What Is Accuracy In Maths?
Accuracy in maths is how close a measured or calculated value is to the true, exact, or accepted value. A result of 9.98 for a true value of 10 is accurate. A result of 7 for the same true value is not.
Accuracy is always measured against something: a known answer, an agreed standard, or an exact quantity. Without a true value to compare to, "accurate" has no meaning.
Three ideas do the work of accuracy across school maths and beyond:
Significant figures tell you how many digits of a value are trustworthy.
Rounding expresses a value to a chosen degree of accuracy.
Error measures the size of the gap between your value and the true one.
The rest of this page takes each in turn, then shows why the whole idea matters.
What Is The Difference Between Accuracy And Precision?
Accuracy is closeness to the true value. Precision is closeness of repeated results to each other. They sound alike and get swapped constantly, but they answer two different questions.
Think about the dartboard in the picture above. Darts scattered evenly around the bullseye are accurate on average but not precise. Darts packed into one tight cluster in the wrong corner are precise but not accurate. Only darts grouped tightly on the bullseye are both.
Table: Accuracy and precision answer two different questions.
Idea | Question it answers | Dartboard picture |
|---|---|---|
Accuracy | How close am I to the true value? | Darts near the bullseye |
Precision | How close are my results to each other? | Darts clustered together |
A measurement can be precise and still be wrong. A scale that always reads 2 grams over is very precise, and consistently inaccurate. This is why every measuring unit and instrument is checked against a known standard, not just against itself.
What Are Significant Figures?
Significant figures are the digits of a number that carry real information about its value. The more significant figures a value has, the more accurately it is stated.
The rules for counting them are short:
Every non-zero digit is significant. In $45.7$ there are three significant figures.
Zeros between non-zero digits are significant. In $3.05$ there are three.
Leading zeros are never significant, they only mark place value. In $0.003$ the first significant figure is the $3$, so it has one significant figure.
Trailing zeros after a decimal point are significant. In $2.50$ there are three.
That third rule is the one students miss most. A leading zero looks like a digit, but it is only holding a decimal place open. It tells you nothing about how accurate the value is.
How Do You Round To A Degree Of Accuracy?
To round a number, keep the digits up to your chosen degree of accuracy, then look at the very next digit: if it is $5$ or more, round up; if it is $4$ or less, round down. The degree of accuracy is usually given as a number of decimal places or a number of significant figures.
Example 1: Round $3.14159$ to 3 significant figures.
Count three significant figures from the first non-zero digit: $3$, $1$, $4$. The next digit is $1$, which is less than $5$, so round down.
$$3.14159 \approx 3.14 ; \text{(3 s.f.)}$$
Example 2: Round $0.004073$ to 2 significant figures.
Leading zeros do not count. The first significant figure is $4$, the second is $0$. The next digit is $7$, which is $5$ or more, so round up.
$$0.004073 \approx 0.0041 ; \text{(2 s.f.)}$$
Example 3: Round $5821$ to 2 significant figures.
The first two significant figures are $5$ and $8$. The next digit is $2$, so round down, and hold the place values with zeros.
$$5821 \approx 5800 ; \text{(2 s.f.)}$$
When a question says "give your answer to a suitable degree of accuracy," 3 significant figures is the usual safe choice. Rounding decimals is closely tied to writing values as fractions, decimals and percentages, since the same value can be expressed in each form to a chosen degree of accuracy.
What Are Absolute Error And Relative Error?
Error measures the size of the gap between a value and the truth. There are two ways to report it, and they answer different questions.
Absolute error is the plain size of the gap, in the same units as the measurement:
$$\text{Absolute error} = \lvert \text{true value} - \text{measured value} \rvert$$
Relative error is that gap compared to the true value, written as a ratio or, multiplied by $100$, as a percent:
$$\text{Relative error} = \frac{\lvert \text{true value} - \text{measured value} \rvert}{\text{true value}}$$
Example 4: A length is measured as $48$ cm; the true value is $50$ cm.
Absolute error first:
$$\lvert 50 - 48 \rvert = 2 \text{ cm}$$
Then relative error, and the percentage error:
$$\frac{2}{50} = 0.04 = 4%$$
Why keep both? An absolute error of $2$ cm is large on a $5$ cm pencil and tiny on a $500$ cm room. Relative error tells you which case you are in, because it scales the gap against the size of the thing being measured. Absolute error tells you the raw size; relative error tells you whether that size matters.
Why Does Accuracy Matter In Maths?
Accuracy matters because a wrong-by-a-little answer is still wrong, and small errors do not stay small. Two forces make this true.
Small errors compound. Round too early in a chain of steps, and each later step builds on a value that was already off. A $1%$ slip repeated across several multiplications can grow into a result that is visibly wrong.
Real decisions sit on the last digit. A dose, a dimension, a bank balance, a launch angle, each is acted on exactly as written. The number is not a suggestion, it is an instruction, and the degree of accuracy decides whether the instruction is safe.
There is a second, quieter reason. Stating a value to the right number of significant figures is a form of honesty. Writing $3.14159$ when you only measured to the nearest tenth claims an accuracy you never had. Good mathematicians round to the accuracy they can actually defend, and no further.
Who Shaped The Maths Of Accuracy?
Controlling accuracy is one of the oldest jobs in mathematics. Long before decimals existed, mathematicians were finding ways to pin a value between two bounds and prove it could not escape.
One more mathematician shaped how we handle the errors that measurement leaves behind:
Carl Friedrich Gauss (1777–1855, Germany) developed the theory of errors and the method of least squares, and in 1801 used it to predict where the newly lost dwarf planet Ceres would reappear in the sky. His methods are still the backbone of how scientists squeeze the most accurate answer out of imperfect data.
Where Is Accuracy Used In The Real World?
The same three tools, significant figures, rounding, and error, run quietly under a wide range of work.
Medicine: a drug dose is rounded to a safe degree of accuracy, and being off by a small relative error can change its effect.
Engineering and construction: parts are cut to a stated tolerance, which is an allowed absolute error, so pieces made in different places still fit together.
Science labs: every measurement is reported to the significant figures the instrument can justify, never more.
Money and banking: currency is rounded to two decimal places, and rounding at the right step keeps totals honest across millions of transactions.
Space and navigation: a tiny error in an angle, left uncorrected, becomes a huge error over a long distance.
One idea, closeness to the truth, connects a pharmacist, a bridge builder, a lab technician, and a spacecraft. Accuracy is the shared discipline underneath all of them.
What Are The Most Common Accuracy Mistakes?
These three errors account for most lost marks on accuracy questions, verified against GCSE rounding notes, Mathematics LibreTexts, and error-analysis guides.
Confusing Accuracy With Precision
Where it slips in:
A student reports a tightly repeated set of readings as "accurate" without ever checking them against the true value.
Don't do this:
Do not treat consistent results as correct results. Repeatable and right are two different claims.
The correct way:
Ask both questions separately. Precision asks whether the readings agree with each other; accuracy asks whether they agree with the true value. Confirm accuracy against a known standard.
Rounding Too Early in a Calculation
Where it slips in:
A student rounds every intermediate step, then rounds again at the end, so each step feeds a slightly wrong value into the next.
Don't do this:
Do not round partway through a multi-step calculation. Early rounding lets small errors compound into a visibly wrong final answer.
The correct way:
Keep full accuracy through every step and round only the final answer to the stated degree of accuracy. Use the unrounded value, or the calculator's stored value, in between.
Miscounting significant figures
Where it slips in:
A student counts a leading zero as a significant figure, so $0.0041$ gets read as having more accuracy than it really has.
Don't do this:
Do not count leading zeros. They mark place value only and carry no information about accuracy.
The correct way:
Start counting from the first non-zero digit. In $0.0041$ the first significant figure is $4$ and the second is $1$, so it has two significant figures.
A Real-World Version Of The Rounding Mistake
Rounding too early is not only a homework slip. The wrong-path calculation below shows how fast it grows. Suppose you need $\sqrt{2} \times \sqrt{8}$.
The exact answer is clean:
$$\sqrt{2} \times \sqrt{8} = \sqrt{16} = 4$$
Now do it the rushed way, rounding each root to one decimal place first:
$$\sqrt{2} \approx 1.4, \qquad \sqrt{8} \approx 2.8$$
$$1.4 \times 2.8 = 3.92$$
The rounded route lands on $3.92$ when the true answer is exactly $4$. A gap of $0.08$ appeared from nothing but early rounding, and in a longer calculation that gap keeps growing. Carry the exact values, round last.
Practice Problems On Accuracy
Try these, then check the answer beside each.
Round $7.4652$ to 2 significant figures. (Answer: $7.5$.)
Round $0.006189$ to 3 significant figures. (Answer: $0.00619$.)
How many significant figures are in $0.0308$? (Answer: three, the $3$, the middle $0$, and the $8$.)
A value is measured as $19$ g; the true value is $20$ g. Find the absolute error. (Answer: $\lvert 20 - 19 \rvert = 1$ g.)
For problem 4, find the relative error as a percentage. (Answer: $\tfrac{1}{20} = 0.05 = 5%$.)
Round $2.999$ to 2 significant figures. (Answer: $3.0$, the carry ripples all the way up.)
Where Should You Go Next After Accuracy?
Accuracy connects to almost every number skill, and a few natural doors open from here.
Percent. Percentage error is relative error written as a percentage, so this is the next tool for describing how large an error really is.
Fractions, Decimals And Percentages. Rounding lives inside these three forms; the same value gets stated to a chosen degree of accuracy in each.
Number. Significant figures are a way of reading the digits of a number, so a firm grip on place value makes accuracy far easier.
If your child is building these number foundations, a live Bhanzu trainer teaches accuracy starting from the "why" (the true value behind every measurement) in the Bhanzu algebra program.
Was this article helpful?
Your feedback helps us write better content