Statistics Calculator

Enter numbers and the calculator finds the mean, median, mode, quartiles, variance, deviation and outliers, and shows a histogram and a box plot. Separate tabs: frequencies and the weighted mean, correlation with regression, the normal distribution. Every result comes with a step-by-step solution.

Enter data

Education
What to find
Settings
Decimal places
Quartile method
Quartiles of small samples depend on the method: “median of the halves” is the textbook one, QUARTILE.INC and QUARTILE.EXC follow the Excel functions.

Result

CalcCore
Enter the known quantities — the result will appear here.

What the statistics calculator does

The “Descriptive statistics” tab computes the main measures of a set of numbers: mean, median, mode, quartiles, range, variance, standard deviation, standard error, coefficient of variation, skewness, kurtosis and outliers, and draws a histogram and a box plot. “Frequencies and weighted mean” works with a table of values and frequencies or weights. “Correlation and regression” looks for the relation between two series, and “Normal distribution” finds the z-score, probabilities and quantiles.

Mean, median and mode

The mean is the sum of the values divided by their number. The median is the middle value of the sorted set (for an even count, the mean of the two middle ones). The mode is the most frequent value. For the set 12, 15, 11.5, 18, 15, 22, 9, 15, 40 the mean is 17.5, the median 15 and the mode 15. One large value (40) pulls the mean strongly but hardly changes the median, so for data with outliers the median is more reliable.

Variance and standard deviation

Variance is the mean squared deviation from the mean. For a sample divide by n − 1 (s² = Σ(x − x̄)² / (n − 1)), for a population by n (σ² = Σ(x − x̄)² / n). For the example above Σ(x − x̄)² = 684, so s² = 684 / 8 = 85.5 and s = √85.5 ≈ 9.2466. The standard deviation has the same units as the data. The coefficient of variation CV = s / x̄ · 100 % shows the spread relative to the mean.

Quartiles and outliers

Q1 and Q3 cut off the lowest and highest quarter of the values, and IQR = Q3 − Q1. Outliers are values outside [Q1 − 1.5·IQR; Q3 + 1.5·IQR]. In the example, by the “median of the halves” method Q1 = 11.75, Q3 = 20, IQR = 8.25, the fences are −0.625 and 32.375, so 40 is an outlier. For small samples the result depends on the method: the calculator supports the textbook one (median of the halves), Excel QUARTILE.INC and QUARTILE.EXC; choose it in the settings.

Frequencies and the weighted mean

When values repeat it is convenient to give them with frequencies: the mean x̄ = Σ(f·x) / Σf. For the values 2, 3, 4, 5, 6 with frequencies 3, 5, 9, 7, 2 the sum of frequencies is 26, Σ(f·x) = 104 and the mean is 4. The weighted mean works the same way, only any weights are used instead of frequencies, for example the credits of subjects or shares in a final grade.

Correlation and linear regression

The Pearson correlation coefficient r = Sxy / √(Sxx · Syy) measures a linear relation from −1 to 1; r² is the share of the variability of Y explained by X. The regression line y = a + bx has b = Sxy / Sxx and a = ȳ − b·x̄. For X = 1…5 and Y = 2, 4, 5, 4, 5 you get b = 0.6, a = 2.2, r ≈ 0.7746, r² = 0.6 and the prediction for x = 6 is 5.8. Correlation does not prove causation, and a prediction far outside the data is unreliable.

Normal distribution and the z-score

The z-score z = (x − μ) / σ shows by how many standard deviations a value differs from the mean. For μ = 100, σ = 15 and x = 115 we get z = 1 and P(X < 115) ≈ 0.8413, P(X > 115) ≈ 0.1587. In a normal distribution about 68 % of the values lie within ±1σ, 95 % within ±2σ and 99.7 % within ±3σ. The inverse task: the 95th percentile has z ≈ 1.6449, that is x ≈ 124.67.

All calculations are exact (fractions and roots are not rounded), decimals are rounded to the chosen number of places. Skewness, kurtosis, the geometric mean and the normal distribution are approximate values.

FAQ

How do I calculate the arithmetic mean?

Add all the numbers and divide the sum by their count. For 12, 15, 11.5, 18, 15, 22, 9, 15, 40 it is 157.5 / 9 = 17.5.

Why is the median sometimes better than the mean?

The median does not depend on outliers: one very large or small value shifts the mean strongly but hardly changes the median.

What is the mode?

The most frequent value. If all values are different or equally frequent there is no mode.

What is the difference between sample and population variance?

The sample variance is divided by n − 1, the population variance by n. For a sample n − 1 gives an unbiased estimate; the calculator has a switch.

What is the standard deviation?

The square root of the variance: the typical spread of the values around the mean in the same units.

How do I find the quartiles?

Q1 and Q3 are the medians of the lower and upper halves of the set (or by the Excel formulas). The method is chosen in the settings because the results differ for small samples.

What is an outlier?

A value far from the bulk of the data. By Tukey's rule it lies outside Q1 − 1.5·IQR and Q3 + 1.5·IQR.

What is the coefficient of variation?

The standard deviation divided by the mean, in percent; handy for comparing the spread of sets with different units or scale.

How do I use the frequency table?

In each row enter a value and how many times it occurs. The calculator expands the set and computes the mean, median, mode and variance.

What does the correlation coefficient mean?

The direction and strength of a linear relation: 1 is a perfect positive one, −1 a negative one, 0 no relation. It does not prove that one quantity causes the other.

What is a z-score?

The number of standard deviations between a value and the mean: z = (x − μ) / σ.

How do I separate numbers?

With a space, a comma, a semicolon or a new line. The decimal separator in the English version is a dot (3.5).