Why measurement is uncertain
In the study of chemistry one constantly handles experimental data alongside theoretical calculations, and the numbers involved must be presented realistically — with certainty to the extent possible, but no further. A measurement is never a single perfect value; it is a best estimate accompanied by a margin of doubt. The mass of an object weighed on a platform balance might read 9.4 g, while the same object on an analytical balance reads 9.4213 g. The platform balance simply cannot resolve the digits beyond the first decimal, so the figure 4 after the decimal is the uncertain digit in that reading.
This uncertainty is communicated by the number of significant figures reported. Significant figures are the meaningful digits known with certainty, plus one final digit that is estimated or uncertain. When a result is written as 11.2 mL, the convention is that 11 is certain and 2 is uncertain; unless otherwise stated, an uncertainty of $\pm 1$ in the last digit is always understood.
A reported value is a promise about its own reliability. Writing 9.4 g and writing 9.40 g are different scientific statements: the second claims knowledge of the hundredths place, the first does not.
Scientific notation
Chemistry is the study of atoms and molecules, which have extremely small masses yet occur in enormous numbers. A chemist routinely meets a quantity as large as 602 200 000 000 000 000 000 000 (the number of molecules in 2 g of hydrogen gas) and as small as 0.000 000 000 000 000 000 000 001 66 g (the mass of one hydrogen atom). Counting zeros is error-prone, and ordinary arithmetic on such strings is a genuine challenge.
The remedy is scientific notation: any number is written in the exponential form $N \times 10^{n}$, where $n$ is an exponent that may be positive or negative, and $N$ is the digit term lying between $1.000\ldots$ and $9.999\ldots$. The decimal point is shifted until exactly one non-zero digit stands to its left; the number of places shifted becomes the exponent.
Express 232.508 and 0.00016 in scientific notation.
For 232.508 the decimal is moved left by two places, so the exponent is $+2$:
$$232.508 = 2.32508 \times 10^{2}$$
For 0.00016 the decimal is moved right by four places, so the exponent is $-4$:
$$0.00016 = 1.6 \times 10^{-4}$$
Scientific notation also resolves the ambiguity of trailing zeros, and it makes the four arithmetic operations systematic. Multiplication and division follow the ordinary index laws — the digit terms are combined and the exponents are added or subtracted. Addition and subtraction require an extra preparatory step.
Add $6.65 \times 10^{4}$ and $8.95 \times 10^{3}$; then subtract $4.8 \times 10^{-3}$ from $2.5 \times 10^{-2}$.
First make the exponents equal, then combine the digit terms:
$$ (6.65 \times 10^{4}) + (0.895 \times 10^{4}) = (6.65 + 0.895)\times 10^{4} = 7.545 \times 10^{4}$$
$$ (2.5 \times 10^{-2}) - (0.48 \times 10^{-2}) = (2.5 - 0.48)\times 10^{-2} = 2.02 \times 10^{-2}$$
Significant-figure rules
The whole apparatus of uncertainty rests on five rules for counting significant figures. They turn the loose idea of "meaningful digits" into a procedure that can be applied to any number without ambiguity. The rules are summarised below with the NCERT illustrations.
| # | Rule | Example | Sig. figs |
|---|---|---|---|
| 1 | All non-zero digits are significant. | 285 cm; 0.25 mL | 3; 2 |
| 2 | Zeros preceding the first non-zero digit are not significant — they only fix the decimal position. | 0.03; 0.0052 | 1; 2 |
| 3 | Zeros between two non-zero digits are significant. | 2.005 | 4 |
| 4 | Trailing zeros are significant only if they lie to the right of the decimal point. | 0.200 g; 100 | 3; 1 |
| 5 | Exact (counted / defined) numbers have infinite significant figures. | 20 eggs = 20.000… | ∞ |
Two consequences of these rules deserve emphasis. First, in any number written in scientific notation all digits of the term $N$ are significant: $4.01 \times 10^{2}$ has three significant figures and $8.256 \times 10^{-3}$ has four. Second, the bare integer 100 is treated as having a single significant figure, because a trailing zero with no decimal point is not significant. To declare two or three significant figures one must use scientific notation, as the table of equivalents below shows.
| Plain form | Intended sig. figs | Unambiguous scientific notation |
|---|---|---|
100 | 1 | $1 \times 10^{2}$ |
100 | 2 | $1.0 \times 10^{2}$ |
100.0 | 4 | $1.000 \times 10^{2}$ |
Leading zeros versus trailing zeros
The most common slip is treating every zero alike. Leading zeros (as in 0.0052) are never significant; trailing zeros after a decimal point (as in 0.200) always are; trailing zeros in a whole number with no decimal point (as in 100) are not. Confusing these changes the sig-fig count and therefore the final reported answer.
If a zero only locates the decimal, drop it from the count. If it certifies a measured place, keep it.
Operations with significant figures
A calculated result can be no more reliable than the least reliable measurement entering the calculation. Two distinct limiting rules apply, and the most frequent error in NEET numericals is interchanging them.
Addition and subtraction — limit by decimal places
For addition and subtraction the result cannot have more digits to the right of the decimal point than the original number with the fewest decimal places.
Add 12.11, 18.0 and 1.012.
$$12.11 + 18.0 + 1.012 = 31.122$$
Here 18.0 has only one digit after the decimal point — the fewest — so the answer is reported to one decimal place: 31.1.
Multiplication and division — limit by significant figures
For multiplication and division the result must be reported with no more significant figures than the factor having the fewest significant figures.
Evaluate $2.5 \times 1.25$ to the correct number of significant figures.
$$2.5 \times 1.25 = 3.125$$
Since 2.5 has only two significant figures, the answer cannot exceed two: 3.1.
Decimal places are not significant figures
Addition is governed by decimal places; multiplication by significant figures. Applying the sig-fig rule to a sum, or the decimal-place rule to a product, is a classic loss of an otherwise free mark. Round only the final answer, never the intermediate steps, to avoid compounding rounding error.
Sum or difference → match the fewest decimal places. Product or quotient → match the fewest significant figures.
These sig-fig rules carry straight into mole and mass problems — see Stoichiometry Calculations to apply them on real reaction data.
Rounding off the numbers
When a result is trimmed to the required number of significant figures, the digit immediately to the right of the last kept digit decides what happens. NCERT states three rounding rules, the third of which — the round-half-to-even convention — is what separates careful students from careless ones.
| Digit being removed | Rule | Example | Rounded |
|---|---|---|---|
| Greater than 5 | Increase the preceding digit by one. | 1.386 (remove 6) | 1.39 |
| Less than 5 | Leave the preceding digit unchanged. | 4.334 (remove 4) | 4.33 |
| Exactly 5 | Leave the preceding digit unchanged if it is even; increase by one if it is odd. | 6.35 → odd 3 | 6.4 |
| Exactly 5 | (same rule) preceding digit even | 6.25 → even 2 | 6.2 |
The round-half-to-even rule for a trailing 5 exists to prevent a systematic upward bias: if one always rounded 0.5 upward, a long series of such roundings would inflate the total. Rounding to the nearest even digit balances the ups and downs over many measurements.
Precision versus accuracy
One always wishes results to be both precise and accurate, yet the two words describe different virtues. Precision is the closeness of repeated measurements of the same quantity to one another — their reproducibility. Accuracy is the agreement of a measured value with the true value. A set can be tightly clustered (precise) yet sit far from the truth (inaccurate) if a systematic error is present.
NCERT illustrates this with three students measuring a quantity whose true value is 2.00 g. Student A reports values close to each other but below the truth; Student B reports scattered values; Student C reports values that are both tight and centred on 2.00 g.
| Student | Measurement 1 / g | Measurement 2 / g | Average / g | Verdict |
|---|---|---|---|---|
| A | 1.95 | 1.93 | 1.940 | Precise, not accurate |
| B | 1.94 | 2.05 | 1.995 | Neither precise nor accurate |
| C | 2.01 | 1.99 | 2.000 | Precise and accurate |
Tight clustering signals precision; landing on the bullseye signals accuracy. The two are independent, and only the rightmost pattern delivers reliable data.
Dimensional analysis
Calculations frequently require converting a quantity from one system of units to another. The method used is the factor-label method, also called the unit-factor method or dimensional analysis. Its engine is the unit factor: a ratio of two equivalent quantities, which therefore equals one. Multiplying by such a factor changes the units of a quantity without changing its physical value, and the units themselves are cancelled, multiplied or squared exactly like algebraic symbols.
From the defined equivalence $1\ \text{in} = 2.54\ \text{cm}$ one can build two unit factors, $\dfrac{2.54\ \text{cm}}{1\ \text{in}}$ and $\dfrac{1\ \text{in}}{2.54\ \text{cm}}$, both equal to one. The one to use is the factor that places the required unit in the numerator so the unwanted unit cancels.
The inch unit appears once in the numerator (3 in) and once in the denominator of the unit factor, so it cancels, leaving the answer in centimetres.
A piece of metal is 3 inch long. What is its length in cm?
$$3\ \text{in} = 3\ \text{in} \times \frac{2.54\ \text{cm}}{1\ \text{in}} = 3 \times 2.54\ \text{cm} = 7.62\ \text{cm}$$
The method extends to compound and cubed units. To convert a volume, the linear unit factor is raised to the third power so that cubic units cancel cleanly. This is where careless students lose marks — they convert the linear factor but forget to cube it.
A jug contains 2 L of milk. Calculate the volume in $\text{m}^{3}$. (Given $1\ \text{L} = 1000\ \text{cm}^3$ and $1\ \text{m} = 100\ \text{cm}$.)
First write the volume in $\text{cm}^3$, then apply the cubed length unit factor:
$$2\ \text{L} = 2 \times 1000\ \text{cm}^3 = 2000\ \text{cm}^3$$
$$2000\ \text{cm}^3 \times \left(\frac{1\ \text{m}}{100\ \text{cm}}\right)^{3} = 2000 \times \frac{1}{10^{6}}\ \text{m}^3 = 2 \times 10^{-3}\ \text{m}^3$$
Cube the unit factor for cubed units
When the quantity carries a cubed (or squared) unit, the conversion factor must be raised to the same power. Using $\frac{1\ \text{m}}{100\ \text{cm}}$ instead of $\left(\frac{1\ \text{m}}{100\ \text{cm}}\right)^{3}$ for a volume gives an answer wrong by a factor of $10^{4}$.
Match the power of the unit factor to the power of the unit being converted.
Uncertainty & Significant Figures in one glance
- Significant figures = certain digits + one estimated digit; an uncertainty of $\pm 1$ in the last digit is implied.
- Scientific notation $N \times 10^{n}$ with $1 \le N < 10$ removes zero-counting errors and disambiguates trailing zeros.
- Sig-fig rules: non-zero digits count; leading zeros do not; sandwiched zeros do; trailing zeros count only after a decimal point; exact numbers are infinite.
- Addition/subtraction → limit by fewest decimal places. Multiplication/division → limit by fewest significant figures.
- Rounding: >5 round up; <5 round down; exactly 5 → round to the nearest even digit (6.35 → 6.4, 6.25 → 6.2).
- Precision = reproducibility; accuracy = closeness to the true value; reliable data is both.
- Dimensional analysis multiplies by unit factors (each equal to 1); cube the factor for cubed units.