Significant Figures: How Honest Is Your Number Really?
JEE/NEET Physics · Units & Measurements series · Part 3 of 6 · All parts →
- A significant figure is a digit you actually measured — writing more claims false precision
- All non-zero digits count; zeros count only if measured or between measured digits
- Multiply/divide → keep the FEWEST sig-figs in the chain; add/subtract → keep the fewest DECIMALS
- Exact numbers (2π, 4 in ‘4 sides’, conversion factors) have infinite sig-figs
- Rounding: 5 or more rounds up; round only at the END of a calculation
A robot says your pizza arrives in 7.2941 minutes. Absurd — not because the maths is wrong, but because the claim is dishonest: nobody can know a delivery time to a tenth of a second. Significant figures are the honesty policy of measurement. Every digit you write is a promise about what your instrument actually saw — and significant figures are the grammar that keeps those promises honest. Part 3 of the Units & Measurements series.
- Why digits carry honesty
- Counting rules: which zeros count
- Scientific notation: the ambiguity killer
- The two arithmetic rules
- Exact numbers: the infinite exceptions
- Rounding done right
- Solved examples
- Common mistakes
- This physics in your daily life
- Frequently asked questions
- Practice set
- Recap & key takeaways
Why Digits Carry Honesty
Write ‘the rod is 12.3 cm’ and you claim three things: the 1 and 2 are certain, the 3 is your best estimate, and you’d be embarrassed to say more. Write 12.30 cm and you additionally claim the 0 was measured. Extra digits aren’t extra information — they’re false advertising. Sig-figs keep you from promising precision your instrument never delivered.
This idea has real consequences. In JEE and NEET numericals, examiners deliberately hand you numbers of different precision and check whether your final answer respects the weakest one. In real laboratories, the stakes are higher: a drug dosage or a satellite trajectory reported with fake precision can mislead an entire engineering or medical team. The rule exists because every measurement ends in one estimated digit — the reading you take between the finest markings on the scale — and everything downstream of that estimate inherits its doubt.
Think of it this way: a measurement is a sentence, and its significant figures are the words that actually carry meaning. Padding it with extra digits is like adding adjectives to a sentence you only half-understood — the sentence gets longer, but not truer.
Counting Rules: Which Zeros Count
Zeros are the whole game. Non-zero digits always count — there’s never any debate about a 3 or a 7. The trouble begins and ends with zero, because zero plays two roles: sometimes it’s a measured digit, and sometimes it’s merely a placeholder pushing the decimal point to the right place. The complete set of rules:
- Rule 1: All non-zero digits are significant.
- Rule 2: Zeros between non-zero digits (sandwiched zeros) are always significant.
- Rule 3: Leading zeros — zeros before the first non-zero digit — are never significant. They’re place-setting, not measurement.
- Rule 4: Trailing zeros after a decimal point are significant (someone measured them).
- Rule 5: Trailing zeros without a decimal point are ambiguous — use scientific notation to resolve.
| Number | Sig-figs | Why |
|---|---|---|
| 204 | 3 | sandwiched zeros always count |
| 0.0204 | 3 | leading zeros NEVER count (they’re place-setting) |
| 2.04 | 3 | same digits, same physics |
| 2.40 | 3 | trailing zero AFTER a decimal point was measured |
| 240 | 2 or 3 | ambiguous — write 2.4×10² or 2.40×10² to say which |
| 0.006 | 1 | all three zeros are leading placeholders |
| 1.00 | 3 | both trailing zeros sit after a decimal point |
Scientific Notation: The Ambiguity Killer
Rule 5 is the reason scientists love scientific notation. When you write 2400, nobody can tell whether you measured two, three, or four digits. But write it as 2.4 × 10³ and you’ve declared exactly two sig-figs; 2.40 × 10³ declares three; 2.400 × 10³ declares four. The power of ten (10³) is not a digit at all — it’s furniture, and it never counts toward the sig-fig total. In JEE problems, any coefficient written in scientific notation counts sig-figs from its first non-zero digit onward, and that’s the entire procedure.
The Two Arithmetic Rules
× and ÷: the answer keeps as many sig-figs as the poorest number in the chain. + and −: the answer keeps as many decimal places as the least precise term. The logic: your answer can never be more honest than your shakiest ingredient.
Why do the rules differ? Because multiplication damages relative precision (percentage-wise), while addition damages absolute precision (place-wise). When you multiply, a number with 2 sig-figs carries roughly a few-percent uncertainty that infects the product’s leading digits. When you add, error lives in the decimal columns: adding 12.11 and 0.3, the hundredths column of the answer is pure fiction because 0.3 never knew anything past its tenths. Match the rule to the operation and the physics stays consistent.
Exact Numbers: The Infinite Exceptions
Counting numbers (12 eggs), constants in formulas (the ½ in KE = ½mv²), and defined conversions (1 m = 100 cm) have unlimited sig-figs — they were never measured, so they carry no doubt. The number π in circumference = 2πr is likewise exact in principle; in practice you carry enough of its digits (say 3.1416) that it never limits your answer. The test is simple: was this number measured? If nobody measured it — if it’s a definition, a count, or a pure coefficient — it can never be the weak link.
| Letter | What it means (plain words) | Value / unit |
|---|---|---|
| s.f. | significant figures — digits that carry real information | count from first non-zero digit |
| decimal place | position after the decimal point | governs + and − only |
Rounding Done Right
Two disciplines make rounding reliable. First, round only once, at the very end. If you round 8.55 to 8.6 mid-calculation and then multiply by something else, you’ve injected an error the final answer never earned. Keep one or two guard digits through every intermediate step, then trim once. Second, follow the standard rule: digit 5 or greater rounds up, 4 or less rounds down. (If your syllabus covers it, the ’round-half-to-even’ convention exists for instrument readings that land exactly on 5 — but for JEE/NEET, the plain rule is what’s tested.)
A worked rounding drill: round 3.14159 to 3, 4, and 5 sig-figs. Three s.f. → 3.14. Four s.f. → 3.142 (the next digit is 5, round up). Five s.f. → 3.1416. Notice how each answer carries exactly as many digits as requested — no more, no fewer. That discipline is what graders look for.
Solved Examples
Skip leading zeros (0.00): start at the 6.
6, 0, 4, 0 — the middle 0 counts (sandwiched), the last 0 counts (after decimal, trailing).
4 sig-figs. ✔
Answer: 4
Raw: 8.55 m². But 2.5 has only 2 sig-figs — the chain’s weakest link.
Round to 8.6 m² (2 s.f.).
Why: the 5 in 2.5 was already an estimate — an estimate multiplied by precision is still an estimate. ✔
Answer: 8.6 m² (2 s.f.)
Raw: 16.66 g. Least decimals: 0.3 (one decimal place).
Round to 16.7 g (one decimal place).
Note the rule switched: addition cares about DECIMALS, not total digits — 0.3 is shakiest because its uncertainty is ±0.1. ✔
Answer: 16.7 g
Step 1 (multiplication): 4.20 × 1.5 = 6.3 raw → weakest is 1.5 (2 s.f.) → keep 6.3.
Step 2 (addition): 6.3 + 6.13 = 12.43 raw → least decimals is one place (6.3) → round to 12.4.
Strategy: apply each rule at each stage, but carry a guard digit between steps and round the final result once. ✔
Answer: 12.4
ρ = 27.4 ÷ 10.2 = 2.686… g/cm³ raw. Both inputs have 3 s.f., so the answer keeps 3 s.f.
ρ = 2.69 g/cm³. ✔
Answer: 2.69 g/cm³
- Counting leading zeros. 0.0034 has TWO sig-figs, not five — zeros before the first non-zero digit never count.
- Applying the × rule to + problems. Addition counts decimal places, multiplication counts sig-figs — mixing the two is the classic slip.
- Rounding mid-calculation. Carry extra digits through, round ONCE at the end; early rounding compounds error.
- Treating exact numbers as 1-sig-fig. The 2 in d = 2r and the 100 in 1 m = 100 cm are exact — infinite sig-figs, they never limit the answer.
- Counting the power of ten. In 6.02 × 10²³, the sig-figs live in 6.02 (three); the 10²³ is notation, not measurement.
- Trimming trailing zeros in answers. Changing 2.40 to 2.4 destroys information — that zero was measured; keep it.
This Physics in Your Daily Life
- Delivery apps say ’25–35 minutes’, not ‘27.4 min’ — honest uncertainty ranges, the sig-fig philosophy in commerce.
- Cricket win-probability ‘68%’ never shows 68.37% — broadcasters rounding to the precision a model can honestly support.
- Bank interest rounded to the paisa — accounting rules enforce decimal-place discipline identical to the addition rule.
- Medical lab reports (‘Hb 12.3 g/dL’, three sig-figs) — calibrated instruments promise exactly the digits they print, no more; doctors diagnose against those digit budgets.
- Every sports statistic you read — bowling speed 140.2 km/h but batting average 45.6 — silently follows a sig-fig convention.
- Fuel prices and odometers — a car showing 45,213 km is honest to the kilometre; no manufacturer would claim 45,213.7 km on a mechanical odometer, because the hardware can’t measure it.
A tailor with a metre stick measures your sleeve as ’61 cm’, estimating the last digit by eye. If he writes 61.000 cm, he’s claiming a laser lab. Digits beyond your instrument’s power aren’t detail — they’re fiction wearing a suit. Sig-figs are the tailor’s oath: I write only what I saw.
A ruler marked in mm gives ~3 honest sig-figs for a book (21.3 cm); a vernier caliper gives 4 (21.34 cm). Multiply book height × 2.5 (a measured factor): the ruler chain must stop at 3 digits, the vernier chain at 4. Poorer instrument wins — 2.5’s honesty caps everything below it.
Picture each number as a pole whose length is certainty: 4 certain digits = tall pole, 1 estimate = wobbly tip. Chain numbers with ropes (×, ÷): the rope can only be as taut as the shortest pole. Draw the chain — the sag marks where your answer must stop being written.
Frequently Asked Questions
Q. Do sig-figs matter in chemistry too, or only physics?
A. Everywhere measurement happens — physics, chemistry, biology labs, engineering. Molarity calculations, titration readings, and gas-law numericals all obey the identical two rules. JEE Chemistry numericals frequently test them.
Q. How many sig-figs should I keep in intermediate steps?
A. Keep one or two guard digits beyond the final requirement, then round once at the end. This is the standard practice in both exam solutions and professional computation.
Q. Is 0 a significant figure on its own?
A. The number 0 by itself is exact (a defined value), so the question doesn’t really arise. Within measurements, follow the zero rules: sandwiched and trailing-after-decimal count, leading never do.
Q. What about numbers like 3 × 10⁸ in c = 3 × 10⁸ m/s?
A. Textbook shorthand often treats it as one sig-fig; when precision matters, the measured value is 2.998 × 10⁸ m/s (4 s.f.). Context decides how honest the ‘3’ is being.
Practice set (answers hidden — try first)
(NEET-level) Sig-figs in 0.0500:
(JEE Main-level) 4.0 × 3.00 ÷ 6.000 → answer’s s.f.:
(NEET-level) 11.2 + 3.14 (proper sig-figs):
(Concept) In d = 2r, the 2 has how many sig-figs?
(JEE Main-level) Volume = 2.1 × 2.1 × 2.1 (m³):
(NEET-level) Express 8500 with 2 and 3 sig-figs, unambiguously:
(JEE Advanced-level) (12.0 × 1.5) + 8.204 = ?
- sandwiched and trailing-after-decimal zeros count; leading never
- × ÷ → fewest sig-figs; + − → fewest decimals
- exact numbers (counts, 2π, defined conversions) = infinite s.f.
- round once, at the end
- 240 is ambiguous — scientific notation settles it
- 🔁 leading zeros never count
- 🔁 × ÷ → fewest s.f.; + − → fewest decimals
- 🔁 exact numbers carry infinite s.f.
- 🧠 Chant: ‘zeros left, never; zeros between, forever; zeros right (with decimal), measured tight’.
- 🧠 Chain rule: ‘the weakest link writes the last digit’.
- 🧠 Round once: guard digits through the middle, trim only at the end.
- 🏠 Daily: ’25–35 min’ delivery windows = honest uncertainty.
- 🏠 Daily: lab reports print exactly the digits machines can honor.
- ✅ Key takeaway: significant figures are not a formality — they are the difference between a measurement and a boast. Every extra digit you write is a claim, and every claim must be earned by the instrument.
- ✅ Key takeaway: in exams, match the answer’s precision to the input’s precision and you harvest the full method marks — most lost marks here come from one extra digit, not from wrong arithmetic.
Quick revision
- A significant figure is a digit you actually measured — writing more claims false precision
- All non-zero digits count; zeros count only if measured or between measured digits
- Multiply/divide → keep the FEWEST sig-figs in the chain; add/subtract → keep the fewest DECIMALS
- Exact numbers (2π, 4 in ‘4 sides’, conversion factors) have infinite sig-figs
- Rounding: 5 or more rounds up; round only at the END of a calculation
- Counting rules: which zeros count
Have a doubt on this topic?
