In short
Every measured value carries a built-in limit of precision — significant figures tell you where that limit is. Trailing zeros after a decimal point count, leading zeros never do, and when you combine measurements, the answer cannot be more precise than the least precise input. Keeping track of significant figures prevents you from writing down digits that your instrument never actually gave you.
You are in the physics lab. The teacher hands you a metre ruler — the kind with millimetre markings — and asks you to measure the length of a glass rod. You line it up carefully. One end sits at the 0 mark. The other end falls somewhere between the 14.7 cm and 14.8 cm lines. You squint. It looks closer to 14.7 than to 14.8. You write down 14.7 cm.
That number has three digits you are confident about: the 1, the 4, and the 7. You know the rod is not 15 cm. You know it is not 14.6 cm. But you do not know whether it is 14.71 cm or 14.73 cm — your ruler cannot tell you that. Those three digits you can trust are the significant figures of your measurement.
Now your lab partner uses a Vernier calliper on the same rod. The calliper reads 14.72 cm — four significant figures. A micrometer screw gauge might give 14.723 cm — five significant figures. The rod did not change. What changed is the precision of the instrument, and significant figures are how you record that precision inside the number itself.
Precision is not accuracy — and the difference matters
Before counting significant figures, you need to separate two ideas that are often confused.
Precision is about how finely an instrument can distinguish between nearby values. Your metre ruler is precise to the nearest millimetre. The Vernier calliper is precise to the nearest 0.01 cm. A micrometer is precise to the nearest 0.001 cm.
Accuracy is about how close the measurement is to the true value. A ruler with a chipped end might consistently read 0.3 cm too high — its measurements are precise (repeatable to the nearest mm) but inaccurate (shifted away from the truth).
Significant figures track precision, not accuracy. They tell you how many digits your instrument can reliably give you. If your weighing scale shows 52.4 g for a packet of dal, the three significant figures mean the scale can distinguish between 52.3 g and 52.5 g — but they say nothing about whether the scale was calibrated correctly.
Rules for counting significant figures
These rules look like a list to memorise, but they are not arbitrary. Each one follows directly from a single principle: a significant figure is any digit that carries real information about the measurement.
Rule 1: All non-zero digits are significant
The number 2574 has four significant figures. Every digit tells you something real — the measurement is not 2575 or 2573.
Rule 2: Zeros between non-zero digits are significant
The number 3007 has four significant figures. Those two zeros are not filler — they tell you the measurement is in the three-thousands, not 3017 or 3107. Similarly, 40.05 has four significant figures.
Why: a zero sandwiched between non-zero digits was placed there deliberately by the instrument. It carries the same information as any other digit in that position.
Rule 3: Leading zeros are never significant
The number 0.0042 has only two significant figures (the 4 and the 2). The leading zeros are just placeholders that tell you where the decimal point is — they shift the number to the right place on the number line but carry no information about the precision of the measurement.
Why: you could rewrite 0.0042 as 4.2 \times 10^{-3} and the leading zeros vanish entirely. They were never part of the measurement — they were an artefact of the decimal system.
Rule 4: Trailing zeros after the decimal point are significant
The number 2.50 has three significant figures, not two. That trailing zero is the physicist saying: "I measured this to the hundredths place, and the hundredths digit is 0." Writing 2.5 would mean you only measured to the tenths place — a less precise measurement. The zero carries information.
Similarly, 0.03700 has four significant figures (3, 7, 0, 0). The leading zeros do not count, but the trailing zeros do.
Why: if the trailing zero did not matter, you would not have written it. Its presence is a deliberate statement about the precision of your instrument.
Rule 5: Trailing zeros in a whole number are ambiguous
This is the one rule that causes genuine confusion, and it deserves a careful explanation.
Suppose someone writes "1500 m" for the length of a road. Does that mean the length is known to the nearest metre (four significant figures: 1, 5, 0, 0)? Or to the nearest hundred metres (two significant figures: 1 and 5, with the zeros just marking the scale)?
Without additional context, you cannot tell. This is a real ambiguity in the decimal system, and there are two ways to resolve it:
-
Use scientific notation. Write 1.500 \times 10^3 m for four significant figures, or 1.5 \times 10^3 m for two. The notation removes all ambiguity.
-
Use a decimal point. Write 1500. m (with an explicit decimal point) to indicate that all four digits are significant. This convention is less common in Indian textbooks but is standard in many lab manuals.
In practice, JEE and CBSE problems resolve the ambiguity by either stating the number of significant figures explicitly or using scientific notation. If a problem says "a mass of 2400 g," treat it as having the number of significant figures that the context requires — and if you are unsure, ask.
Summary table
| Number | Significant figures | Why |
|---|---|---|
| 4307 | 4 | All non-zero + sandwiched zero |
| 0.0056 | 2 | Leading zeros do not count |
| 2.50 | 3 | Trailing zero after decimal counts |
| 8.002 | 4 | Sandwiched zeros count |
| 1500 | Ambiguous | Could be 2, 3, or 4 — use scientific notation |
| 1.500 \times 10^3 | 4 | Notation resolves the ambiguity |
| 0.07080 | 4 | Leading zeros out, trailing zero in, sandwiched zero in |
| 100.0 | 4 | Trailing zero after decimal counts |
Rounding rules
When you round a number to fewer significant figures, you need a rule for what to do with the digit you are dropping. The standard rules are:
If the dropped digit is less than 5, leave the preceding digit unchanged.
- 3.842 rounded to 3 sig figs → 3.84
If the dropped digit is greater than 5, increase the preceding digit by 1.
- 3.847 rounded to 3 sig figs → 3.85
If the dropped digit is exactly 5 (with nothing after it, or only zeros), round to the nearest even digit. This is called the round-half-to-even rule (or banker's rounding).
- 3.845 → 3.84 (the 4 is already even, so keep it)
- 3.835 → 3.84 (the 3 is odd, so round up to 4)
Why round-half-to-even? If you always round 5 upward, you introduce a systematic bias — your rounded numbers will, on average, be slightly too high. Rounding to the nearest even digit removes this bias because about half the time you round up and half the time you round down. In a long chain of calculations, the errors cancel out instead of accumulating in one direction.
If the dropped digit is 5 followed by non-zero digits, always round up.
- 3.8451 rounded to 3 sig figs → 3.85
Why: the number 3.8451 is genuinely larger than 3.845, so it belongs in the "greater than 5" category. The round-half-to-even rule only applies to the exact halfway point.
How significant figures propagate through calculations
Measurements do not live in isolation. You measure the length and width of a table, then multiply to get the area. You measure two times and subtract to get a duration. Each operation has its own rule for how many significant figures the result can claim.
Multiplication and division: count significant figures
When you multiply or divide measured quantities, the result has as many significant figures as the input with the fewest significant figures.
Suppose you measure a rectangular carrom board: length = 73.2 cm (3 sig figs), width = 73.0 cm (3 sig figs). The area is:
But the inputs only have 3 significant figures each, so the result is rounded to 3 significant figures: 5340 cm² (or 5.34 \times 10^3 cm² to avoid the trailing-zero ambiguity).
Why: multiplication combines relative uncertainties. If each side is uncertain by roughly 1 part in 730 (about 0.14%), the area is uncertain by about 0.28% — which means the fourth digit of 5343.6 is unreliable. Keeping only 3 significant figures is the honest thing to do.
Addition and subtraction: count decimal places
When you add or subtract, the result should have as many decimal places as the input with the fewest decimal places.
Suppose you weigh three ingredients for chai:
- Tea leaves: 5.2 g (1 decimal place)
- Sugar: 12.35 g (2 decimal places)
- Milk powder: 3.0 g (1 decimal place)
The sum is:
The inputs with the fewest decimal places (5.2 and 3.0) have only 1 decimal place. So the result is rounded to 1 decimal place: 20.6 g.
Why: when you add, it is the absolute uncertainty that matters, not the relative uncertainty. The tea leaves are known only to the nearest 0.1 g, so the total cannot possibly be known to the nearest 0.01 g, no matter how precisely you weighed the sugar. The weakest link limits the chain.
Notice the difference: multiplication/division cares about significant figures, while addition/subtraction cares about decimal places. The underlying logic is the same — do not claim more precision than your data supports — but the way precision propagates is different for the two types of operations.
A practical rule for multi-step calculations
When a calculation involves several steps, keep at least one extra significant figure in all intermediate results. Round only at the very end. If you round at every intermediate step, the rounding errors accumulate and the final answer drifts — sometimes enough to change the last significant figure.
Worked examples
Example 1: Measuring a cricket pitch
A cricket pitch is officially 22 yards long. In metres, this is 20.1168 m (exact conversion). A student uses a measuring tape marked in centimetres to measure the pitch and gets 20.12 m. How many significant figures does the student's measurement have? If the width of the pitch is measured as 3.05 m, what is the area of the pitch in the correct number of significant figures?
Step 1. Count the significant figures in 20.12.
All four digits are significant: 2, 0, 1, 2. The zero is sandwiched between non-zero digits (Rule 2), so it counts. The measurement has 4 significant figures.
Why: the student's tape can distinguish between 20.11 m and 20.13 m. All four digits carry real measurement information.
Step 2. Count significant figures in the width: 3.05 m.
Three significant figures: 3, 0, 5. The zero is sandwiched (Rule 2).
Step 3. Compute the area.
The inputs have 4 and 3 significant figures. For multiplication, the result takes the fewer — that is 3. So the area rounds to 61.4 m².
Why: the width is the least precise input (3 sig figs). Keeping more digits in the area would imply you know the width more precisely than you actually do.
Result: The length has 4 significant figures. The area of the pitch is 61.4 m² (3 significant figures).
What this shows: Even when one measurement is very precise (4 sig figs), the final answer is limited by the least precise input (3 sig figs). The weakest measurement controls the precision of the result.
Example 2: Weighing chemicals in a lab
In a chemistry-physics practical, you need to find the density of a metal block. You weigh it on an electronic balance that reads 215.3 g. You measure its volume by water displacement: the initial water level in the measuring cylinder is 40.0 mL and the final level (after immersing the block) is 68.5 mL. Calculate the density in the correct number of significant figures.
Step 1. Find the volume of the block.
This is a subtraction, so count decimal places. Both readings have 1 decimal place, so the result has 1 decimal place: 28.5 cm³ — that is 3 significant figures.
Why: subtracting two volumes, both known to the nearest 0.1 mL, gives a difference known to the nearest 0.1 mL. The 3 significant figures in 28.5 reflect this precision.
Step 2. Compute the density.
The mass has 4 significant figures, the volume has 3. For division, the result takes the fewer: 3 significant figures.
Why: the volume is the least precise input. Reporting the density as 7.5543 would imply you know the volume to 4 or 5 significant figures — you do not.
Step 3. Identify the metal.
A density of 7.55 g/cm³ is close to iron (7.87 g/cm³) or tin (7.31 g/cm³). The uncertainty in your measurement (about ±0.1 g/cm³ from the volume measurement) means you cannot distinguish between them from density alone — you would need a more precise volume measurement.
Result: The density is 7.55 g/cm³ (3 significant figures).
What this shows: In a division, the least precise input controls the result. The electronic balance gave 4 significant figures, but the measuring cylinder limited the answer to 3. If you wanted higher precision, you would need a better way to measure volume — perhaps a Vernier-depth gauge or an overflow can.
Common confusions
-
"5.0 and 5 are the same number." Mathematically, yes. Physically, no. The number 5.0 cm means you measured to the nearest 0.1 cm and got exactly 5. The number 5 cm means you measured to the nearest 1 cm. The trailing zero in 5.0 is not decoration — it is a claim about your instrument's precision.
-
"More digits means a better answer." Your calculator will happily give you 12 digits for 22/7 = 3.14285714286... But if the inputs to your calculation had only 3 significant figures, those extra digits are fiction. They look precise, but they come from the calculator, not from your measurement. Never copy all the digits from your calculator.
-
"Leading zeros count as significant." They do not. The number 0.00340 has three significant figures (3, 4, 0), not six. The leading zeros just position the decimal point. The quick test: write the number in scientific notation (3.40 \times 10^{-3}). The leading zeros vanish, proving they were never significant.
-
"Exact numbers have significant figures." They do not. The "2" in the formula C = 2\pi r is exactly 2 — not 2.0 or 2.00 or 2.000. It has infinite significant figures because it is a definition, not a measurement. The same applies to conversion factors like "1 km = 1000 m" — both numbers are exact. When you multiply a measurement by an exact number, the measurement's significant figures alone determine the result.
-
"Addition uses the same rule as multiplication." It does not. For addition and subtraction, match decimal places. For multiplication and division, match significant figures. Mixing up the two is one of the most common errors in JEE Mains physics.
Why trailing zeros and leading zeros behave differently
This is worth pausing on, because the asymmetry feels arbitrary until you see the reason.
A leading zero is forced on you by the number system. The measurement 0.0042 kg is the same as 4.2 g — the leading zeros appear only because you chose to write the number in kilograms. They tell you about the unit, not the measurement. Switch units, and the zeros disappear.
A trailing zero after the decimal point is a choice you made as a measurer. When you write 4.20 g instead of 4.2 g, you are saying: "I checked the hundredths place, and it was zero." That zero is a genuine measurement result — it is a digit your instrument reported. You could have measured 4.21 or 4.19, but you did not — you got 4.20. Dropping that zero would throw away information.
The acid test is always the same: rewrite the number in scientific notation. Any digit that survives the rewrite is significant.
- 0.0042 \to 4.2 \times 10^{-3} — two sig figs survive
- 4.20 \to 4.20 \times 10^{0} — three sig figs survive
- 1500 \to either 1.5 \times 10^3 or 1.500 \times 10^3 — ambiguous until you decide
If you understand how to count significant figures, apply the rounding rules, and propagate through calculations, you have the working toolkit. What follows covers scientific notation, order of magnitude, and the connection to formal uncertainty — material that appears in JEE Advanced and competitive physics.
Scientific notation
Scientific notation writes every number as a product of two parts:
where 1 \leq |a| < 10 and n is an integer. The number a is called the significand (or mantissa, though that term has a different meaning in logarithms) and n is the exponent or order of magnitude.
The speed of light: c = 2.998 \times 10^8 m/s — four significant figures, order of magnitude 10^8.
The charge on an electron: e = 1.602 \times 10^{-19} C — four significant figures, order of magnitude 10^{-19}.
The mass of Earth: M = 5.972 \times 10^{24} kg — four significant figures, order of magnitude 10^{24}.
ISRO's GSLV Mk III carries about 2.8 \times 10^5 kg of propellant. Writing "280000 kg" would leave the number of significant figures ambiguous. Writing 2.8 \times 10^5 kg makes it clear: two significant figures. If the mass were known more precisely, you would write 2.80 \times 10^5 kg (three sig figs) or 2.800 \times 10^5 kg (four sig figs).
Scientific notation solves three problems at once:
- It eliminates the trailing-zero ambiguity.
- It makes very large and very small numbers easier to read and compare.
- It separates the precision (the significand) from the scale (the exponent), so you can see both at a glance.
Order of magnitude — the physicist's approximation
Two quantities have the same order of magnitude if they are within a factor of 10 of each other. More precisely, the order of magnitude of a number is the power of 10 closest to it.
The mass of a cricket ball is about 0.16 kg — order of magnitude 10^{-1} kg. The mass of an autorickshaw is about 300 kg — order of magnitude 10^{2.5}, which rounds to 10^3 kg. The ratio of their masses is about 10^{3.3}, meaning the autorickshaw is roughly a thousand times heavier — about 3.3 orders of magnitude apart.
Order-of-magnitude estimates are not about getting the right answer to three decimal places. They are about knowing whether your answer is in the right ballpark. If you calculate that a cricket ball weighs 0.016 kg and do not notice that this is off by a factor of 10 from reality, you have made an error. Order-of-magnitude thinking catches errors like that before they propagate.
The connection to formal uncertainty
Significant figures are a shorthand for something deeper: measurement uncertainty. When you write 20.12 m, you are implicitly saying the true value lies somewhere in the range 20.12 \pm 0.005 m — the uncertainty is half a unit in the last significant digit.
In advanced physics and engineering, this implicit uncertainty is replaced by an explicit one:
The \pm 0.01 m tells you exactly how uncertain the measurement is. This is more informative than significant figures alone, because:
- It can express asymmetric uncertainty: T = 2.34\,^{+0.02}_{-0.03} s.
- It distinguishes between a measurement with a large systematic error and one with a small random error.
- It propagates through calculations using the formal rules of error propagation (covered in Errors in Measurement).
Significant figures are the training wheels for uncertainty analysis. They give you the right habits — do not claim more precision than your data supports, propagate precision through calculations, round only at the end — without requiring the full mathematical machinery. When you reach the level where you need that machinery, the transition is natural because the underlying principle is the same.
Powers and roots
When you raise a measured value to a power, the number of significant figures in the result equals the number in the base. If a cube has side a = 2.3 cm (2 sig figs), its volume is:
Round to 2 significant figures: V = 12 cm³.
Why: raising to a power is repeated multiplication. The same rule applies — the result cannot have more significant figures than the input. Formally, the relative uncertainty in a^3 is 3 times the relative uncertainty in a, which generally costs about one significant figure.
For square roots, the same principle applies. If A = 150 cm² (assuming 3 sig figs), then \sqrt{A} = \sqrt{150} = 12.247... cm, which rounds to 12.2 cm (3 sig figs).
Where this leads next
- Errors in Measurement — the formal treatment of uncertainty: absolute and relative errors, systematic vs random errors, and how to propagate uncertainty through any calculation.
- Units and the SI System — the seven base units, derived units, and dimensional analysis as a tool for building physical intuition.
- Estimation and Order of Magnitude — how physicists make quick, back-of-the-envelope calculations using powers of 10 to check whether an answer is reasonable.