← Mathematics

Indices, Roots and Standard Form

Repair guide · your check broke here at step 2 · non-calculator throughout

What the check actually found

You sat 77 questions across eleven strands. In this strand — indices, roots and standard form — you were secure at step 1 and the first gap opened at step 2. Step 2 is the age 11–12 rung. That is the lowest break of all eleven strands, which is why this guide exists before any of the others.

A break that low is almost never a sign that the hard material is too hard. It is a sign that one early idea — what a power actually means — got learned as a rule instead of as a picture, and everything stacked on top of it has been wobbling ever since.

Why this one gap is expensive

It maps onto four syllabus points, and every one of them is rated high non-calculator risk:

CodeTopicWhere it bites
E1.3Powers and rootsSquares, cubes and their roots by recall
E1.7Indices IThe three laws, zero, negative, fractional
E1.8Standard formWriting it, and multiplying and dividing by hand
E1.9EstimationRounding to 1 significant figure to check an answer

Paper 2 is non-calculator: 2 hours, 100 marks, 50% of your grade. So nothing in this guide is done with a calculator, and nowhere will you be told to reach for one. Everything is by hand, every line shown.

Indices also feed upward. E2.4 Indices II (indices with algebra), surds, and every standard-form calculation in physics and chemistry all sit on this brick. One loose brick is quietly taxing several other topics at once.

The good news, and it is real

Number sense and the four operations: solid on all seven rungs. No gap anywhere. Your arithmetic mechanics work. You can add, subtract, multiply, divide, handle negatives in ordinary sums, and hold a calculation together in your head.

That matters more than it sounds. The usual pattern in a weak maths student is shaky arithmetic underneath everything, which makes every topic hard and slow to fix. Yours is the opposite: a good foundation with two loose bricks sitting on it — indices, and algebraic manipulation. That is a much easier problem, and a much faster one.

One more thing the check picked up: sign errors recurred across several different topics. Negative indices are exactly where sign errors do the most damage, so they get their own warnings in section 4.

How to use this guide

Each idea appears four times, and it gets less help each time. That is deliberate — it is the fastest way to learn a method you do not yet own.

  1. Worked in full — every step, with a reason for each one. Read it, do not skim it.
  2. Last step is yours — work out the final line before pressing the button.
  3. Last two are yours — the same, harder.
  4. All yours — type an answer and check it.

Sections 1 to 7 rebuild from below the break upward. Section 8 is a mixed set with the topics deliberately unlabelled and out of order, because in an exam nobody tells you which idea applies.

step 1–21 · What a power actually means▼

Start here, below the break. This is the idea that step 2 tested, and it is almost certainly the thing that went wrong — because there is one specific mistake nearly everybody makes at this rung.

The mistake: reading 25 as “2 times 5” and answering 10. The small raised number is not something you multiply by. It is a count of how many 2s are multiplied together. 25 = 2 × 2 × 2 × 2 × 2 = 32, not 10.
2 5 the base — the thing being multiplied the index — how many of them = 2×2×2×2×2 = 32 not 2×5 = 10 Read it out loud as “five 2s multiplied together”, never as “two five”.
The habit that fixes it: whenever you are unsure, write the multiplication out in full first. 34 becomes 3 × 3 × 3 × 3 before you work out a single thing. It costs four seconds and it removes the error completely.

The facts you must simply know

The syllabus names these explicitly, and Paper 2 assumes them. Recall, not working out.

n12345678
n21491625364964
n9101112131415
n281100121144169196225
n1234510
n31827641251000
Non-calculator: 64 appears in both tables — it is 82 and 43. That double life is why 64 turns up constantly in index questions. Also worth having: the powers of 2 up to 210 — 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024. They come up more than any other list.
Worked in full
Work out 34
1
34 = 3 × 3 × 3 × 3
Write the multiplication out. Four 3s, because the index is 4.
2
3 × 3 = 9
Take them two at a time; never try to do all four at once.
3
9 × 3 = 27
Multiply the running total by the next 3.
4
27 × 3 = 81
And the last 3. So 34 = 81.
5
Check: is 81 sensible?
3 × 4 would be 12. The answer being far bigger than 12 is exactly what you expect from a power.
Last step is yours
Work out 26
1
26 = 2 × 2 × 2 × 2 × 2 × 2
Six 2s.
2
2 × 2 = 4, then 4 × 2 = 8
Running total after three 2s.
3
8 × 2 = 16, then 16 × 2 = 32
Five 2s used.
4
32 × 2 = 64
The sixth and last 2.
Last two are yours
Work out 53 − 42
1
53 = 5 × 5 × 5
Three 5s. Powers are dealt with before the subtraction.
2
5 × 5 = 25, then 25 × 5 = 125
This is one of the cubes you are meant to recall.
3
42 = 16
Straight from the squares table.
4
125 − 16 = 109
Only now do you subtract. 125 − 16 = 125 − 20 + 4 = 109.
All yours
Work out 43
Work out 25 — the exact question type that broke at step 2
Work out 132
Work out 104
Work out 32 + 24
step 2–32 · Roots — powers run backwards▼
▶  Watch: E1.3 Powers and roots
Short hand-worked explanations for exactly this sub-topic. Opens on YouTube in a new tab. Watch one, then come straight back and try the questions below — watching without testing yourself feels like learning but is not.

A root is not a new operation to learn. It is the reverse of a power, in exactly the way that dividing is the reverse of multiplying.

132 = 169   ⇒   √169 = 13
43 = 64   ⇒   3√64 = 4

So every square you know gives you a square root for free. That table in section 1 is doing double duty. √169 is not something you compute; it is something you recognise, because you already know 13².

The mistake: treating √169 as “169 ÷ 2” and answering 84.5. Halving is not rooting. The test is always the same: does your answer, multiplied by itself, give you back the number you started with? 84.5 × 84.5 is nowhere near 169, so it is wrong.
Non-calculator: if a square root is not one you recognise, hunt for a square factor. √144 × √4 style splitting works because √(a×b) = √a × √b. For example √900 = √9 × √100 = 3 × 10 = 30.

Roots that are not exact: trap them

Paper 2 will ask you to estimate a root like √50. You cannot work it out exactly by hand, and you are not meant to. You trap it between the two squares either side of it.

49 64 50 7² 8² √49 = 7 √64 = 8 √50 sits here — just above 7, so about 7.1 50 is only 1 above 49 but 14 below 64, so the root is much nearer 7 than 8.
The method that works: name the square below and the square above. Then say which one the number is closer to, and lean the answer that way. “Between 7 and 8, much nearer 7” earns the mark; a random decimal does not.
Worked in full
Work out √196, and estimate √30 to one decimal place
1
√196: which number squared gives 196?
Ask the reverse question. Do not try to divide.
2
13² = 169, 14² = 196
Walk up the squares table. 14 lands exactly.
3
√196 = 14
Check: 14 × 14 = 196. It goes back where it came from.
4
√30: trap it. 5² = 25 and 6² = 36
30 lies between 25 and 36, so the root lies between 5 and 6.
5
30 is 5 above 25 and 6 below 36 — almost exactly halfway
So the root sits near the middle of 5 and 6.
6
√30 ≈ 5.5
Test it: 5.5² = 30.25. Slightly over, so 5.5 is right to one decimal place.
Last step is yours
Work out 3√125
1
A cube root asks: what number, times itself, times itself again, gives 125?
Three copies, because the index on the root is 3.
2
43 = 64 — too small
Walk up the cubes: 1, 8, 27, 64, 125.
3
3√125 = 5
Because 5 × 5 × 5 = 125.
Last two are yours
Work out √225 − 3√27
1
√225: 14² = 196, 15² = 225
Walking up the squares table again.
2
√225 = 15
Check: 15 × 15 = 225.
3
3√27 = 3
Because 3 × 3 × 3 = 27.
4
15 − 3 = 12
Roots first, subtraction last.
All yours
Work out √169
Work out 3√64
√40 lies between which two whole numbers? Give the smaller one.
Work out √400 by splitting it
Work out √121 + 3√8
step 3–43 · The index laws — derived, not memorised▼
▶  Watch: E1.7 Indices I
Short hand-worked explanations for exactly this sub-topic. Opens on YouTube in a new tab. Watch one, then come straight back and try the questions below — watching without testing yourself feels like learning but is not.

There are three laws. You are not going to memorise them, because memorised rules are the thing that collapsed at step 2. You are going to see why each one has to be true, from what a power means. Once you have seen it, you can rebuild any of them in ten seconds in an exam.

Law 1: multiplying — add the indices

a3 × a4 = (a×a×a) × (a×a×a×a) = a×a×a×a×a×a×a = a7

Three as in one bracket and four in the other. Push them together and you have seven as. That is all “add the indices” means: you are counting how many copies there are in total.

am × an = am+n

Law 2: dividing — subtract the indices

a5 ÷ a2 = (a×a×a×a×a) ÷ (a×a)

Two of the as on the top cancel against the two on the bottom. Five copies, take away two, leaves three.

am ÷ an = am−n   so   a5 ÷ a2 = a3

Law 3: a power of a power — multiply the indices

(a3)2 = a3 × a3 = a6

Two lots of three as is six as. Not three plus two.

(am)n = amn
A power is a count of copies a⁵ means five copies of a multiplied together put the copies together aⁿ × aⁿ = a add the indices cancel copies off aⁿ ÷ aⁿ = a subtract the indices groups of groups (aⁿ)ⁿ = a multiply the indices The bases never change. 2³ × 2⁴ = 2⁷, never 4⁷.
The mistake: adding the bases as well as the indices. 23 × 24 is 27, not 47. Check it with real numbers: 8 × 16 = 128, and 27 = 128. Meanwhile 47 is 16384. The laws move the indices around; the base stays put.
The second mistake: using the laws when the bases are different. 23 × 32 cannot be simplified into a single power — the copies are not the same thing, so you cannot count them together. Just work it out: 8 × 9 = 72.
Worked in full
Simplify 56 × 53 ÷ 57, leaving your answer as a single power of 5
1
All three have base 5
Check this first, every time. Same base means the laws apply.
2
56 × 53 = 56+3 = 59
Six copies and three copies makes nine copies.
3
59 ÷ 57 = 59−7
Seven of the nine copies cancel against the bottom.
4
= 52
Two copies left. The base is still 5, exactly as it started.
5
Check: 52 = 25
If the question wants a number rather than a power, say 25. Read what it asks for.
Last step is yours
Simplify (32)4 ÷ 35 as a single power of 3
1
(32)4 = 32×4 = 38
Four groups, each holding two copies. Multiply, do not add.
2
38 ÷ 35 = 38−5
Dividing, so subtract.
3
= 3³
8 − 5 = 3, so three copies remain. As a number that is 27.
Last two are yours
Simplify (23 × 24)2 ÷ 29 as a single power of 2
1
Inside the bracket first: 23 × 24 = 27
Add the indices. The base stays 2 — it does not become 4.
2
(27)2 = 214
Power of a power, so 7 × 2 = 14.
3
214 ÷ 29 = 2⁵
14 − 9 = 5. As a number, 25 = 32.
All yours
Simplify a4 × a6. Write it as a with an index, like a^10.
Simplify a9 ÷ a4
Simplify (a5)3
Work out 23 × 22 as an ordinary number
Work out 22 × 32 as an ordinary number
step 4–54 · Zero and negative indices — and the sign traps▼
▶  Watch: E1.7 Indices I
Short hand-worked explanations for exactly this sub-topic. Opens on YouTube in a new tab. Watch one, then come straight back and try the questions below — watching without testing yourself feels like learning but is not.

This is where the check flagged sign errors recurring across topics, and this is where they cost the most marks. Slow down here.

Why a0 = 1 — derived, not asserted

“Zero copies of a multiplied together” is a sentence that means nothing, so the picture does not help here. The division law does.

a5 ÷ a5 = a5−5 = a0
but a5 ÷ a5 is a number divided by itself, which is 1

Both lines describe the same thing, so a0 and 1 must be the same thing. That is the whole argument. 70 = 1. 10000 = 1. (−4)0 = 1.

The mistake: answering 0. The index being zero does not make the answer zero — it makes the answer 1. Rebuild the derivation above if you ever doubt it in an exam; it takes one line.

Why a negative index means “one over”

Same trick, keep dividing.

a3 ÷ a5 = a3−5 = a−2
but writing it out: (a×a×a) ÷ (a×a×a×a×a) = 1 ÷ (a×a) = 1⁄a2
so a−n = 1⁄an

A negative index is an instruction to flip, never an instruction to make the answer negative.

Walk down the powers of 2. Each step down divides by 2. Nothing ever turns negative. 2³ = 8 2² = 4 2¹ = 2 2⁰ = 1 2⁻¹ = 1/2 2⁻² = 1/4 ÷ 2 ÷ 2 ÷ 2 ÷ 2 ÷ 2 The three answers people give for 2⁻³ 1/8 ✓ correct −8 ✗ sign copied from the index −6 ✗ multiplied 2 by −3 The values stay positive all the way down.
The sign trap: 2−3 = 1⁄8. It is not −8 and not −6. The minus sign lives on the index, not on the value. A positive base raised to any power — positive, zero or negative — gives a positive answer, always.
The other sign trap: brackets. (−3)2 = −3 × −3 = 9, but −32 means −(32) = −9. The bracket decides whether the minus is inside the squaring or outside it. Read the brackets before you do anything.
The rule for negative bases: an even index makes the answer positive, because the minuses pair up. An odd index leaves one minus unpaired, so the answer is negative. (−2)4 = 16, but (−2)3 = −8.
Non-calculator: for a negative index, do the flip last. Work out the positive power as an ordinary number first, then put 1 over it. Trying to handle the fraction and the power at the same time is where the errors creep in.
Worked in full
Work out 3−2 + 50
1
3−2 — negative index, so flip
3−2 = 1⁄32. The minus does not touch the value.
2
32 = 9, so 3−2 = 1⁄9
Power first, flip second. Positive answer.
3
50 = 1
Any non-zero number to the power zero is 1, not 0.
4
1⁄9 + 1 = 11⁄9 = 10⁄9
Either form is acceptable. As a top-heavy fraction it is 10/9.
5
Sense check: is the answer a bit more than 1?
Yes — a small positive fraction added to 1. Nothing negative anywhere.
Last step is yours
Work out 4−2
1
Negative index means reciprocal: 4−2 = 1⁄42
Flip. Do not attach the minus to the answer.
2
42 = 16
From the squares table.
3
4−2 = 1/16
Positive, and smaller than 1 — which is what a negative index always does to a base bigger than 1.
Last two are yours
Work out (−2)3 + 2−2
1
(−2)3: the bracket means the minus is included in the base
So it is −2 × −2 × −2.
2
−2 × −2 = 4, then 4 × −2 = −8
Odd index, so one minus is left unpaired and the answer is negative.
3
2−2 = 1⁄22 = 1⁄4
Different kind of minus. This one flips, it does not negate.
4
−8 + 1⁄4 = −7¾
Adding a quarter to −8 moves it towards zero, giving −7.75.
All yours
Work out 2−3. Give a fraction or a decimal.
Work out 120
Work out (−5)2
Work out −52 — note there is no bracket
Work out 5−2
Work out (−3)3
step 5–65 · Fractional indices▼
▶  Watch: E1.7 Indices I
Short hand-worked explanations for exactly this sub-topic. Opens on YouTube in a new tab. Watch one, then come straight back and try the questions below — watching without testing yourself feels like learning but is not.

An index of 1⁄2 looks strange until you push the multiplying law at it.

a½ × a½ = a½+½ = a1 = a

So a½ is the thing that gives you a when multiplied by itself. That is the definition of a square root. Therefore:

a½ = √a     a⅓ = 3√a     a1/n = n√a

The bottom of the fraction is the root. Now the top. Using the power-of-a-power law:

am/n = (a1/n)m = (n√a)m
Read a fractional index as two instructions: the bottom tells you which root to take, the top tells you what power to raise it to. Bottom = root, top = power.

Always root first

You are allowed to do them in either order — (n√a)m and n√(am) give the same answer. But by hand the difference is enormous.

163/4 — root first163/4 — power first
4th root of 16 = 2
23 = 8
Two easy steps
163 = 4096
4th root of 4096 = 8
You have to cube 16 and then root a four-digit number
Non-calculator: root first, every time. Rooting first keeps the numbers small enough to handle in your head; powering first blows them up into numbers you then have to un-blow. This is a Paper 2 survival habit, not a preference.
The mistake: reading 163/4 as 16 × 3⁄4 = 12. A fractional index is still an index. It is not a multiplier — same error as reading 25 as 10, wearing a different hat.
The sign version: 25−½ is 1⁄5. Two instructions, not one: the − flips it, the ½ roots it. It is not −5, and it is not −½.
Worked in full
Work out 82/3
1
Bottom is 3, so take the cube root. Top is 2, so then square.
Split the fraction into its two jobs before touching a number.
2
3√8 = 2
Because 2 × 2 × 2 = 8. Root first, so the numbers stay small.
3
22 = 4
Now apply the top of the fraction.
4
82/3 = 4
Check the size: the index is between 0 and 1, so the answer should be less than 8. It is.
Last step is yours
Work out 163/4
1
Bottom 4 ⇒ fourth root. Top 3 ⇒ then cube.
Bottom = root, top = power.
2
Fourth root of 16: 2 × 2 × 2 × 2 = 16, so it is 2
From the powers of 2 list. Rooting first keeps this trivial.
3
23 = 8
So 163/4 = 8. Doing it the other way round would have meant rooting 4096.
Last two are yours
Work out 27−2/3
1
Three instructions: minus ⇒ flip, bottom 3 ⇒ cube root, top 2 ⇒ square
Deal with them one at a time. Do the flip last so you work with whole numbers.
2
3√27 = 3
Because 3 × 3 × 3 = 27.
3
32 = 9
So 272/3 = 9. Now handle the minus.
4
27−2/3 = 1/9
Flip 9. The answer is positive and less than 1 — never −9.
All yours
Work out 491/2
Work out 641/3
Work out 253/2
Work out 25−1/2
Work out 813/4
Work out 8−1/3
step 5–66 · Standard form▼
▶  Watch: E1.8 Standard form
Short hand-worked explanations for exactly this sub-topic. Opens on YouTube in a new tab. Watch one, then come straight back and try the questions below — watching without testing yourself feels like learning but is not.

Standard form is a way of writing any number as

A × 10n   where   1 ≤ A < 10   and   n is a whole number

The condition on A is the whole game. A must have exactly one non-zero digit before the decimal point. 0.9 × 104 and 12 × 103 are both correct values and both wrong form — and in an exam the form is what is being marked.

10⁻³ 10⁻² 10⁻¹ 10⁰ 10¹ 10² 10³ 10⁴ 10⁵ 0.001 0.01 0.1 1 10 100 1000 10000 100000 n is just a label for which column the number starts in. 4200 = 4.2 × 10³ 0.036 = 3.6 × 10⁻² Positive n means big. Negative n means small — small, not negative. 3.6 × 10⁻² is a positive number.

Writing a number in standard form

Put the decimal point after the first non-zero digit, then count how many places it moved.

4200 → 4.2   (point moved 3 places left) → 4.2 × 103
0.00058 → 5.8   (point moved 4 places right) → 5.8 × 10−4
Which way is the sign? Ask whether the original number is bigger or smaller than 1. Bigger than 1 ⇒ n is positive. Smaller than 1 ⇒ n is negative. Never guess from the direction of the arrow — check against 1. This is the standard-form version of your recurring sign error, and this is the check that stops it.

Multiplying and dividing by hand

This is the part that actually gets examined, and it is easier than it looks because the two halves separate completely.

(a × 10m) × (b × 10n) = (a × b) × 10m+n
(a × 10m) ÷ (b × 10n) = (a ÷ b) × 10m−n

Numbers with numbers, powers of ten with powers of ten. Then check the front number, because it often lands outside 1 to 10 and has to be fixed.

The classic correction: 3 × 4 = 12, so you get 12 × 103. That is not standard form — 12 is not between 1 and 10. Rewrite 12 as 1.2 × 101, and the extra ten joins the others: 1.2 × 104. The number got ten times smaller at the front, so the power went up by one to compensate. Big front number goes down, power goes up.
The other direction: if you land on 0.5 × 106, the front number is too small. 0.5 becomes 5, ten times bigger, so the power drops by one: 5 × 105.
Worked in full
Work out (3 × 105) × (4 × 10−2), giving your answer in standard form
1
Separate: (3 × 4) × (105 × 10−2)
Multiplication can be reordered, so gather the plain numbers together and the tens together.
2
3 × 4 = 12
The front number.
3
105 × 10−2 = 105+(−2) = 103
Same base, so add the indices. 5 + (−2) = 3 — adding a negative means going down.
4
So far: 12 × 103
Correct value, wrong form. 12 is not between 1 and 10.
5
12 = 1.2 × 101, so 12 × 103 = 1.2 × 101 × 103
Rewrite the front number in standard form and let the spare ten join the rest.
6
= 1.2 × 104
Sense check: 300000 × 0.04 = 12000, and 1.2 × 104 is 12000.
Last step is yours
Write 0.00072 in standard form
1
First non-zero digit is 7, so A = 7.2
Exactly one non-zero digit before the point.
2
0.00072 is smaller than 1, so the power is negative
Check against 1 rather than guessing the sign.
3
The point moves 4 places: 0.00072 → 7.2
Count them: 0.0007.2 is four hops right.
4
0.00072 = 7.2 × 10⁻⁴
Check by undoing it: 7.2 ÷ 10000 = 0.00072.
Last two are yours
Work out (8 × 103) ÷ (2 × 10−4) in standard form
1
Separate: (8 ÷ 2) × (103 ÷ 10−4)
Numbers with numbers, tens with tens.
2
8 ÷ 2 = 4
Already between 1 and 10, so no correction will be needed.
3
103 ÷ 10−4 = 103−(−4) = 10⁷
Subtracting a negative adds: 3 − (−4) = 3 + 4 = 7. This exact step is where sign errors live.
4
Answer: 4 × 10⁷
Sense check: dividing by a very small number should make it much bigger. It did.
All yours
Write 63000 in standard form. Type it like 6.3 x 10^4.
Write 0.0045 in standard form
Write 4.7 × 103 as an ordinary number
Correct 45 × 106 into proper standard form
Work out (2 × 104) × (6 × 103) in standard form
Work out (9 × 105) ÷ (3 × 102) in standard form

Adding and subtracting in standard form

Multiplying and dividing let the two halves separate. Adding and subtracting do not: you may only add the front numbers when the powers of 10 are the same. Rewrite one number so the powers match, add, then put the answer back into standard form.

Worked in full
Work out 3.2 × 10⁴ + 5 × 10³, giving the answer in standard form.
1
3.2 × 10⁴ = 32 × 10³
Make the powers match: move one place, so 3.2 becomes 32 and 10⁴ becomes 10³.
2
32 × 10³ + 5 × 10³ = 37 × 10³
Now the fronts can be added: 32 + 5 = 37.
3
= 3.7 × 10⁴
37 is not between 1 and 10, so tidy it: 3.7 × 10⁴. Check: 32 000 + 5000 = 37 000 ✓
The mistake: 3.2 × 10⁴ + 5 × 10³ = 8.2 × 10⁷ (adding the fronts and the powers) or 8.2 × 10⁴ (adding the fronts as if the powers were equal). Write both numbers out in full once, as a check, until matching the powers is automatic.
All yours
Work out (4 × 10⁵) + (3 × 10⁴). Give the answer in standard form. Type it like 6.3 x 10^4.
Work out (2.5 × 10⁻³) − (4 × 10⁻⁴). Give the answer in standard form. Type it like 6.3 x 10^-4.
Work out (9 × 10⁶) + (8 × 10⁵). Give the answer in standard form. Type it like 6.3 x 10^4.
step 67 · Estimation — rounding to 1 significant figure▼
▶  Watch: E1.9 Estimation
Short hand-worked explanations for exactly this sub-topic. Opens on YouTube in a new tab. Watch one, then come straight back and try the questions below — watching without testing yourself feels like learning but is not.

E1.9. This is pure Paper 2 material, and it is the cheapest topic on the syllabus: no method to learn, just one rule applied honestly. It also rescues you elsewhere — when you are unsure whether an answer is roughly right, an estimate settles it in fifteen seconds.

One significant figure

The first significant figure is the first non-zero digit. Round to it and turn everything after it into zeros (or nothing, after a decimal point).

NumberTo 1 s.f.Why
4.875First digit is 4; the next digit is 8, so round up
19.620First digit is 1; the next digit is 9, so round the 1 up to 2
0.04120.04First non-zero digit is 4; the leading zeros are placeholders, not significant
612600First digit 6; next digit 1, so it stays
0.04890.05First non-zero digit 4; next digit 8, so round up
The mistake: rounding 0.0412 to 0.0 because “one figure” is read as “one decimal place”. Significant figures start counting at the first non-zero digit, not at the decimal point. 0.0412 to 1 s.f. is 0.04.
Non-calculator: dividing by a decimal is where estimates fall over. Turn it into a multiplication instead: ÷ 0.5 is the same as × 2, ÷ 0.2 is the same as × 5, ÷ 0.1 is the same as × 10. Or scale the top and bottom by the same power of ten until the bottom is whole: 24 ÷ 0.4 = 240 ÷ 4 = 60.
Answering the exam question: if it says “estimate”, you must show the rounded values before you work anything out. The mark is for the line “≈ (5 × 20) ÷ 0.5”, not only for the final number.
Worked in full
Estimate the value of (4.87 × 19.6) ÷ 0.51
1
Round every number to 1 significant figure first
All of them, before any arithmetic. Write this line down — it carries a mark.
2
4.87 ≈ 5, 19.6 ≈ 20, 0.51 ≈ 0.5
19.6 rounds to 20, not 19 — the 1 is carried up by the 9.
3
(5 × 20) ÷ 0.5 = 100 ÷ 0.5
Top first.
4
100 ÷ 0.5 = 100 × 2 = 200
Dividing by a half doubles. Never leave a division by a decimal standing.
5
Estimate: 200
The true value is 187.2, so the estimate is doing its job.
Last step is yours
Estimate (612 × 0.0398) ÷ 1.97
1
612 ≈ 600, 0.0398 ≈ 0.04, 1.97 ≈ 2
Each to 1 significant figure.
2
600 × 0.04 = 24
600 × 4 = 2400, and 0.04 is 4 divided by 100, so 2400 ÷ 100 = 24.
3
24 ÷ 2 = 12
Estimate 12. The true value is about 12.4.
Last two are yours
Estimate √(38.2 × 2.13)
1
38.2 ≈ 40 and 2.13 ≈ 2
Round inside the root first.
2
40 × 2 = 80
Now the root has a manageable number under it.
3
√80 is between √64 = 8 and √81 = 9, so √80 ≈ 9
80 is almost exactly 81, so the answer is just under 9. The true value is about 9.02.
All yours
Write 0.0387 correct to 1 significant figure
Estimate 39 × 61
Estimate 41.3 ÷ 0.196
Estimate √(9.4 × 105)

Decimal places, significant figures, and the nearest 10, 100 or 1000

instructioncount fromexample
decimal places (d.p.)the decimal point7.4651 to 2 d.p. is 7.47; 3.8962 to 2 d.p. is 3.90 (write the 0)
significant figures (s.f.)the first non-zero digit0.04056 to 2 s.f. is 0.041; 48 520 to 3 s.f. is 48 500
the nearest 10, 100, 1000that place value5764 to the nearest 1000 is 6000

In every case look at the next digit: 5 or more rounds up. Keep the size of the number: 48 520 to 3 s.f. is 48 500, not 485. And when a rounded answer ends in 0 after the point, write the 0: 3.90 to 2 d.p., not 3.9.

Sensible accuracy in context

Decide what the answer means before you round. People or things that must all be carried: round up. Complete items you can make: round down. Money: 2 decimal places. Lengths: 3 significant figures unless told otherwise.

Worked in full
380 students go on a trip. A bus holds 52 students. How many buses are needed?
1
380 ÷ 52 = 7.3…
7 × 52 = 364, which is 16 short of 380.
2
round UP: 8 buses
7 buses would leave 16 students behind.
All yours
Write 5764 correct to the nearest thousand.
Write 0.04056 correct to 2 significant figures.
Write 48 520 correct to 3 significant figures.
Write 7.4651 correct to 2 decimal places.
A baker has 250 eggs. Each box holds 12 eggs. How many FULL boxes can be filled?
mixed8 · Mixed set — no labels, no order▼

Sixteen questions drawn from everything above, shuffled and unlabelled. Ordinary revision does one topic at a time, which quietly does the hardest part for you — deciding which idea applies. Here nobody tells you. Before you calculate anything, say to yourself which of these it is: a power, a root, an index law, a zero or negative index, a fractional index, standard form, or an estimate.

nothing answered yet
1. Work out 4−2
2. Work out 811/2
3. Write 0.000306 in standard form. Type it like 5.2 x 10^-6.
4. Work out 30 + 32
5. Work out 3√1000
6. Simplify a7 ÷ a7 and give the answer as a number
7. Estimate (5.8 × 41) ÷ 0.21
8. Work out (−4)2
9. Work out 272/3
10. Work out (5 × 10−3) × (8 × 106) in standard form
11. Work out √196 − 33
12. Simplify (25)2 ÷ 28 and give the answer as a number
13. Work out 100−1/2
14. Write 8 900 000 in standard form
15. Work out −24 — there is no bracket
16. Work out 2−1 + 2−2

What to take away

Five things, and they cover most of what an examiner can ask in this strand:

  1. A power counts copies. Write the multiplication out whenever you are unsure — it costs seconds and removes the step-2 error entirely.
  2. Learn the squares to 15 and the cubes of 1, 2, 3, 4, 5 and 10 by heart. Every root question then becomes a recognition question.
  3. The three laws come out of that picture: multiply add, divide subtract, power of a power multiply. The base never changes.
  4. A minus on an index means flip, never negate. A minus on a base only survives the squaring if it is inside a bracket.
  5. In standard form, split the numbers from the tens, then check the front number is between 1 and 10 before you write the answer down.

Come back to section 8 in a week without reading anything above it. If the score bar sits high on a cold run, this brick is set and the next repair guide can start.