Understand Data & Uncertainty
Statistics & Probability Calculators for Averages, Variability, Z-Scores & Probability
Statistics helps you describe and understand data. Probability helps you work with uncertain events and possible outcomes. Start by deciding whether you need to find the centre of some data, measure its spread, understand a value’s position or uncertainty, or calculate probability.
Choose Your Path
What Do You Need to Calculate?
Most questions on this page fit one of four statistical pathways. Choose the description that best matches your question.
Centre & Averages
What value best represents the middle or centre?
Use this path when you need to summarize a dataset with a representative value.
- Arithmetic mean
- Median
- Mode
- Weighted average
Spread & Variability
How spread out are the values?
Use this path when you want to understand how much observations vary around their centre.
- Variance
- Standard deviation
- Mean absolute deviation
- Sample vs population
Position & Uncertainty
Where does a value sit, or how uncertain is an estimate?
Use this path for standardized values, distributions and supported statistical inference.
- Z-scores & percentiles
- Standard error
- Critical values
- Confidence intervals
Probability & Possibilities
How likely is an event, or how many outcomes are possible?
Use this path for uncertain events, odds, expected values and counting arrangements.
- Probability & odds
- Conditional probability
- Expected value
- Permutations & combinations
How the Topics Connect
Statistical questions often build from describing data toward understanding variation and uncertainty.
Quick Access to Statistics & Probability Tools
Already know what you need? Open the dedicated tool. If you are unsure which method applies, use the four pathways above first.
Core Concepts & Relationships
Understand the Statistical Question Before Choosing the Formula
Statistics is not one calculation. Different methods answer different questions. Some describe the centre of a dataset, some measure its spread, some show where a value sits within a distribution, and others help describe uncertainty or the chance of an event.
Start With One Simple Question
Choose the type of information you are trying to get from the numbers.
What value represents the middle or typical result?
Think about mean, median, mode or a weighted average.
Explore Averages →How much do the observations differ from each other?
Think about variance, standard deviation or absolute deviation.
Explore Variability →Where does a value sit, or how uncertain is an estimate?
Think about z-scores, percentiles, standard errors, critical values and confidence intervals.
Explore Distributions →How likely is an outcome or how many arrangements exist?
Think about probability, odds, expected value, permutations and combinations.
Explore Probability →Four Core Areas of This Statistics & Probability Guide
Each area answers a different kind of question, but the concepts often connect.
1. Averages — Describe the Centre
Measures of centre reduce a set of observations to a useful summary value. Mean, median and mode do this in different ways.
2. Variability — Describe the Spread
Two datasets can have the same mean but very different amounts of spread. Variance, standard deviation and absolute deviations help describe that difference.
3. Distributions — Position & Statistical Uncertainty
Distribution-based methods can show how far an observation lies from a mean, convert positions into probabilities, or help quantify uncertainty around an estimate when the required assumptions are appropriate.
4. Probability — Model Possible Outcomes
Probability describes uncertainty. Combinatorics helps count possible selections or arrangements when those counts are needed for a probability problem.
Key Statistics Terms in Plain English
These terms appear throughout the calculators and educational pages in this pillar.
- Dataset
- A collection of observations or values being studied.
- Population
- The complete group of observations or individuals the statistical question concerns.
- Sample
- A subset of a population used to obtain information about that population.
- Mean
- The arithmetic average: add the values and divide by the number of values.
- Median
- The middle ranked value after observations have been placed in order.
- Mode
- The value or category occurring most often. A dataset can have more than one mode or no mode.
- Weighted Mean
- An average in which some values have more influence because they have different weights.
- Variance
- A measure based on squared deviations that describes how spread out observations are around the mean.
- Standard Deviation
- The square root of variance. It expresses spread in the same units as the original data.
- Mean Absolute Deviation
- The average size of absolute deviations from a chosen centre.
- Z-Score
- A standardized score showing how many standard deviations a value lies above or below the mean.
- Percentile
- A way of describing relative position. Its exact calculation depends on the selected method or distribution.
- Standard Error
- A measure of the sampling variability of a statistic, such as a sample mean.
- Critical Value
- A value taken from an appropriate probability distribution for a selected confidence or significance calculation.
- Confidence Interval
- An interval calculated from sample information using a statistical procedure with a stated confidence level.
- Probability
- A numerical description of how likely an event is under a specified probability model.
- Expected Value
- A probability-weighted average of possible numerical outcomes.
- Combination
- A selection in which order does not matter.
How Statistical Concepts Build on Each Other
Many calculations use results from an earlier step. These pathways show some of the most useful relationships.
Centre Journey
Spread Journey
Confidence Interval Journey
Probability Journey
Important Statistical Distinctions
These pairs are related, but they do not mean the same thing. Keeping them separate prevents many common errors.
Mean vs Median
The mean uses every numerical value. The median uses ranked position and is generally less sensitive to extreme values.
Ordinary Mean vs Weighted Mean
An ordinary mean gives each observation equal influence. A weighted mean allows different observations to carry different weights.
Variance vs Standard Deviation
Variance is expressed in squared units. Standard deviation is the square root of variance and returns the measure of spread to the original data units.
Population vs Sample
A population is the complete group being studied. A sample is only part of that population. Some formulas change depending on which one you have.
Raw Score vs Z-Score
A raw score uses the original measurement units. A z-score expresses the value as a number of standard deviations from the mean.
Confidence Level vs Confidence Interval
The confidence level describes the long-run coverage target of the procedure. The confidence interval is the numeric interval calculated from a particular sample.
Probability vs Odds
Probability expresses likelihood as a share of all possible probability mass. Odds compare occurrence with non-occurrence.
Permutation vs Combination
In a permutation, order matters. In a combination, order does not matter.
Formula Overview
These formulas show how the main concepts connect. The complete calculation procedures and worked substitutions belong in the next sections.
Arithmetic Mean
x̄ = (x₁ + x₂ + … + xₙ) ÷ nx̄ = sample mean; x = observation; n = number of observations.
Weighted Mean
Weighted Mean = Σ(wx) ÷ Σwx = value and w = weight assigned to that value.
Population Variance
σ² = Σ(x − μ)² ÷ Nμ = population mean; N = population size; σ² = population variance.
Sample Variance
s² = Σ(x − x̄)² ÷ (n − 1)x̄ = sample mean; n = sample size; s² = sample variance.
Z-Score
z = (x − μ) ÷ σx = raw value; μ = population mean; σ = population standard deviation.
Standard Error of a Mean
SE = σ ÷ √nFor the population-standard-deviation form: σ = population standard deviation and n = sample size.
Margin of Error
ME = Critical Value × SESE = standard error. The critical value depends on the selected distribution and confidence procedure.
Confidence Interval Structure
CI = Estimate ± Margin of ErrorThe exact standard error and critical value depend on the statistical method and its assumptions.
Match the Question to the Statistical Concept
Use this table as a quick concept check before selecting a calculator.
| Question | Main concept | Typical information needed | Typical result |
|---|---|---|---|
| What is the arithmetic average? | Mean | Numerical observations | One central value |
| What is the middle ranked value? | Median | Ordered observations | Middle value or midpoint of two values |
| Which value occurs most often? | Mode | Values or categories and frequencies | Most frequent value or values |
| How spread out is the dataset? | Variance / standard deviation / MAD | Dataset and population/sample choice | Measure of variability |
| How far is this value from the mean? | Z-score | Value, mean and standard deviation | Standardized position |
| How precise is this estimate? | Confidence interval | Estimate, variability, sample size, confidence level and applicable assumptions | Lower and upper interval bounds |
| How likely is this event? | Probability | Events, outcomes or probability model | Probability |
| How many selections are possible? | Combination | Total items and number selected | Number of unordered selections |
| How many ordered arrangements are possible? | Permutation | Total items and number arranged | Number of ordered arrangements |
Formulas, Methods & Manual Calculation
How to Calculate Statistics & Probability Step by Step
A statistical formula only works when it matches the question and the information you have. Use the method below to choose the correct calculation, keep sample and population formulas separate, calculate carefully, and interpret what the result actually tells you.
The Five-Step Manual Calculation Method
Use this sequence whether you are calculating an average, standard deviation, z-score, confidence interval or probability.
Match the method to the statistical question and data.
Write down the data, variables and required assumptions.
Put each known value into the correct part of the formula.
Keep enough precision through intermediate steps.
Explain what the number means in the statistical context.
1. Measures of Centre
Use these methods when the question asks for an average, middle value, most frequent value or weighted average.
Mean, Median, Mode & Weighted Mean
These calculations all describe centre, but they use different information from the dataset.
Sample Arithmetic Mean
x̄ = Σx ÷ nx̄ = sample mean; Σx = sum of the observations; n = number of observations.
Population Mean
μ = Σx ÷ Nμ = population mean; Σx = sum of population values; N = population size.
Weighted Mean
Weighted Mean = Σ(wx) ÷ Σwx = value; w = weight. Divide the weighted sum by the total weight, not simply by the number of rows.
Median
Odd n: middle ordered value Even n: (two middle values) ÷ 2Sort the observations before identifying the middle position.
Mode
Mode = most frequent value or categoryA dataset may have one mode, more than one mode, or no mode.
2. Variance, Standard Deviation & Absolute Deviation
Measures of spread describe how observations differ from a centre. Sample and population formulas must not be mixed.
Core Variability Formulas
The usual variance pathway is dataset → mean → deviations → squared deviations → variance → standard deviation.
Population Variance
σ² = Σ(x − μ)² ÷ Nσ² = population variance; x = observation; μ = population mean; N = population size.
Sample Variance
s² = Σ(x − x̄)² ÷ (n − 1)s² = sample variance; x̄ = sample mean; n = sample size.
Population Standard Deviation
σ = √[Σ(x − μ)² ÷ N]Standard deviation is the square root of variance and uses the original measurement units.
Sample Standard Deviation
s = √[Σ(x − x̄)² ÷ (n − 1)]Use the sample form when the observations are a sample and the method calls for sample variance.
Mean Absolute Deviation About the Mean
MAD = Σ|x − x̄| ÷ nThis definition uses absolute distances from the sample mean. State the chosen centre because “MAD” can be used for different absolute-deviation measures.
Range
Range = Maximum − MinimumRange uses only the two extreme values, unlike standard deviation, which uses all observations.
3. Z-Scores, Standard Error & Confidence Intervals
These methods move from describing observations toward standardized position and statistical inference. The required distribution and assumptions matter.
Standardization & Mean-Inference Formulas
A confidence interval generally combines a point estimate, standard error and an appropriate critical value.
Z-Score
z = (x − μ) ÷ σx = raw value; μ = mean; σ = standard deviation. The result is measured in standard deviations from the mean.
Raw Value From a Z-Score
x = μ + zσThis is the z-score formula rearranged to recover the raw value.
Standard Error — Known Population SD
SE = σ ÷ √nσ = population standard deviation; n = sample size.
Estimated Standard Error — Sample SD
SE = s ÷ √ns = sample standard deviation. This estimated standard error is used in applicable sample-based mean-inference procedures.
Margin of Error
ME = Critical Value × SEThe critical value depends on the distribution, confidence level or significance level, tail choice and, for t methods, degrees of freedom.
General Confidence Interval
CI = Estimate ± MECI = confidence interval; ME = margin of error. This is the general interval structure.
Z Confidence Interval for a Mean
x̄ ± z* × (σ ÷ √n)Used for a mean under the applicable z-method conditions when σ is known.
One-Sample t Confidence Interval for a Mean
x̄ ± t* × (s ÷ √n)For the applicable one-sample t procedure, degrees of freedom are df = n − 1.
4. Probability, Expected Value & Combinatorics
Probability methods depend on how events relate to each other and, in many problems, how possible outcomes are counted.
Core Probability & Counting Formulas
Check whether events overlap, are independent, are conditional, or require ordered or unordered counting before choosing a formula.
Complement
P(not A) = 1 − P(A)The probability that event A does not occur.
Addition Rule
P(A or B) = P(A) + P(B) − P(A and B)Subtract the overlap so outcomes belonging to both events are not counted twice.
Conditional Probability
P(A given B) = P(A and B) ÷ P(B)Requires P(B) > 0. It asks for the probability of A after B is known to have occurred.
Independent Events
P(A and B) = P(A) × P(B)Use this multiplication form when A and B are independent.
Odds in Favour
Odds = P(A) : [1 − P(A)]Odds compare occurrence with non-occurrence and should not be confused with probability itself.
Probability From Odds a:b
P(A) = a ÷ (a + b)Here a represents favourable odds units and b represents unfavourable odds units.
Expected Value
E(X) = Σ[xᵢ × P(xᵢ)]xᵢ = possible numerical outcome; P(xᵢ) = probability of that outcome.
Factorial
n! = n × (n − 1) × … × 1By definition, 0! = 1. Factorials are used in permutations and combinations.
Permutations — Order Matters
P(n,r) = n! ÷ (n − r)!n = number of available items; r = number arranged or selected in order.
Combinations — Order Does Not Matter
C(n,r) = n! ÷ [r!(n − r)!]Use combinations when changing the order does not create a different selection.
Statistical Symbols & Variables
Keep sample and population notation consistent. These symbols are used throughout the formulas above.
Manual Calculation Procedures
These procedures show the calculation order. Request 4 will apply the same process to full numerical examples.
Arithmetic Mean
Use when every observation receives equal weight.
- Add all numerical observations.
- Count the observations.
- Divide the sum by the count.
- Check that no observation was omitted or counted twice.
- Interpret the result as the arithmetic centre.
Median
Use ranked position rather than arithmetic magnitude.
- Place all observations in ascending order.
- Count the observations.
- If n is odd, identify the single middle value.
- If n is even, average the two middle values.
- Interpret the median as the central ranked position.
Variance & Standard Deviation
Decide sample or population before choosing the divisor.
- Calculate the appropriate mean.
- Subtract the mean from every observation.
- Square each deviation.
- Add the squared deviations.
- Divide by N for population variance or n − 1 for sample variance.
- Take the square root to obtain standard deviation.
Z-Score
Standardize a value relative to a mean and standard deviation.
- Identify the raw value x.
- Identify the applicable mean μ.
- Identify the applicable standard deviation σ.
- Calculate x − μ.
- Divide the difference by σ.
- Interpret the sign and distance in standard deviations.
Confidence Interval for a Mean
Select the statistical procedure before calculating the bounds.
- Identify the point estimate, such as x̄.
- Check assumptions and whether σ is known or estimated by s.
- Calculate the applicable standard error.
- Select z or t and determine the critical value.
- Calculate ME = critical value × SE.
- Calculate estimate − ME and estimate + ME.
- Interpret the interval using the chosen procedure and confidence level.
Probability of Combined Events
Identify the event relationship before adding or multiplying.
- Define events A and B clearly.
- Decide whether you need “A or B” or “A and B”.
- Check whether events overlap, are conditional or are independent.
- Choose the addition, conditional or multiplication rule that applies.
- Substitute the probabilities.
- Check that the final probability is between 0 and 1 inclusive.
Expected Value
Find the probability-weighted average of possible outcomes.
- List each numerical outcome xᵢ.
- List its corresponding probability P(xᵢ).
- Check that the probabilities form the intended probability model.
- Multiply each outcome by its probability.
- Add the probability-weighted outcomes.
- Interpret the result as a long-run model average, not a guaranteed next outcome.
Permutation or Combination
Start by asking whether order changes the outcome.
- Identify n, the number of available items.
- Identify r, the number selected or arranged.
- Ask whether changing order creates a different outcome.
- If yes, use the permutation formula.
- If no, use the combination formula.
- Check that n and r satisfy the requirements of the counting problem.
Units, Conventions & Interpretation
Statistical outputs do not all use the same units. Check the meaning of the result before reporting it.
| Calculation | Units / scale | Convention to check | Interpretation |
|---|---|---|---|
| Mean / Median | Usually same units as numerical data | Dataset and measure of centre | Central or representative value |
| Variance | Squared data units | Sample or population | Squared measure of spread |
| Standard deviation | Same units as data | Sample or population | Spread around the mean |
| Z-score | Unitless | Correct mean and SD | Standard deviations from the mean |
| Standard error | Same units as the statistic for a mean | Correct SE formula and sample size | Sampling variability of an estimate |
| Confidence interval for a mean | Same units as the estimated mean | Confidence level, distribution and assumptions | Interval produced by the stated inferential procedure |
| Probability | 0 to 1, or equivalent percentage | Event definition and probability model | Likelihood under the specified model |
| Odds | Ratio | In favour or against | Occurrence compared with non-occurrence |
| Expected value | Same units as numerical outcomes | Probabilities and outcome definitions | Probability-weighted model average |
| Permutations / combinations | Whole-number count | Whether order matters | Number of possible arrangements or selections |
Assumptions, Edge Cases & Precision
Correct arithmetic is only one part of a statistically useful result.
Check the Method
- Separate sample formulas from population formulas.
- Check independence where required.
- Check sampling and distributional conditions for inference.
- Use the correct degrees of freedom for t methods.
- Do not assume entering valid numbers makes an inferential method appropriate.
Watch for Edge Cases
- An empty dataset cannot produce an ordinary mean.
- A dataset may have multiple modes or no mode.
- A zero standard deviation prevents division by σ in a z-score.
- Conditional probability requires a non-zero conditioning probability.
- Outliers should not be silently removed.
- Permutation and combination inputs must fit the counting problem.
Keep Enough Precision
- Do not round intermediate means too early.
- Retain precision in squared deviations.
- Retain precision in standard errors and critical values.
- Round displayed probabilities sensibly.
- Round confidence bounds only after the main calculation.
- Do not report more decimal places than the context can support.
Check Your Manual Calculation With a Related Tool
Use the dedicated calculator after you understand which method applies. The tool should support the calculation, not replace method selection or interpretation.
Worked Examples & Practical Applications
See Statistics & Probability Calculations Worked Out Step by Step
The examples below turn the formulas into complete calculations. Each one follows the same path: Formula → Identify values → Substitute → Calculate → Interpret. The aim is not only to get an answer, but to understand what the answer means and what the calculation assumes.
1. Centre & Averages
These examples show how different measures of centre answer different questions about the same type of data.
Example 1 · Arithmetic Mean
Find the Mean of Four Measurements
A set of four measurements is 4, 7, 9 and 10. What is the arithmetic mean?
Formula
x̄ = Σx ÷ nIdentify the values
Observations = 4, 7, 9, 10
n = 4
Substitute
x̄ = (4 + 7 + 9 + 10) ÷ 4Calculate
x̄ = 30 ÷ 4 = 7.5Interpret
The arithmetic mean of the four observations is 7.5.
Example 2 · Weighted Average
Calculate a Weighted Course Result
A course has three assessments: 70 worth 20%, 80 worth 30%, and 90 worth 50%.
Formula
Weighted Mean = Σ(wx) ÷ ΣwIdentify the values
70 with weight 0.20
80 with weight 0.30
90 with weight 0.50
Substitute
Weighted Mean = [70(0.20) + 80(0.30) + 90(0.50)] ÷ (0.20 + 0.30 + 0.50)Calculate
= (14 + 24 + 45) ÷ 1.00 = 83Interpret
The weighted course result is 83. The final assessment has the greatest effect because it carries 50% of the total weight.
2. Spread & Variability
Averages tell you about centre. Variance and standard deviation show how observations are spread around that centre.
Example 3 · Population Variance & Standard Deviation
Measure the Spread of Four Values
Treat 2, 4, 6 and 8 as the complete population of interest. Find the population variance and population standard deviation.
Formula
σ² = Σ(x − μ)² ÷ N σ = √σ²Identify the values
Population = 2, 4, 6, 8
N = 4
Find and square each deviation
| x | x − μ | (x − μ)² |
|---|---|---|
| 2 | −3 | 9 |
| 4 | −1 | 1 |
| 6 | 1 | 1 |
| 8 | 3 | 9 |
Calculate
σ² = 20 ÷ 4 = 5 σ = √5 ≈ 2.236Interpret
The population variance is 5 squared units. The population standard deviation is about 2.24 units.
3. Relative Position & Statistical Uncertainty
Standardized scores compare a value with a distribution. Confidence intervals use sample information to quantify uncertainty under a stated statistical procedure.
Example 4 · Z-Score
How Far Is a Score From the Mean?
A value is 85. The relevant population mean is 70 and the population standard deviation is 10.
Formula
z = (x − μ) ÷ σIdentify the values
x = 85
μ = 70
σ = 10
Substitute
z = (85 − 70) ÷ 10Calculate
z = 15 ÷ 10 = 1.5Interpret
The value 85 lies 1.5 standard deviations above the specified population mean.
Example 5 · 95% Z Confidence Interval
Estimate a Population Mean With Known Population SD
Suppose a sample has mean 50, the known population standard deviation is 12, and the sample size is 36. Under the applicable z-method conditions, calculate a two-sided 95% confidence interval.
Formula
CI = x̄ ± z* × (σ ÷ √n)Identify the values
x̄ = 50
σ = 12
n = 36
z* ≈ 1.96 for this two-sided 95% z procedure
Calculate the standard error
SE = 12 ÷ √36 SE = 12 ÷ 6 = 2Calculate the margin of error and bounds
ME = 1.96 × 2 = 3.92 Lower = 50 − 3.92 = 46.08 Upper = 50 + 3.92 = 53.92Interpret
The calculated 95% confidence interval is 46.08 to 53.92, using this z-based procedure and its assumptions.
4. Probability, Expected Value & Counting
These examples show why event relationships and the way possible outcomes are counted must be identified before calculating.
Example 6 · Independent Events
Find the Probability That Two Independent Events Both Occur
Event A has probability 0.60 and event B has probability 0.50. Assume the events are independent.
Formula
P(A and B) = P(A) × P(B)Identify the values
P(A) = 0.60
P(B) = 0.50
Events are stated to be independent.
Substitute
P(A and B) = 0.60 × 0.50Calculate
P(A and B) = 0.30 0.30 = 30%Interpret
Under the stated probability model, the probability that both events occur is 0.30, or 30%.
Example 7 · Expected Value
Calculate the Expected Value of a Simple Outcome
An outcome pays £10 with probability 0.30 and £0 with probability 0.70.
Formula
E(X) = Σ[xᵢ × P(xᵢ)]Identify the values
Outcome 1 = £10, probability = 0.30
Outcome 2 = £0, probability = 0.70
Substitute
E(X) = £10(0.30) + £0(0.70)Calculate
E(X) = £3 + £0 = £3Interpret
The expected value is £3 per trial in the long-run mathematical model.
Example 8 · Combinations
How Many Groups of 3 Can Be Chosen From 8 People?
Choose 3 people from a group of 8. The order of selection does not matter.
Formula
C(n,r) = n! ÷ [r!(n − r)!]Identify the values
n = 8 available people
r = 3 people selected
Order does not matter.
Substitute
C(8,3) = 8! ÷ [3!(8 − 3)!] = 8! ÷ (3! × 5!)Calculate
C(8,3) = (8 × 7 × 6) ÷ (3 × 2 × 1) = 336 ÷ 6 = 56Interpret
There are 56 different unordered groups of 3 that can be chosen from 8 people.
Where These Calculations Are Used
The same mathematical methods can appear in very different settings. The statistical question determines which method is appropriate.
Education & Coursework
Mean, median, weighted grades, variability, probability, standardized scores and confidence intervals.
Research & Surveys
Descriptive summaries, sampling variability, standard errors, confidence intervals and weighted data.
Business & Quality Control
Performance averages, process variability, uncertainty, probability and repeated-outcome analysis.
Science & Data Analysis
Measurements, spread, standardized observations, sampling uncertainty and probability models.
Worked Example Summary
Use this table to connect the question, required inputs, method and related calculator.
| Question | Main inputs | Method | Example result | Related tool |
|---|---|---|---|---|
| What is the average? | 4, 7, 9, 10 | Arithmetic mean | 7.5 | Central Tendency Calculator |
| What is the weighted result? | 70 at 20%, 80 at 30%, 90 at 50% | Weighted mean | 83 | Central Tendency Calculator |
| How spread out is the population? | 2, 4, 6, 8 | Population variance and SD | Variance 5; SD ≈ 2.24 | Variance & Standard Deviation Calculator |
| How far is a value from the mean? | x = 85, μ = 70, σ = 10 | Z-score | z = 1.5 | Z-Score & Confidence Interval Tool |
| What is the 95% interval? | x̄ = 50, σ = 12, n = 36 | 95% z confidence interval | 46.08 to 53.92 | Z-Score & Confidence Interval Tool |
| What is the chance both independent events happen? | P(A) = 0.60, P(B) = 0.50 | Independent-event multiplication | 0.30 or 30% | Probability & Expected Value Calculator |
| What is the expected value? | £10 at 0.30; £0 at 0.70 | Probability-weighted mean | £3 | Probability & Expected Value Calculator |
| How many unordered groups are possible? | n = 8, r = 3 | Combination | 56 | Probability & Expected Value Calculator |
Featured Article
Tool Selection & Related Calculators
Which Statistics or Probability Calculator Should You Use?
You do not need to know the formula name before you start. First decide what you are trying to find: the centre of data, its spread, the position of a value, statistical uncertainty, or the probability of an outcome. Then choose the calculator that matches that question.
Start With the Question, Not the Formula
Statistics contains formulas that can look similar while answering different questions. Choosing the correct method starts with understanding the data and the result you need.
Find Your Calculation by Question
Choose the closest match. You can move to a more detailed statistical method later if your problem needs it.
A Simple Tool-Selection Journey
Use these four checks before entering numbers into a calculator.
Statistics & Probability Calculator Directory
The pillar is organised around four main calculator pathways. Each tool handles a related group of statistical questions without forcing every calculation into one oversized tool.
Centre & Averages
Central Tendency Calculator
Use this when the main question is about the centre, typical value or weighted average of a dataset.
Mean, median, mode, weighted average, frequency-weighted mean or a missing value from a known target mean.
Choose carefully: the mean uses every value and can be affected strongly by extreme observations. The median is based on ranked position and may respond differently to outliers.
Spread & Variability
Variance & Standard Deviation Calculator
Use this when you want to measure how observations vary around their centre.
Population or sample variance, standard deviation, mean absolute deviation, supported median absolute deviation or a full deviation breakdown.
Important: decide whether the observations represent the complete population of interest or a sample used to estimate population variability. That choice changes the variance and standard-deviation calculation.
Distributions & Uncertainty
Z-Score & Confidence Interval Tool
Use this for standardized position, supported distribution probabilities, critical values, standard errors and confidence intervals.
A z-score, raw score from z, normal-distribution percentile, standard error, z or t critical value, margin of error or a supported confidence interval.
Do not treat these as one calculation: standardizing an observation, finding a distribution probability, selecting a critical value and building a confidence interval are related but distinct tasks.
Probability & Combinatorics
Probability & Expected Value Calculator
Use this when the question involves chances, event relationships, odds, expected outcomes or counting possible arrangements and selections.
Basic or compound probability, complement, conditional probability, probability and odds, expected value, permutations or combinations.
Check the event structure first: independent and mutually exclusive events are not the same. For counting problems, ask whether changing the order creates a different outcome.
Which Calculation Should I Use?
Use the question in the first column to identify the method before choosing a calculator.
| If you want to… | Use | Check first | Calculator |
|---|---|---|---|
| Find the arithmetic centre | Mean | Numerical data; consider whether extreme observations affect usefulness. | Central Tendency Calculator |
| Find the middle ranked value | Median | Sort the observations first. | Central Tendency Calculator |
| Find the most frequent value | Mode | A dataset may have one mode, several modes or no mode. | Central Tendency Calculator |
| Give some observations more influence | Weighted mean | Use the correct value-weight pairs. | Central Tendency Calculator |
| Measure overall squared spread | Variance | Population or sample? | Variance & SD Calculator |
| Measure spread in the original units | Standard deviation | Population or sample? | Variance & SD Calculator |
| Standardize a value relative to a mean and SD | Z-score | Confirm the mean and standard deviation used for comparison. | Z-Score & CI Tool |
| Estimate uncertainty around a mean | Confidence interval | Distribution, sample size, SD information, confidence level and assumptions. | Z-Score & CI Tool |
| Find the chance of an event | Probability | How outcomes are counted and how events relate to each other. | Probability Calculator |
| Find a probability-weighted average outcome | Expected value | Possible outcomes and their probabilities. | Probability Calculator |
| Count selections where order does not matter | Combination | Confirm that rearranging the selected items does not create a new outcome. | Probability Calculator |
| Count arrangements where order matters | Permutation | Confirm that a different ordering counts as a different outcome. | Probability Calculator |
Similar Terms That Need Different Calculations
These distinctions prevent some of the most common tool-selection errors.
Mean uses every value. Median uses ranked position. Extreme values can affect them differently.
Variance uses squared units. Standard deviation is expressed in the original measurement units.
This distinction changes formulas such as variance and standard deviation.
SD describes variation among observations. SE describes uncertainty in a sample statistic such as a mean.
A z-score counts standard deviations from a mean. It is not itself a percentage or percentile.
A confidence interval for a mean concerns uncertainty about a parameter. A prediction interval concerns a future individual observation under a model.
Both can describe event likelihood, but they use different numerical representations.
Independent events do not alter each other’s probability. Mutually exclusive events cannot occur together.
Permutation: order matters. Combination: order does not matter.
Mistakes, Limitations & FAQ
Common Statistics & Probability Mistakes — and How to Avoid Them
A calculator can perform the arithmetic correctly while the statistical method, data or interpretation is wrong. Before trusting a result, check what was calculated, what the data represent, which assumptions apply and what the result actually means.
Common Calculation & Interpretation Mistakes
Use these checks when your answer looks unusual, differs from another calculator or does not match a manual result.
Averages
Dividing a mean by the wrong count
The arithmetic mean divides the sum of the observations by the number of observations included.
Averages
Finding the median before sorting the values
Median is based on ranked position, not the original order in which the values were entered.
Averages
Assuming every dataset has exactly one mode
A dataset can have one mode, more than one mode or no repeating value that qualifies as a mode.
Weighted Data
Forgetting the total weight
A weighted average is based on weighted values divided by the total weight. Do not assume weights always sum to 1 or 100 unless they actually do.
Spread
Mixing sample and population variance
Population and sample variance are not interchangeable. A sample variance used to estimate population variability commonly uses n − 1 in the denominator.
Spread
Forgetting to square the deviations
Variance uses squared deviations. Simply adding signed deviations around the mean will not produce variance.
Spread
Confusing variance with standard deviation
Variance is expressed in squared units. Standard deviation is the square root of variance and returns to the original measurement units.
Spread
Using “MAD” without defining it
The abbreviation can refer to different absolute deviation measures in different contexts.
Inference
Confusing standard deviation with standard error
Standard deviation describes variability among observations. Standard error describes uncertainty in a sample statistic such as a sample mean.
Distributions
Treating a z-score as a percentage
A z-score expresses how many standard deviations a value lies above or below a specified mean.
Distributions
Confusing percentile with percentage
A percentage is a proportion out of 100. A percentile describes relative position within a dataset or distribution.
Confidence
Using the wrong critical value
Critical values can depend on the distribution, confidence or significance level, tail choice and, for a t distribution, degrees of freedom.
Confidence
Treating a 95% confidence interval as a guarantee
A confidence interval is produced by a statistical procedure with stated assumptions. It is not a promise that one calculated interval must contain the true value.
Probability
Multiplying dependent probabilities as if independent
The simple multiplication P(A) × P(B) for both events requires independence. Dependent events need conditional probability information.
Probability
Adding probabilities without considering overlap
When two events can both occur, their overlap must be considered when finding the probability of A or B.
Probability
Confusing independent with mutually exclusive
Independent events do not change each other’s probability. Mutually exclusive events cannot occur together.
Counting
Mixing permutations and combinations
A permutation counts arrangements where order matters. A combination counts selections where order does not.
Expected Value
Expecting the expected value on the next trial
Expected value is a probability-weighted average over possible outcomes. It does not need to equal any individual possible outcome.
What a Statistics Calculator Cannot Tell You by Itself
Calculation and statistical validity are different. These limitations become especially important when using sample data to make claims about a larger population.
Accurate arithmetic cannot make a poorly selected sample representative of the population you want to study.
A calculation may still be possible, but uncertainty and the suitability of particular methods can depend on sample size and other conditions.
Results depend on which observations are available and how missing observations are handled.
A calculator normally treats the numbers entered as the intended inputs. Incorrect observations can therefore produce a correctly calculated but wrong result.
Repeated, paired, clustered or otherwise related observations may require methods that account for that dependence.
Some probability and inferential procedures depend on a particular model or on conditions that make the chosen approximation appropriate.
An estimate is difficult to interpret if it is unclear which wider group the sample is intended to represent.
Repeatedly examining many hypotheses or comparisons can create issues that are not represented by one isolated calculator result.
A numerical relationship or difference does not by itself establish why that relationship exists.
Outliers: Do Not Delete Them Automatically
An unusual observation may be an error, a valid extreme case or important information about the data. Removing it can materially change the mean, variance, standard deviation and confidence interval.
Rounding: Keep Precision Until the End
Rounding intermediate values too early can alter later calculations, especially standard errors, squared deviations, critical-value calculations, tail probabilities and confidence bounds.
Computing a Statistic vs Drawing a Statistical Conclusion
Keeping these two stages separate helps prevent a correct calculation from being given an unsupported interpretation.
Computing the Statistic
This is mainly the mathematical stage. For example, a calculator can calculate a mean, variance, z-score, probability or confidence interval from supplied inputs.
- Check the formula or method.
- Check the numerical inputs.
- Check units and notation.
- Keep sufficient internal precision.
Drawing a Wider Conclusion
This requires more than arithmetic. The conclusion may depend on how observations were obtained and whether the method’s assumptions fit the situation.
- Was sampling suitable?
- Are observations appropriately independent?
- Are distributional conditions reasonable?
- Is the population clearly defined?
- Are missing values and unusual observations understood?
Statistics & Probability FAQ
Short answers to common questions about choosing, calculating and interpreting statistical results.
Which average should I use: mean, median or mode?
It depends on the question and data. Mean uses every numerical value. Median uses the middle ranked position and is generally less sensitive to extreme values. Mode identifies the most frequent value or values.
Can a dataset have more than one mode?
Yes. More than one value can share the highest frequency. A dataset may also have no repeated value that provides a useful mode.
What is the difference between sample and population standard deviation?
Population standard deviation describes the complete population of interest. Sample standard deviation is calculated from a sample and commonly uses the n − 1 adjustment when estimating population variability.
Why does variance use squared units?
Variance averages squared deviations from the mean, so its units are squared. Standard deviation takes the square root of variance and returns to the original measurement units.
Are standard deviation and standard error the same?
No. Standard deviation describes variability among observations. Standard error describes uncertainty in a sample statistic, such as a sample mean.
Is a z-score the same as a percentile?
No. A z-score describes standardized distance from a mean. A percentile describes relative position. A distribution calculation can be used to connect the two when the relevant model is appropriate.
Is a percentile the same as a percentage?
No. A percentage expresses a proportion out of 100. A percentile describes relative position within a dataset or distribution. A score of 80% therefore does not automatically mean the 80th percentile.
Should I always use 1.96 for a 95% confidence interval?
No. About 1.96 is a commonly used critical value for an appropriate two-sided 95% z procedure. Critical values depend on the selected distribution, confidence level or significance level, tail choice and, for t procedures, degrees of freedom.
Does a 95% confidence interval mean there is a 95% chance the true value is inside it?
Not under the standard frequentist interpretation for an already-calculated interval. The interval is produced by a procedure designed to achieve its stated coverage rate over repeated samples when the method’s assumptions hold.
What is the difference between a confidence interval and a prediction interval?
A confidence interval for a mean concerns uncertainty about a population parameter. A prediction interval concerns a future individual observation under a model. They should not be treated as interchangeable.
Are independent events the same as mutually exclusive events?
No. Independent events do not change one another’s probabilities. Mutually exclusive events cannot both occur in the same trial or situation being modelled.
When can I multiply two probabilities?
For independent events, the probability that both occur can be found by multiplying their probabilities. For dependent events, the relevant conditional probability must be taken into account.
What is the difference between probability and odds?
They are different ways of representing event likelihood. For example, probability 0.25 corresponds to odds in favour of 0.25:0.75, which simplifies to 1:3.
When should I use a permutation instead of a combination?
Use a permutation when changing the order creates a different outcome. Use a combination when only the selected items matter and their order does not.
Is expected value the most likely outcome?
Not necessarily. Expected value is a probability-weighted average over possible outcomes and does not need to equal any single possible outcome.
Should a calculator automatically remove outliers?
No. An unusual value should be investigated rather than silently removed. If an observation is excluded, the exclusion should be explicit because it can materially change statistical results.
Why does my answer differ slightly from another calculator?
Differences can come from rounding, sample-versus- population settings, critical-value methods, definitions of a statistic or different handling of inputs. Compare the method and settings before assuming either result is wrong.
Can a statistically correct calculation still be misleading?
Yes. Correct arithmetic does not fix biased sampling, poor data quality, dependence, missing observations, unsuitable assumptions or an interpretation that goes beyond what the data support.