• Maximum, xₘₐₓ
• Range, Rₓ
• Population variance, σₓ²
• Population standard deviation, σₓ
• Sample variance, sₓ²
• Sample standard deviation, sₓ

Corresponding Y statistics are reported using Mdnᵧ, yₘᵢₙ, yₘₐₓ, Rᵧ, σᵧ², σᵧ, sᵧ², and sᵧ.

Paired Statistics

Example 3: Paired Descriptive Statistics

A statistics teacher wants to examine the relationship between study time and quiz scores for five students.

From the question, we know that:

X: 2, 3, 4, 5, 6

Y: 65, 70, 76, 82, 88

Enter these values under Paired Data and click Calculate.

The preliminary totals are:

• Number of pairs, n = 5
• Sum of X, Σx = 20
• Sum of Y, Σy = 381
• Sum of squared X values, Σx² = 90
• Sum of squared Y values, Σy² = 29,369
• Sum of cross products, Σxy = 1,582
• Mean of X, x̄ = 4
• Mean of Y, ȳ = 76.2
• Sum of squares for X, SSₓₓ = 10
• Sum of squares for Y, SSᵧᵧ = 336.8
• Sum of cross deviations, SSₓᵧ = 58

Using these quantities, the paired statistics are:

• Sample covariance, sₓᵧ = 14.5
• Population covariance, σₓᵧ = 11.6
• Pearson correlation, r ≈ 0.999406
• Coefficient of determination, r² ≈ 0.998812

This implies that study time and quiz score have a very strong positive linear relationship in this small dataset.

However, correlation only measures association. It does not prove that the additional study time caused the higher quiz scores.

How to Calculate Covariance From the Paired Totals

The preliminary values provided by the calculator make it easier to show the manual calculation.

For the study-time example:

SSₓᵧ = 58

and:

n = 5

Sample Covariance

For sample data:

sₓᵧ = SSₓᵧ / (n − 1)

Substituting the values gives:

sₓᵧ = 58 / (5 − 1)

= 58 / 4

= 14.5

Population Covariance

If the five pairs represent the complete population:

σₓᵧ = SSₓᵧ / n

= 58 / 5

= 11.6

How to Calculate Pearson Correlation From the Paired Totals

Pearson’s correlation coefficient can be calculated from SSₓₓ, SSᵧᵧ, and SSₓᵧ.

Use:

r = SSₓᵧ / √(SSₓₓ × SSᵧᵧ)

From the example:

• SSₓₓ = 10
• SSᵧᵧ = 336.8
• SSₓᵧ = 58

Substituting gives:

r = 58 / √(10 × 336.8)

≈ 0.999406

Since r is very close to +1, the observed data have a very strong positive linear relationship.

Using the Paired Totals for Simple Linear Regression

Although this calculator does not perform a complete regression analysis, its paired-data results provide the main quantities needed to calculate a least-squares regression line manually.

The regression slope can be found from:

b₁ = SSₓᵧ / SSₓₓ

Once the slope is known, calculate the intercept using:

b₀ = ȳ − b₁x̄

The regression equation is then written as:

ŷ = b₀ + b₁x

For a full regression analysis, including the fitted equation and graphical relationship, use the appropriate regression or scatter plot calculator rather than treating this descriptive summary as a complete regression procedure.

Covariance vs. Correlation

Covariance and correlation both describe how two numerical variables vary together, but they answer slightly different questions.

A positive covariance indicates that larger X values tend to occur with larger Y values. Negative covariance suggests that one variable tends to decrease as the other increases.

Because covariance is affected by the units used to measure X and Y, its magnitude is difficult to compare across different datasets.

Pearson correlation removes those units and standardizes the linear relationship between −1 and 1:

• r close to +1 indicates a strong positive linear relationship.
• r close to −1 indicates a strong negative linear relationship.
• r close to 0 indicates little or no linear relationship.

Squaring r gives the coefficient of determination, r².

Five-Number Summary

A five-number summary gives a compact description of the position and spread of numerical data.

It consists of:

1. Minimum
2. First quartile, Q1
3. Median, Q2
4. Third quartile, Q3
5. Maximum

These five values are also commonly used to construct a box-and-whisker plot.

Unlike the mean and standard deviation, the five-number summary makes it easier to see the location of the middle half of the dataset and the distance between its lower and upper extremes.

Descriptive Statistics vs. Inferential Statistics

Descriptive statistics summarize the observations that were actually collected. For example, calculating the mean, median, standard deviation, or frequency distribution of a dataset is descriptive analysis.

On the other hand, inferential statistics address questions about a larger population. Common procedures include confidence intervals, hypothesis tests, regression inference, and analysis of variance.

Suppose a sample of college students has a mean exam score of 78.6. Reporting that sample mean is descriptive. However, testing whether the population mean differs from 75 would require an inferential statistical test.

This calculator therefore focuses on describing the data rather than testing statistical hypotheses.

Descriptive Statistics Calculator vs. Statistics Calculator

The two calculators serve different purposes. Use this Descriptive Statistics Calculator when you want to enter a dataset and obtain a complete summary at once. However, if you want to compute individual statistics and perform mathematical operations such as fractions, powers, logarithms, factorials, combinations, permutations, and complete expressions, use the Statistics Calculator.

A simple distinction is:

• Descriptive Statistics Calculator summarizes an entire dataset.
• Statistics Calculator calculates individual statistics or mathematical expressions.

Frequently Asked Questions