Rounding, Estimation and Accuracy

Q: What is the general rule for deciding whether to round a target digit up or let it stay the same?

Look at the decider digit immediately to the right: if it is \(0, 1, 2, 3,\) or \(4\), round down (keep the target digit unchanged); if it is \(5, 6, 7, 8,\) or \(9\), round up (add \(1\) to the target digit).

Q: Why are placeholder zeros essential when rounding large whole numbers?

Placeholder zeros preserve the correct place value of the digits; without them, the magnitude of the number changes drastically (e.g. \(67,420\) would become \(67\) instead of \(67,000\)).

Q: How do you round a number to a given number of decimal places (d.p.)?

Count the specified number of digits to the right of the decimal point, check the decider digit directly to the right of the cut-off line, round up or down accordingly, and keep exactly that number of decimal digits.

Q: When must you include a trailing zero after rounding to a specific number of decimal places?

You must keep trailing zeros whenever necessary to show the exact precision requested by the question (e.g. \(3.198\) rounded to \(2\text{ d.p.}\) must be written as \(3.20\)).

Q: How do you identify the first significant figure (1st s.f.) in any number?

The first significant figure is always the first non-zero digit when reading the number from left to right.

Q: What is the standard KS3 strategy for estimating the answer to a calculation?

Round every number in the calculation to \(1\text{ significant figure}\) (1 s.f.) first, then perform the simplified calculation using the approximation symbol \(\approx\).

Q: How do you calculate the upper and lower bounds of a rounded measurement?

Find the unit of accuracy, divide it by \(2\) to find the maximum error, then subtract this half-unit to find the Lower Bound (LB) and add it to find the Upper Bound (UB).

Q: How is an error interval written using inequality notation?

An error interval is written in the form \(\text{Lower Bound} \le x

Mathematics · Key Stage 3 (KS3) · 9 min read · Updated: 27 Aug 2026

Welcome to Rounding, Estimation and Accuracy!

Have you ever looked at a crowded football stadium and said, "There are about \(50,000\) people here"? Or checked a supermarket receipt and thought, "That should cost roughly \(£20\)"? If so, you have already used the skills of rounding and estimation in everyday life!

In this chapter, we will learn how to make numbers simpler to work with, how to predict answers quickly, and how to describe the exact limits of a rounded measurement. Don't worry if you find numbers confusing at times; we will break down every single rule step-by-step with clear examples.

Section 1: Rounding Whole Numbers and Powers of \(10\)

Rounding whole numbers makes large values easier to read, remember, and calculate without changing their overall size.

The Golden Rule of Rounding

When rounding to any target place value (such as the nearest \(10\), \(100\), or \(1000\)):

Step 1: Find your target digit (the place value you are rounding to).
Step 2: Look at the very next digit to its right. We call this the decider digit.
Step 3: Apply the simple rule:
• If the decider digit is \(0\), \(1\), \(2\), \(3\), or \(4\), you round down (the target digit stays exactly the same).
• If the decider digit is \(5\), \(6\), \(7\), \(8\), or \(9\), you round up (add \(1\) to the target digit).
Step 4: Fill in any remaining whole number columns up to the decimal point with placeholder zeros.

A handy rhyme to remember: "\(4\) or less, let it rest; \(5\) or more, raise the score!"

Worked Example: Whole Numbers

Round the number \(3,748\) to:

• The nearest \(10\): Target digit is \(4\) (tens). The decider is \(8\). Since \(8 \ge 5\), round up: the \(4\) becomes \(5\). Answer: \(3,750\).
• The nearest \(100\): Target digit is \(7\) (hundreds). The decider is \(4\). Since \(4 \le 4\), round down: the \(7\) stays as \(7\). Answer: \(3,700\).
• The nearest \(1000\): Target digit is \(3\) (thousands). The decider is \(7\). Since \(7 \ge 5\), round up: the \(3\) becomes \(4\). Answer: \(4,000\).

Why Placeholder Zeros Matter

Imagine rounding \(67,420\) to the nearest thousand. The target is \(7\) and the decider is \(4\). If you write just \(67\), you have changed sixty-seven thousand into sixty-seven! You must keep the placeholder zeros: \(67,000\).

Key Takeaway: Always check the decider digit directly to the right. Never forget your placeholder zeros in whole numbers to protect their place value.

Section 2: Rounding to Decimal Places (d.p.)

Decimal places refer to how many digits appear strictly to the right of the decimal point. The abbreviation d.p. is commonly used.

• \(1\) decimal place (\(1\text{ d.p.}\)): tenths column (e.g. \(0.4\))
• \(2\) decimal places (\(2\text{ d.p.}\)): hundredths column (e.g. \(0.45\))
• \(3\) decimal places (\(3\text{ d.p.}\)): thousandths column (e.g. \(0.458\))

Step-by-Step Method for Decimal Places

Step 1: Count to the required number of digits after the decimal point and draw a cut-off line.
Step 2: Look at the next digit to the right (your decider).
Step 3: If it is \(5\) or more, increase the last kept digit by \(1\). If it is \(4\) or less, leave the last kept digit unchanged.
Step 4: Drop all the digits after the cut-off line. Do not add trailing zeros unless they are needed to show precision!

Worked Examples: Decimal Places

Example A: Round \(4.638\) to \(1\text{ d.p.}\)
Count \(1\) place after the point: \(4.6 \mid 38\). Decider is \(3\). Round down \(\rightarrow 4.6\).

Example B: Round \(12.875\) to \(2\text{ d.p.}\)
Count \(2\) places after the point: \(12.87 \mid 5\). Decider is \(5\). Round up \(\rightarrow 12.88\).

Example C (The Trailing Zero Trap): Round \(3.198\) to \(2\text{ d.p.}\)
Count \(2\) places after the point: \(3.19 \mid 8\). Decider is \(8\), so \(19\) hundredths rounds up to \(20\) hundredths. We write \(3.20\).
Important: You must keep the \(0\) at the end because the question specifically asked for \(2\text{ d.p.}\)! Writing \(3.2\) only gives \(1\text{ d.p.}\)

Key Takeaway: For \(n\text{ d.p.}\), count \(n\) digits past the decimal point, check the decider, and keep exactly \(n\) digits after the point.

Section 3: Rounding to Significant Figures (s.f.)

Rounding to significant figures focuses on the most important digits in a number, regardless of how huge or tiny the number is.

How to Find the First Significant Figure

The first significant figure (1st s.f.) is always the first non-zero digit when reading from left to right.
Every digit after the 1st s.f. is also significant (even if it is a zero!).

Let's find the significant figures in two different numbers:

• In the number \(80,450\): The \(8\) is the 1st s.f., \(0\) is the 2nd s.f., \(4\) is the 3rd s.f., and \(5\) is the 4th s.f.
• In the number \(0.00506\): The leading zeros are just placeholders. The first non-zero digit is \(5\) (1st s.f.). The next digit is \(0\) (2nd s.f.), and \(6\) is the 3rd s.f.

Worked Examples: Significant Figures

Example 1: Round \(47,382\) to \(2\text{ s.f.}\)
• 1st s.f. is \(4\), 2nd s.f. is \(7\). Line: \(47 \mid 382\).
• Decider is \(3\) (round down).
• Add placeholder zeros up to the decimal point: \(47,000\).

Example 2: Round \(0.003681\) to \(1\text{ s.f.}\)
• 1st s.f. is \(3\). Line: \(0.003 \mid 681\).
• Decider is \(6\) (round up \(\rightarrow 3\) becomes \(4\)).
• Answer: \(0.004\).

Example 3: Round \(0.04096\) to \(2\text{ s.f.}\)
• 1st s.f. is \(4\), 2nd s.f. is \(0\). Line: \(0.040 \mid 96\).
• Decider is \(9\) (round up \(\rightarrow 40\) becomes \(41\)).
• Answer: \(0.041\).

Key Takeaway: Never start counting significant figures on zeros at the front of a decimal. Start at the first digit from \(1\) to \(9\).

Section 4: Estimation and Approximation

Estimation allows us to find a quick, sensible approximation to a calculation without doing long, messy arithmetic. It is an essential real-world skill for checking whether a calculator answer makes sense!

The Approximation Symbol

We use the symbol \(\approx\) to mean "is approximately equal to".
For example, we write \(19.8 \times 5.1 \approx 20 \times 5 = 100\).

The Standard KS3 Strategy: Round to \(1\text{ s.f.}\)

To estimate the answer to any calculation:
Step 1: Round every single number in the question to \(1\text{ significant figure}\).
Step 2: Perform the calculation with these simplified numbers.

Worked Examples: Estimation

Example 1: Estimate the value of \(312 \times 4.89\)
• Round \(312\) to \(1\text{ s.f.} \rightarrow 300\)
• Round \(4.89\) to \(1\text{ s.f.} \rightarrow 5\)
• Calculation: \(300 \times 5 = 1500\)
• So, \(312 \times 4.89 \approx 1500\).

Example 2: Estimate the value of \(\frac{41.6 + 18.9}{0.19}\)
• Round \(41.6\) to \(1\text{ s.f.} \rightarrow 40\)
• Round \(18.9\) to \(1\text{ s.f.} \rightarrow 20\)
• Round \(0.19\) to \(1\text{ s.f.} \rightarrow 0.2\)
• Calculation: \(\frac{40 + 20}{0.2} = \frac{60}{0.2}\)
• Dividing by \(0.2\) is the same as multiplying by \(5\): \(\frac{60}{0.2} = 300\)
• So, the estimated answer is \(300\).

Key Takeaway: For estimations, round each number to \(1\text{ s.f.}\) first before calculating. This gives you a fast sense of the correct order of magnitude.

Section 5: Accuracy, Bounds, and Error Intervals

When any measurement is rounded to a certain degree of accuracy, the true original value could have been slightly higher or slightly lower than the rounded number.

What Are Bounds?

• Lower Bound (LB): The smallest possible number that would round up to our given value.
• Upper Bound (UB): The boundary number that marks the cut-off point where numbers would round to the next value up.
• Rule for Half the Unit of Precision: The maximum possible error is always half of the unit of accuracy (\(\pm 0.5 \times \text{unit of accuracy}\)).

Finding Bounds Step-by-Step

Step 1: Identify the unit of accuracy (e.g. nearest \(10\), nearest whole number (\(1\)), nearest \(0.1\)).
Step 2: Divide that unit by \(2\).
Step 3: Subtract this value to find the Lower Bound, and add it to find the Upper Bound.

Example: A length is measured as \(8\text{ cm}\) to the nearest centimeter (\(1\text{ cm}\)).
• Half the unit: \(1\text{ cm} \div 2 = 0.5\text{ cm}\)
• Lower Bound: \(8 - 0.5 = 7.5\text{ cm}\)
• Upper Bound: \(8 + 0.5 = 8.5\text{ cm}\)

Writing Error Intervals Using Inequality Notation

To show all the possible values that the true measurement \(x\) could be, we write an error interval using inequalities:

\(\text{Lower Bound} \le x < \text{Upper Bound}\) or \(a < x \le b\)

Why is there a \(\le\) on the left and a \(<\) on the right?
• If \(x = 7.5\), it rounds UP to \(8\), so \(7.5\) is included (\(\le\)).
• If \(x = 8.5\), it would round UP to \(9\), so \(8.5\) is strictly excluded (\(<\)). Any number just under \(8.5\) (like \(8.4999\)) rounds to \(8\).
Therefore, the error interval for \(8\text{ cm}\) rounded to the nearest integer is:
\(7.5 \le x < 8.5\)

More Error Interval Examples

• A mass \(m = 60\text{ kg}\) rounded to the nearest \(10\text{ kg}\):
Unit is \(10\text{ kg}\) \(\rightarrow\) half unit is \(5\text{ kg}\).
\(\text{LB} = 60 - 5 = 55\text{ kg}\), \(\text{UB} = 60 + 5 = 65\text{ kg}\).
Error Interval: \(55 \le m < 65\)

• A distance \(d = 4.3\text{ m}\) rounded to \(1\text{ d.p.}\) (nearest \(0.1\text{ m}\)):
Unit is \(0.1\text{ m}\) \(\rightarrow\) half unit is \(0.05\text{ m}\).
\(\text{LB} = 4.3 - 0.05 = 4.25\text{ m}\), \(\text{UB} = 4.3 + 0.05 = 4.35\text{ m}\).
Error Interval: \(4.25 \le d < 4.35\)

Key Takeaway: Divide the accuracy unit by \(2\), then subtract to get the lower bound and add to get the upper bound. Express the interval as \(\text{LB} \le x < \text{UB}\).

Common Pitfalls to Avoid!

1. Counting leading zeros as significant:
In \(0.0072\), the first significant figure is \(7\), not \(0\). The zeros at the front are simply place-value holders.

2. Forgetting placeholder zeros in big numbers:
Rounding \(4,812\) to \(1\text{ s.f.}\) gives \(5,000\), not \(5\)!

3. Writing \(8.49\) or \(8.499\) instead of \(8.5\) for the upper bound:
The boundary value is exactly \(8.5\). Because we write \(x < 8.5\), the strictly-less-than symbol handles all values up to \(8.4999...\) perfectly.

4. Dropping required trailing zeros in decimals:
If a question asks for \(5.996\) to \(2\text{ d.p.}\), the answer is \(6.00\). You must keep both decimal zeros to demonstrate two decimal places of precision.

5. Rounding too early in multi-step problems:
Keep full values on your calculator during intermediate steps, and only round your final answer at the very end to avoid compounding errors.

Quick Review Summary

• Rounding Rules: Look at the decider digit directly to the right: \(0\) to \(4\) rounds down, \(5\) to \(9\) rounds up.
• Decimal Places (d.p.): Count digits strictly after the decimal point.
• Significant Figures (s.f.): Start counting from the first non-zero digit on the left.
• Estimation: Round all values to \(1\text{ s.f.}\) before calculating, using the \(\approx\) symbol.
• Error Intervals: Add and subtract half of the unit of precision to state bounds as \(a \le x < b\) or \(a < x \le b\).

Quick check

Can you answer these now?

Open each question to check the key ideas from this chapter.

What is the general rule for deciding whether to round a target digit up or let it stay the same?

Look at the decider digit immediately to the right: if it is \(0, 1, 2, 3,\) or \(4\), round down (keep the target digit unchanged); if it is \(5, 6, 7, 8,\) or \(9\), round up (add \(1\) to the target digit).

Why are placeholder zeros essential when rounding large whole numbers?

Placeholder zeros preserve the correct place value of the digits; without them, the magnitude of the number changes drastically (e.g. \(67,420\) would become \(67\) instead of \(67,000\)).

How do you round a number to a given number of decimal places (d.p.)?

Count the specified number of digits to the right of the decimal point, check the decider digit directly to the right of the cut-off line, round up or down accordingly, and keep exactly that number of decimal digits.

When must you include a trailing zero after rounding to a specific number of decimal places?

You must keep trailing zeros whenever necessary to show the exact precision requested by the question (e.g. \(3.198\) rounded to \(2\text{ d.p.}\) must be written as \(3.20\)).

How do you identify the first significant figure (1st s.f.) in any number?

The first significant figure is always the first non-zero digit when reading the number from left to right.

What is the standard KS3 strategy for estimating the answer to a calculation?

Round every number in the calculation to \(1\text{ significant figure}\) (1 s.f.) first, then perform the simplified calculation using the approximation symbol \(\approx\).

How do you calculate the upper and lower bounds of a rounded measurement?

Find the unit of accuracy, divide it by \(2\) to find the maximum error, then subtract this half-unit to find the Lower Bound (LB) and add it to find the Upper Bound (UB).

How is an error interval written using inequality notation?

An error interval is written in the form \(\text{Lower Bound} \le x < \text{Upper Bound}\).

Ready to test yourself?

Turn these notes into exam-style practice. Get unlimited AI questions on this topic with instant marking and explanations.

Practise This Topic

More Mathematics chapters

* The content provided by thinka is generated by AI and may not always be accurate or up-to-date. Please use it as a supplementary resource and verify with official materials.

Put These Notes into Practice

Reading the notes is a great start. Now practise with unlimited AI-generated questions and get instant feedback. 100,000+ students are already improving their grades.

Start Practising Now View Pricing

Done reading? Test yourself with AI practice questions

Practise This Topic Now