Chapter 5 of 9 · 18 min read

Chapter 3: Errors in Measurement

What you will learn

  • The difference between accuracy and precision

  • The kinds of error and which ones repeated readings can reduce

  • How to calculate absolute, mean absolute, relative and percentage error

  • How errors combine when quantities are added, multiplied or raised to powers

3.1 Accuracy and precision

No measurement is perfect. The difference between the measured value and the true value is the error.

  • Accuracy tells how close a measurement is to the true value.

  • Precision tells how close repeated measurements are to each other, and how fine the instrument's steps are (its least count).

A thermometer with 0.1 °C marks is precise. If it is wrongly calibrated and reads 2 °C high, it is precise but not accurate.

Figure 3.1 · Accuracy and precision as shots on a target
Figure 3.1 · Accuracy and precision as shots on a target

The centre is the true value and each dot is one reading. A miscalibrated but fine instrument gives the second pattern.

3.2 Types of error

Systematic errors push every reading in the same direction, always too high or always too low. Repeating the measurement does not remove them. Their sources:

  • Instrumental: a badly calibrated scale, a zero error, a stretched tape.

  • Imperfect method: measuring body temperature by placing the thermometer under the arm always reads lower than the true core temperature.

  • Personal: a student who always reads the scale from slightly above, causing the same parallax every time.

Systematic errors are reduced by better instruments, correct methods and removing personal bias.

Random errors change in size and sign from one reading to the next, because of unpredictable changes such as small voltage fluctuations or air currents. They are reduced by taking many readings and averaging.

Least count error comes from the resolution of the instrument. It is reduced by using an instrument with a smaller least count.

Gross errors are plain blunders: recording 3.52 as 5.32, or reading the wrong scale. Careful work avoids them.

3.3 Absolute, relative and percentage error

Suppose you take n readings a₁, a₂, …, aₙ.

Step 1. Mean value (the best estimate of the true value):

amean=a1+a2+⋯+anna_{\text{mean}} = \frac{a_1 + a_2 + \dots + a_n}{n}

Step 2. Absolute error of each reading:

Δai=amean−ai\Delta a_i = a_{\text{mean}} - a_i

This can be positive or negative.

Step 3. Mean absolute error (average of the magnitudes):

Δamean=∣Δa1∣+∣Δa2∣+⋯+∣Δan∣n\Delta a_{\text{mean}} = \frac{|\Delta a_1| + |\Delta a_2| + \dots + |\Delta a_n|}{n}

The result is written as a = a_mean ± Δa_mean.

Step 4. Relative and percentage error:

δa=Δameanamean,Percentage error=Δameanamean×100%\delta a = \frac{\Delta a_{\text{mean}}}{a_{\text{mean}}}, \qquad \text{Percentage error} = \frac{\Delta a_{\text{mean}}}{a_{\text{mean}}} \times 100\%

Absolute error has the same unit as the quantity. Relative and percentage errors have no unit.

Worked example 3.1: A student times 20 oscillations of a pendulum five times and finds the period as 1.52 s, 1.48 s, 1.55 s, 1.45 s and 1.50 s. Express the result with its error.

Solution:

Readingaᵢ (s)Δaᵢ = mean − aᵢ (s)Magnitude (s)
11.52−0.020.02
21.48+0.020.02
31.55−0.050.05
41.45+0.050.05
51.500.000.00

Mean = 7.50/5 = 1.50 s. Mean absolute error = 0.14/5 = 0.028 ≈ 0.03 s.

Result: T = (1.50 ± 0.03) s. Relative error = 0.03/1.50 = 0.02, so the percentage error is 2%.

3.4 How errors combine

When we calculate a result from measured quantities, their errors carry into the result. NEET always takes the worst case, so errors always add.

Rule 1: Sum or difference, add the absolute errors.

Z=A±B⇒ΔZ=ΔA+ΔBZ = A \pm B \quad\Rightarrow\quad \Delta Z = \Delta A + \Delta B

Rule 2: Product or quotient, add the relative errors.

Z=AB  or  Z=AB⇒ΔZZ=ΔAA+ΔBBZ = AB \;\text{or}\; Z = \frac{A}{B} \quad\Rightarrow\quad \frac{\Delta Z}{Z} = \frac{\Delta A}{A} + \frac{\Delta B}{B}

Rule 3: Powers, multiply each relative error by its power (ignore the sign of the power).

Z=ApBqCr⇒ΔZZ=pΔAA+qΔBB+rΔCCZ = \frac{A^{p} B^{q}}{C^{r}} \quad\Rightarrow\quad \frac{\Delta Z}{Z} = p\frac{\Delta A}{A} + q\frac{\Delta B}{B} + r\frac{\Delta C}{C}

Constants such as 2, π or 1/2 carry no error.

Worked example 3.2: Two rods measure (10.0 ± 0.1) cm and (5.0 ± 0.2) cm. Find their total length and their difference.

Solution: Sum = (15.0 ± 0.3) cm. Difference = (5.0 ± 0.3) cm. The error is added in both cases.

Worked example 3.3: In an Ohm's law experiment, V = (100 ± 5) V and I = (10 ± 0.2) A. Find the percentage error in R = V/I.

Solution: ΔR/R = 5/100 + 0.2/10 = 0.05 + 0.02 = 0.07, so 7%.

Worked example 3.4: Z = A² B³ / √C. The percentage errors in A, B and C are 1%, 1% and 2%. Find the percentage error in Z.

Solution: 2(1%) + 3(1%) + (1/2)(2%) = 2 + 3 + 1 = 6%.

Worked example 3.5: In a simple pendulum, g = 4π²L/T². L is measured with 0.5% error and T with 1% error. Find the percentage error in g.

Solution: Δg/g = ΔL/L + 2ΔT/T = 0.5% + 2% = 2.5%.

Worked example 3.6: A rectangle has l = (4.0 ± 0.1) cm and b = (2.0 ± 0.1) cm. Find its area with error.

Solution: A = 8.0 cm². ΔA/A = 0.1/4.0 + 0.1/2.0 = 0.025 + 0.05 = 0.075. ΔA = 0.075 × 8.0 = 0.6 cm². Area = (8.0 ± 0.6) cm².

Error from least count alone. If only one reading is taken, the absolute error is taken as the least count. A length of 5.0 cm read on a scale with LC 0.1 cm has a percentage error of 0.1/5.0 × 100 = 2%.

Quick revision

  • Accuracy: close to true value. Precision: close to each other, finer least count.

  • Systematic errors have one direction; repeating does not remove them. Random errors fall with averaging.

  • a = a_mean ± Δa_mean; % error = (Δa_mean / a_mean) × 100.

  • Add or subtract: add absolute errors.

  • Multiply or divide: add relative errors.

  • Powers: multiply relative error by the power.

NEET traps

  • In a difference, errors still add. (10.0 ± 0.2) − (9.8 ± 0.2) = 0.2 ± 0.4, an error of 200%. Subtracting close values gives large percentage errors.

  • A power of 1/2 halves the error; a power of −3 still triples it (use the magnitude).

  • The mean absolute error uses magnitudes, so it can never be negative.

  • Round the mean absolute error to match the decimal places of the mean.

Practice MCQs: Chapter 3

1. The precision of a measurement depends mainly on the: (a) true value (b) least count of the instrument (c) number of observers (d) unit used

2. A stopwatch that runs slow causes: (a) random error (b) systematic error (c) gross error (d) no error

3. Random errors can be reduced by: (a) using a new unit (b) taking many readings and averaging (c) correcting zero error (d) changing the method

4. Readings of a length are 2.31, 2.33, 2.30, 2.34 and 2.32 cm. The mean value is: (a) 2.31 cm (b) 2.32 cm (c) 2.33 cm (d) 2.30 cm

5. For the data in Q4, the mean absolute error is about: (a) 0.001 cm (b) 0.01 cm (c) 0.02 cm (d) 0.06 cm

6. For the data in Q4, the percentage error is about: (a) 0.05% (b) 0.5% (c) 5% (d) 1.2%

7. (10.0 ± 0.1) m + (5.0 ± 0.2) m equals: (a) 15.0 ± 0.1 m (b) 15.0 ± 0.2 m (c) 15.0 ± 0.3 m (d) 15.0 ± 0.02 m

8. (10.0 ± 0.1) m − (5.0 ± 0.2) m equals: (a) 5.0 ± 0.1 m (b) 5.0 ± 0.3 m (c) 5.0 ± 0.2 m (d) 5.0 ± 0.0 m

9. V has 3% error and I has 2% error. The percentage error in R = V/I is: (a) 1% (b) 1.5% (c) 5% (d) 6%

10. The radius of a sphere has a 2% error. The percentage error in its volume is: (a) 2% (b) 4% (c) 6% (d) 8%

11. Z = A² B³ / C^(1/2). Errors in A, B, C are 1%, 1%, 2%. The error in Z is: (a) 4% (b) 5% (c) 6% (d) 7%

12. In g = 4π²L/T², L has 0.5% error and T has 1% error. The error in g is: (a) 1.5% (b) 2% (c) 2.5% (d) 3%

13. The momentum of a body is measured with 3% error. The percentage error in its kinetic energy (mass exact) is: (a) 3% (b) 6% (c) 9% (d) 1.5%

14. The side of a cube has 1% error and its mass is exact. The percentage error in its density is: (a) 1% (b) 2% (c) 3% (d) 4%

15. A length of 5.0 cm is read once on a scale with least count 0.1 cm. The percentage error is: (a) 0.2% (b) 1% (c) 2% (d) 5%

16. A rectangle has l = (4.0 ± 0.1) cm and b = (2.0 ± 0.1) cm. Its area is: (a) 8.0 ± 0.2 cm² (b) 8.0 ± 0.6 cm² (c) 8.0 ± 0.1 cm² (d) 8.0 ± 0.3 cm²

17. Which of these is not a systematic error? (a) Zero error of a scale (b) Always reading from the same wrong angle (c) Random voltage fluctuations (d) Faulty calibration

18. Readings that are close to each other but far from the true value are: (a) accurate and precise (b) precise but not accurate (c) accurate but not precise (d) neither

19. a = (10.0 ± 0.2) and b = (9.8 ± 0.2). The percentage error in (a − b) is: (a) 2% (b) 4% (c) 100% (d) 200%

20. Heat produced is H = I²Rt. Errors in I, R and t are 2%, 1% and 1%. The error in H is: (a) 4% (b) 5% (c) 6% (d) 8%

21. For T = 2π√(L/g), L is measured with 2% error and g is exact. The error in T is: (a) 1% (b) 2% (c) 4% (d) 0.5%

22. Relative error is: (a) in the unit of the quantity (b) dimensionless (c) always greater than 1 (d) in the square of the unit

23. Assertion (A): Systematic errors cannot be removed by repeating the measurement many times. Reason (R): Systematic errors affect every reading in the same direction. (a) Both true, R explains A (b) Both true, R does not explain A (c) A true, R false (d) A false, R true

24. Statement I: Relative error has no unit. Statement II: The absolute error of a single reading can be negative. (a) Both true (b) Both false (c) I true, II false (d) I false, II true

25. Pressure P = F/(πr²). F has 2% error and r has 1% error. The error in P is: (a) 1% (b) 3% (c) 4% (d) 5%

Answer key with reasons

QAnsReason
1bFiner least count, finer readings.
2bEvery reading is off in the same direction.
3bPositive and negative errors cancel in the mean.
4b11.60/5 = 2.32 cm.
5b(0.01 + 0.01 + 0.02 + 0.02 + 0)/5 = 0.012 cm.
6b0.012/2.32 × 100 ≈ 0.52%.
7cAbsolute errors add.
8bAbsolute errors add, even in subtraction.
9c3% + 2%.
10cV ∝ r³: 3 × 2%.
11cWorked example 3.4.
12cWorked example 3.5.
13bK = p²/2m: 2 × 3%.
14cρ = m/a³: 3 × 1%.
15c0.1/5.0 × 100.
16bWorked example 3.6.
17cFluctuations vary in sign and size: random error.
18bAgreement with each other is precision; nearness to truth is accuracy.
19d0.2 ± 0.4; 0.4/0.2 × 100 = 200%.
20c2 × 2% + 1% + 1%.
21aT ∝ L^(1/2): (1/2) × 2%.
22bIt is a ratio of like quantities.
23aSame direction means averaging cannot cancel them.
24aΔaᵢ = mean − aᵢ can be either sign.
25c2% + 2 × 1%.

Chapter 3 revision sheet

One-page summary

  • Error = measured value − true value. Every measurement has some error.

  • Accuracy: how close a reading is to the true value. Precision: how close repeated readings are to each other.

  • Systematic errors push readings one way. Types: instrumental (zero error), imperfect technique, personal (reading a scale at an angle). Repeating does not remove them; correction does.

  • Random errors change irregularly from reading to reading. Averaging many readings reduces them.

  • Least count error comes from the instrument's resolution. With a single reading, take the error as the least count.

  • Gross errors are plain mistakes (wrong recording, faulty set-up). Discard such readings.

  • Mean value a_mean = (a₁ + a₂ + … + aₙ)/n is the best estimate.

  • Absolute error of each reading = magnitude of (a_mean − aᵢ). Mean absolute error = average of these magnitudes.

  • Relative error = Δa_mean / a_mean. Percentage error = relative error × 100. Both have no unit.

  • Sum or difference: absolute errors always add, even for a difference.

  • Product or quotient: relative errors add.

  • Powers: relative error is multiplied by the power, and the sign of the power is ignored.

  • Exact constants (2, π, ½) carry no error.

  • The quantity with the highest power gives the biggest share of the error, so measure it most carefully.

  • Subtracting two nearly equal values gives a large percentage error.

If Z=ApBqCr,ΔZZ=pΔAA+qΔBB+rΔCC\text{If } Z = \frac{A^p B^q}{C^r}, \quad \frac{\Delta Z}{Z} = p\frac{\Delta A}{A} + q\frac{\Delta B}{B} + r\frac{\Delta C}{C}

Must-remember table

RelationError rule
Z = A + B or A − BΔZ = ΔA + ΔB
Z = AB or A/BΔZ/Z = ΔA/A + ΔB/B
Z = AⁿΔZ/Z = n ΔA/A
Z = kA (k exact)ΔZ/Z = ΔA/A
g = 4π²L/T²Δg/g = ΔL/L + 2ΔT/T
R = V/IΔR/R = ΔV/V + ΔI/I
ρ = m/a³ (cube)Δρ/ρ = Δm/m + 3Δa/a
K = p²/2mΔK/K = 2Δp/p + Δm/m
Single readingΔa = least count

Memory aids

  • "SAD PAR": Sum And Difference add Absolute errors; Products And Ratios add Relative errors.

  • "Power is a multiplier": the power in the formula multiplies that quantity's relative error.

  • "Errors never cancel": every term is added in the worst case, so no minus signs appear in error sums.

  • "SIR-P" for systematic error types: Instrumental, Reading technique (imperfect method), Personal.

  • "Repeat for Random": repetition helps only random errors.

Concept connections

  • Physics Book 20: Experimental Skills asks for the error in g from a pendulum and in resistance from a meter bridge.

  • Physics Book 12: Current Electricity uses R = V/I, where the relative errors in V and I add.

  • Physics Book 7: Properties of Solids and Liquids: Young's modulus Y = FL/(πr²Δl) has an error dominated by the radius term (power 2).

Practice MCQs: Chapter 3, new NEET pattern (Q26–40)

26. Match the error in List I with its example in List II.

List I:

  • A. Instrumental error

  • B. Personal error

  • C. Random error

  • D. Gross error

List II:

  • i. Writing 2.8 as 8.2 in the notebook

  • ii. Zero error of a screw gauge

  • iii. Always reading a scale from slightly to one side

  • iv. Scatter caused by unpredictable air currents

(a) A-iii, B-ii, C-iv, D-i (b) A-ii, B-iv, C-iii, D-i (c) A-i, B-iii, C-iv, D-ii (d) A-ii, B-iii, C-iv, D-i

27. The percentage errors in A, B and C are 1%, 2% and 3%. Match each Z in List I with its maximum percentage error in List II.

List I:

  • A. Z = AB

  • B. Z = A²B

  • C. Z = A³/C

  • D. Z = B²C

List II:

  • i. 6%

  • ii. 3%

  • iii. 7%

  • iv. 4%

(a) A-iv, B-ii, C-i, D-iii (b) A-ii, B-iv, C-i, D-iii (c) A-ii, B-iv, C-iii, D-i (d) A-iii, B-iv, C-i, D-ii

28. Match the relation in List I with the error rule in List II.

List I:

  • A. Z = A − B

  • B. Z = A/B

  • C. Z = Aⁿ

  • D. Z = kA, k an exact number

List II:

  • i. ΔZ/Z = ΔA/A + ΔB/B

  • ii. ΔZ/Z = ΔA/A

  • iii. ΔZ = ΔA + ΔB

  • iv. ΔZ/Z = nΔA/A

(a) A-iii, B-i, C-iv, D-ii (b) A-i, B-iii, C-iv, D-ii (c) A-iii, B-i, C-ii, D-iv (d) A-iv, B-i, C-iii, D-ii

29. Match each measurement in List I with its percentage error in List II.

List I:

  • A. (20.0 ± 0.1) cm

  • B. (5.0 ± 0.1) cm

  • C. (4.0 ± 0.1) s

  • D. (50 ± 0.5) g

List II:

  • i. 2.5%

  • ii. 1%

  • iii. 0.5%

  • iv. 2%

(a) A-iv, B-iii, C-i, D-ii (b) A-iii, B-iv, C-ii, D-i (c) A-iii, B-iv, C-i, D-ii (d) A-ii, B-iv, C-i, D-iii

30. How many of the following statements are correct?

  • A. Repeating a measurement many times reduces random error.

  • B. Repeating a measurement many times reduces systematic error.

  • C. A precise instrument always gives accurate readings.

  • D. When two quantities are subtracted, their absolute errors add.

  • E. Relative error has no unit.

(a) Two (b) Four (c) Three (d) Five

31. g is found from g = 4π²L/T². Choose the correct set of statements.

  • A. Δg/g = ΔL/L + 2ΔT/T

  • B. The factor 4π² adds to the error in g.

  • C. If L has 1% error and T has 0.5% error, g has 2% error.

  • D. A 1% error in T affects g more than a 1% error in L.

(a) A and C only (b) A, C and D only (c) B, C and D only (d) A, B and D only

32. Five readings of a time period are 4.2 s, 4.4 s, 4.3 s, 4.1 s and 4.5 s. Choose the correct set.

  • A. The mean is 4.3 s.

  • B. The mean absolute error is 0.12 s.

  • C. The percentage error is about 2.8%.

  • D. The result is written as (4.3 ± 0.5) s.

(a) A, B and C only (b) A and D only (c) B and C only (d) A, B, C and D

33. How many of the following statements are correct?

  • A. Accuracy is closeness to the true value.

  • B. Precision is closeness of repeated readings to one another.

  • C. Readings 5.1, 5.1 and 5.2 of a true value 6.0 are precise but not accurate.

  • D. Precision depends on the least count of the instrument.

  • E. A mean absolute error can be negative.

(a) Five (b) Three (c) Two (d) Four

34. Four students measure a rod whose true length is 10.00 cm. Their readings are in the table.

StudentReading 1 (cm)Reading 2 (cm)Reading 3 (cm)
P10.0110.009.99
Q9.509.519.50
R9.610.410.0
S11.28.79.4

Whose readings are precise but not accurate?

(a) P (b) Q (c) R (d) S

35. Single lengths are measured with a scale of least count 1 mm. A graph of percentage error (y-axis) against length measured (x-axis) is drawn. At 5 cm the percentage error is 2%. What is it at 20 cm, and what shape does the graph have?

(a) 8%, straight line rising (b) 0.5%, straight line through the origin (c) 0.5%, falls along a rectangular hyperbola (d) 2%, horizontal line

36. The volume of a cylinder is V = πr²h. The data are below.

QuantityValueAbsolute error
Radius r2.00 cm0.02 cm
Height h10.0 cm0.2 cm

The percentage error in V and the absolute error in V are about

(a) 3%, 3.8 cm³ (b) 5%, 6.3 cm³ (c) 4%, 2.5 cm³ (d) 4%, 5.0 cm³

Case for Q37–38: Meena finds the density of a metal cube in the school lab. She measures its side as (2.00 ± 0.01) cm with vernier callipers. Its mass on a balance is (64.0 ± 0.4) g. She uses ρ = m/a³.

37. The percentage error in the density is about (a) 2.1% (b) 1.1% (c) 0.6% (d) 1.6%

38. Meena should report the density as (a) (8.0 ± 0.1) g/cm³ (b) (8.0 ± 0.4) g/cm³ (c) (8.0 ± 0.2) g/cm³ (d) (4.0 ± 0.2) g/cm³

39. The momentum of a body is measured with 2% error and its mass with 1% error. The maximum percentage error in its kinetic energy is (a) 3% (b) 4% (c) 6% (d) 5%

40. A = (50.0 ± 0.2) cm and B = (48.0 ± 0.2) cm. The percentage error in Z = A − B is (a) 20% (b) 0.4% (c) 0.8% (d) 10%

Answer key (Q26–40)

QAnsReason
26dZero error is instrumental, side view is personal, air currents random, miswriting gross
27b1+2 = 3, 2+2 = 4, 3+3 = 6, 4+3 = 7
28aDifference adds absolute errors; quotient adds relative errors; power multiplies; exact k adds nothing
29c0.1/20 = 0.5%, 0.1/5 = 2%, 0.1/4 = 2.5%, 0.5/50 = 1%
30cA, D, E true
31b4π² is exact; 1 + 2(0.5) = 2%; T has power 2
32aDeviations 0.1, 0.1, 0, 0.2, 0.2; mean 0.12 s; 0.12/4.3 = 2.8%; result (4.3 ± 0.1) s
33dA to D true; it is a mean of magnitudes, never negative
34bQ's readings agree closely but sit 0.5 cm below the true value
35c% error = 0.1/L × 100, so 0.1/20 × 100 = 0.5%; y × x is constant
36d2(1%) + 2% = 4%; V = 125.7 cm³; 4% of it is 5.0 cm³
37a0.4/64 + 3(0.01/2) = 0.625% + 1.5% ≈ 2.1%
38cρ = 64/8 = 8.0 g/cm³; 2.125% of 8.0 = 0.17, rounded to 0.2
39dK = p²/2m, so 2(2%) + 1% = 5%
40aZ = 2.0 cm, ΔZ = 0.4 cm, 0.4/2.0 = 20%
18 min left
‹
›