Math Calculators
Scientific, percentage & statistics calculators. All tools are free, instant, and updated for 2026.
Percentage Calculator
Quick percentage calculations
Use calculatorScientific Calculator
Advanced math calculations
Use calculatorFraction Calculator
Calculate with fractions
Use calculatorGPA Calculator
Calculate your grade point average
Use calculatorStatistics Calculator
Mean, median, standard deviation
Use calculatorAbout Math calculators
These are the tools where the arithmetic is not in doubt and the conventions are. Percentage change depends on which number is the base. Sample and population statistics use different denominators. A basic calculator and a scientific one give different answers to the same keystrokes because one applies operator precedence and the other does not.
Getting a wrong answer here almost always means the calculator did what you asked rather than what you meant.
The ideas these tools share
- Percentage change is not symmetric
- A 50% fall followed by a 50% rise does not return you to where you started; it leaves you 25% down. Percentages compose by multiplication, not addition, which is why percentage change and percentage-of are different operations that people routinely swap.
- Mean, median and mode answer different questions
- The mean is pulled by outliers, the median is not, and the mode describes what is most common rather than what is typical. When a distribution is skewed — incomes, house prices, response times — the mean and median can be far apart, and quoting the wrong one is how statistics mislead without anyone lying.
- Sample and population formulas differ deliberately
- Standard deviation divides by n for a population and by n minus one for a sample. That correction exists because a sample underestimates the spread of the population it came from. Using the wrong one on a small sample produces a materially wrong answer.
- Correlation is not causation, and is easy to manufacture
- Two series can correlate strongly through coincidence, a shared driver, or the way the data was selected. A correlation coefficient is a description of a pattern, never evidence of a mechanism.
- Precision is not accuracy
- Carrying a result to eight decimal places does not make it more correct than the inputs allow. If a measurement is good to two significant figures, the answer is too, no matter what the display shows.
Which one do you need?
- You need trig, logs or powers
- The scientific calculator — and check whether it is set to degrees or radians before trusting a trig result.
- You are working out an increase or decrease
- Percentage, which shows both the change and what it is a percentage of.
- You have a data set
- Statistics, which distinguishes sample from population variance.
- You are working with fractions
- The fraction calculator, which reduces and converts to decimal.
- You are tracking academic performance
- GPA, weighted by credit hours.
Where these go wrong
Adding percentages together
A 10% rise followed by a 10% fall leaves you down 1%, not level. Percentage changes compose by multiplication.
Quoting the mean of a skewed distribution
Incomes, prices and response times all have long tails, which drag the mean away from anything typical. The median is usually the honest summary.
Using the population formula on a sample
Sample standard deviation divides by n minus one for a reason. On a small sample the difference is not cosmetic.
Reporting more precision than the inputs support
A result cannot be more accurate than the measurements behind it. Extra decimal places are arithmetic, not information.
Common questions
Why does my phone give a different answer?
Almost always operator precedence or angle mode. A basic calculator evaluates left to right, so 2 + 3 × 4 gives 20; a scientific one applies PEMDAS and gives 14. For trig, check whether one is in degrees and the other in radians — sin(90) should be exactly 1 in degrees.
What is the difference between sample and population statistics?
The denominator. Population variance divides by n; sample variance divides by n−1, which is Bessel's correction and compensates for the fact that a sample underestimates the spread of the population it came from. Use sample statistics unless you genuinely have every member of the group.
Why does 0.1 + 0.2 not equal 0.3?
Because computers store numbers in binary, and one tenth is a repeating fraction in binary just as one third is in decimal. It has to be truncated. This affects every language and every spreadsheet; calculators hide it by rounding the display to fewer digits than they store.
Which average should I report?
The median, unless you have a reason to prefer the mean. Most real-world distributions are skewed, and the median describes the typical case while the mean describes the arithmetic centre. Reporting both, where they differ, is more honest than choosing one.
What does a p-value actually tell me?
How surprising the data would be if there were no real effect. It does not tell you the probability that your hypothesis is true, how large the effect is, or whether it matters. A very small p-value on a trivial effect is common in large samples.
How many decimal places should I keep?
As many as your least precise input supports, and no more. If a measurement is good to two significant figures, the result is too, whatever the calculator displays.