21 min read
Accuracy and Precision
Accuracy describes how closely a measured value agrees with the true or accepted reference value.
Choose Your Level
Accuracy and Precision
Learning Outcomes
- Define measurement accuracy and precision.
- Distinguish accuracy from precision.
- Explain measurement trueness, repeatability, and bias.
- Identify accurate and precise sets of readings.
- Calculate the mean of repeated measurements.
- Calculate measurement error, percentage error, and range.
- Explain how random and systematic effects influence results.
- Describe methods for improving measurement quality.
- Avoid confusing instrument resolution with precision.
Imagine measuring the length of a pencil several times. If all readings are near the correct length, they show good accuracy. If all readings are almost identical, they show good precision. A measuring instrument can repeatedly give almost the same wrong value. In that case, it is precise but inaccurate. The best measurements should be: Close to the reference value Close to one another Obtained using a correctly calibrated instrument Reported with suitable uncertainty
Recommended for Class 11
Accuracy and Precision
Chapter 1 — Physical Quantities and Measurements · Topic: Accuracy and Precision
Learning Outcomes
- Define measurement accuracy and precision.
- Distinguish accuracy from precision.
- Explain measurement trueness, repeatability, and bias.
- Identify accurate and precise sets of readings.
- Calculate the mean of repeated measurements.
- Calculate measurement error, percentage error, and range.
- Explain how random and systematic effects influence results.
- Describe methods for improving measurement quality.
- Avoid confusing instrument resolution with precision.
Definition
Introduction
No practical measurement is perfectly exact. Measurements are influenced by: Instrument limitations Calibration Observational technique Environmental changes Random variations Systematic effects Suppose the accepted length of an object is 10.00 cm. The readings: 10.01 cm, 10.00 cm, 9.99 cm are close to the reference value and close to one another. They are both accurate and precise. However, the readings: 10.50 cm, 10.51 cm, 10.49 cm are close to one another but not to the reference value. They are precise but inaccurate.
Key Terms
Key Terms
| Term | Meaning |
|---|---|
| Measured value | Value obtained from an experiment |
| Reference value | Accepted value used for comparison |
| Accuracy | Closeness of a result to a reference value |
| Precision | Agreement among repeated results |
| Trueness | Closeness of the average result to a reference value |
| Repeatability | Precision under the same measurement conditions |
| Reproducibility | Precision under appropriately changed conditions |
| Measurement error | Measured value minus reference value |
| Bias | Difference between the average measured value and reference value |
| Range | Difference between highest and lowest readings |
| Resolution | Smallest change an instrument can display or distinguish |
| Calibration | Comparison and adjustment using a measurement standard |
Detailed Concept Explanation
5.1 Accuracy and reference value
Accuracy can be assessed only when a suitable reference value is available. Suppose: x_ref = 20.00 cm If the measured value is: x = 19.98 cm the reading is close to the reference value and has a small error. If the measured value is: x = 18.50 cm the reading has a much larger error and is less accurate.
5.2 Precision and repeated readings
Precision requires repeated measurements. Consider: 5.01 cm, 5.00 cm, 4.99 cm These readings have a small spread and therefore good precision. Now consider: 5.30 cm, 4.80 cm, 5.10 cm These readings have a large spread and therefore poor precision. A single measurement cannot properly establish repeatability or precision.
5.3 Measurement trueness
Trueness describes how closely the average of repeated measurements agrees with a reference value. The VIM relates better trueness to smaller systematic error or bias. JCGM Vocabulary: Measurement Trueness
5.4 Four possible cases
Four possible cases
| Case | Meaning |
|---|---|
| Accurate and precise | Readings are close to the reference and one another |
| Precise but inaccurate | Readings are grouped together but displaced from the reference |
| Close average but low precision | Readings are scattered, although their average may approach the reference |
| Neither accurate nor precise | Readings are scattered and displaced from the reference |
5.5 Relationship with measurement errors
Random effects generally increase the spread of repeated readings and reduce precision. Systematic effects may shift most readings in the same direction, reducing trueness and accuracy without necessarily reducing precision. For example, an incorrectly calibrated balance may repeatedly give: 51.0 g for a 50.0 g reference mass. The results may be precise but inaccurate.
5.6 Resolution is not precision
Resolution is the smallest change an instrument can display. An instrument displaying many decimal places has fine resolution, but its readings are not automatically accurate or precise. For example, a faulty balance might repeatedly display: 25.378 g when the reference value is: 24.500 g The additional decimal places do not make the result accurate.
Concept in Easy Words
Imagine measuring the length of a pencil several times. If all readings are near the correct length, they show good accuracy. If all readings are almost identical, they show good precision. A measuring instrument can repeatedly give almost the same wrong value. In that case, it is precise but inaccurate. The best measurements should be: Close to the reference value Close to one another Obtained using a correctly calibrated instrument Reported with suitable uncertainty
Important Facts
Important Facts
- Accuracy and precision are different concepts.
- Accuracy requires comparison with a reference value.
- Precision requires repeated measurements.
- A precise measurement is not necessarily accurate.
- An accurate-looking single result does not establish precision.
- Random variation mainly affects precision.
- Systematic bias mainly affects trueness.
- Calibration can help identify or reduce systematic bias.
- Repetition can help reveal random variation.
- More decimal places do not guarantee better accuracy.
- A smaller range or standard deviation generally indicates greater precision.
- Modern metrology recommends reporting a measurement result with its uncertainty.
Formulas and Mathematical Explanation
- x̄
- mean measured value
- x₁, x₂, …, xₙ
- individual readings
- n
- number of readings
8.1 Mean of repeated measurements For n readings: x̄ = x₁ + x₂ + ⋯ + xₙn
8.2 Measurement error e = xmeasured − xreference A positive error means the measured value is above the reference value. A negative error means the measured value is below the reference value.
8.3 Absolute error |e| = |xmeasured − xreference|
8.4 Percentage error At school level, percentage error provides a numerical indication of closeness when a reliable reference value is known: Percentage error = (|e| / |x_reference|) × 100%
8.5 Measurement bias b = x̄ − xreference Where: b = estimated bias x̄ = mean of repeated readings xreference = reference value
8.6 Range R = xmax − xmin A smaller range usually indicates better precision when comparing similar sets measured under the same conditions.
8.7 Half-range estimate A simple school-level estimate of the spread is: Δx = xmax − xmin2 This is useful for introductory work but is not a complete statistical evaluation of uncertainty.
8.8 Standard deviation A common statistical measure of spread is the sample standard deviation: s = √[ Σ(xᵢ − x̄)² / (n − 1) ] A smaller standard deviation generally represents better precision under the same measurement conditions.
Units, Symbols, and Measurement Language
Quantity, symbol, and unit
| Quantity | Symbol | Unit |
|---|---|---|
| Measured value | x | Same as measured quantity |
| Reference value | x_ref | Same as measured quantity |
| Mean | x̄ | Same as measured quantity |
| Error | e | Same as measured quantity |
| Bias | b | Same as measured quantity |
| Range | R | Same as measured quantity |
| Percentage error | — | % |
| Relative error | — | Dimensionless |
| Standard deviation | s | Same as measured quantity |
Accuracy and precision are qualitative concepts. They should not be carelessly treated as numerical quantities with units. Numerical measures such as error, bias, range, and standard deviation should be named explicitly. NIST Technical Note 1297
Solved Examples
Example 1: Accuracy and percentage error
The accepted length of a rod is 25.0 cm. A student measures it as 24.8 cm. Calculate the error and percentage error.
- xmeasured = 24.8 cm
- xreference = 25.0 cm
e = xmeasured − xreference
e = 24.8 cm − 25.0 cm
e = −0.2 cm
|e| = 0.2 cm
Percentage error = (0.225.0) × 100%
Percentage error = 0.008 × 100%
Percentage error = 0.8%
Key step: Use the absolute error when calculating percentage error.
Final answer: e = −0.2 cm; Percentage error = 0.8%
Example 2: Accurate and precise readings
A reference length is 10.00 cm. The readings are 10.01 cm, 10.00 cm, and 9.99 cm. Examine their accuracy and precision.
x̄ = 10.01 + 10.00 + 9.993 cm
x̄ = 30.003 cm
x̄ = 10.00 cm
R = xmax − xmin
R = 10.01 cm − 9.99 cm
R = 0.02 cm
Key step: Compare the mean with the reference and the readings with one another.
Final answer: The mean equals the reference value and the range is small, so the readings show excellent trueness and precision.
Example 3: Precise but inaccurate readings
A reference value is 10.00 cm. The readings are 10.50 cm, 10.51 cm, and 10.49 cm. Classify them.
x̄ = 10.50 + 10.51 + 10.493 cm = 10.50 cm
R = 10.51 cm − 10.49 cm = 0.02 cm
b = x̄ − xreference = 10.50 cm − 10.00 cm = 0.50 cm
Final answer: The small range shows good precision, but the large bias shows poor agreement with the reference value. The readings are precise but inaccurate.
Example 4: Comparing precision
Two students obtain these readings: Student A: 5.01, 5.00, 4.99 cm Student B: 5.20, 4.90, 5.10 cm Which set is more precise?
RA = 5.01 cm − 4.99 cm = 0.02 cm
RB = 5.20 cm − 4.90 cm = 0.30 cm
Comparison: RA < RB
Final answer: Student A’s measurements are more precise because their spread is smaller.
Example 5: Finding estimated bias
A 50.0 g reference mass gives readings of 51.0 g, 50.8 g, and 51.2 g. Calculate the mean and bias.
x̄ = 51.0 + 50.8 + 51.23 g = 51.0 g
b = x̄ − xreference = 51.0 g − 50.0 g = 1.0 g
The positive bias suggests a systematic problem such as incorrect calibration or zero error.
Final answer: x̄ = 51.0 g; b = +1.0 g
Real-Life Examples and Applications
Applications
- Medical thermometers must provide accurate temperature readings.
- Laboratory balances require calibration for accurate mass measurements.
- Repeated timing in sports improves confidence in the result.
- Manufacturing requires precise dimensions so parts fit together.
- Navigation systems require accurate position measurements.
- Electrical instruments must provide repeatable voltage and current readings.
- Pharmaceutical quantities require both accuracy and precision.
- Construction measurements must be close to specified values.
- Scientific experiments rely on repeated measurements and uncertainty evaluation.
Comparison Table
Accuracy versus precision
| Feature | Accuracy | Precision |
|---|---|---|
| Main idea | Closeness to reference value | Agreement among repeated values |
| Requires reference value | Yes | Not necessarily |
| Requires repeated readings | Not always | Yes |
| Mainly affected by | Both systematic and random effects | Random variability |
| Improved by calibration | Often | Not necessarily |
| Evaluated using | Error or bias as indicators | Range or standard deviation |
| Can exist without the other | Yes | Yes |
| Example | Mean near accepted value | Readings closely grouped |
Precision, resolution, and trueness
| Term | Meaning |
|---|---|
| Precision | Closeness among repeated measurements |
| Resolution | Smallest detectable or displayed change |
| Trueness | Closeness of the average result to a reference value |
| Accuracy | Overall closeness of a measured result to a reference value |
Key Logic
Use two separate questions: Are the readings close to the reference value? This evaluates accuracy or trueness. Are the repeated readings close to one another? This evaluates precision. For numerical comparison: Smaller error ⇒ better agreement with reference Smaller spread ⇒ better precision A good measurement procedure aims for both.
Tips and Tricks
Accuracy → Accepted value Precision → Packed readings Both words begin with helpful matching letters: Accuracy and Accepted Precision and Packed When comparing data: Calculate the mean. Compare the mean with the reference value. Calculate the range. Compare the spread of repeated readings. Look for a possible common shift or bias. Do not judge measurement quality only by the number of decimal places.
Common Mistakes
Common Mistakes
| Common mistake | Correct idea |
|---|---|
| Treating accuracy and precision as synonyms | They describe different properties |
| Calling close readings automatically accurate | They may all be displaced from the reference |
| Judging precision from one reading | Precision requires repeated measurements |
| Assuming more decimal places mean accuracy | Display resolution alone is insufficient |
| Ignoring calibration error | It may shift every reading |
| Using signed error in percentage-error magnitude | Use absolute error |
| Comparing ranges with different units | Convert to the same unit first |
| Assuming repetition removes systematic bias | Repetition mainly reveals random variation |
| Calling range “precision” with units | State that range is a measure of spread |
| Reporting an answer beyond instrument capability | Match reporting to measurement quality |
Topic Summary
Accuracy describes how closely a measurement agrees with a reference value, while precision describes agreement among repeated measurements. A precise set may still be inaccurate if systematic bias shifts all readings. Mean and bias help evaluate agreement with a reference, while range and standard deviation describe spread. Calibration can improve trueness, and repeated measurements help evaluate random variability. Fine resolution or many decimal places alone does not guarantee measurement quality.
Exam Facts
Exam Facts
- Accuracy concerns closeness to a reference value.
- Precision concerns agreement among repeated readings.
- High precision does not guarantee high accuracy.
- A small random spread indicates good precision.
- A systematic shift can produce precise but inaccurate readings.
- Trueness concerns the average result relative to a reference.
- Mean is calculated from repeated readings.
- Bias is mean minus reference value.
- Range is maximum reading minus minimum reading.
- Percentage error compares absolute error with the reference value.
- Resolution is different from precision.
- A single measurement cannot establish repeatability.
- The best measurements are both accurate and precise.
- Measurement results should be reported with appropriate uncertainty.
Exam Focus
Exam Focus
Important points
- Accuracy concerns closeness to a reference value.
- Precision concerns agreement among repeated readings.
- High precision does not guarantee high accuracy.
- A small random spread indicates good precision.
- A systematic shift can produce precise but inaccurate readings.
- Trueness concerns the average result relative to a reference.
- Mean is calculated from repeated readings.
- Bias is mean minus reference value.
- Range is maximum reading minus minimum reading.
- Percentage error compares absolute error with the reference value.
- Resolution is different from precision.
- A single measurement cannot establish repeatability.
- The best measurements are both accurate and precise.
- Measurement results should be reported with appropriate uncertainty.
Important formulas
- x̄ = x₁ + x₂ + ⋯ + xₙn
- e = xmeasured − xreference
- |e| = |xmeasured − xreference|
- Percentage error = (|e| / |x_reference|) × 100%
- b = x̄ − xreference
- R = xmax − xmin
- Δx = xmax − xmin2
Common exam traps
- Treating accuracy and precision as synonyms
- Calling close readings automatically accurate
- Judging precision from one reading
- Assuming more decimal places mean accuracy
- Using signed error in percentage-error magnitude
- Assuming repetition removes systematic bias
Long question structure
Every experimental measurement contains some uncertainty. Accuracy describes agreement between a measured value and a reference value. A smaller measurement error indicates better accuracy. Precision describes agreement among repeated measurement results. Closely grouped readings show good precision. Widely scattered readings show poor precision. Trueness describes agreement between the average reading and a reference value. Random effects generally increase the spread of readings. Therefore, random variation mainly reduces precision. Systematic effects can shift all readings in the same direction. Such readings may be precise but inaccurate. The mean is used to represent the central value of repeated readings. Bias is found by subtracting the reference value from the mean. Range or standard deviation can be used to describe spread. Calibration helps identify or reduce systematic bias. Repeated measurements help evaluate random variation. Fine instrument resolution does not automatically guarantee accuracy. A reliable experiment aims for both accurate and precise measurements.
Revision Zone
60-Second Revision
Quick cues from this topic — use them for last-minute revision.
Accuracy means closeness to a reference value.
Precision means agreement among repeated readings.
Trueness concerns the mean compared with the reference.
Error equals measured value minus reference value.
Accuracy describes how closely a measured value agrees with the true or accepted reference value.
x̄=Σxᵢn; e=xmeasured−xreference; |e|; % error=(|e|/|xref|)×100%; b=x̄−xref; R=xmax−xmin
- Definition
- Accuracy describes how closely a measured value agrees with the true or accepted reference value.
- Formula
- x̄=Σxᵢn; e=xmeasured−xreference; |e|; % error=(|e|/|xref|)×100%; b=x̄−xref; R=xmax−xmin
- Example
- 24.8 cm vs 25.0 cm → e=−0.2 cm, percentage error=0.8%; readings 10.50, 10.51, 10.49 are precise but inaccurate
- Critical warning
- Treating accuracy and precision as synonyms — They describe different properties
Did you understand this topic?
Your answer is saved with your learning record. It does not mark the topic as mastered.
Ready to test yourself?
You've completed the learning section.
Practice
Conceptual
Q1. Define accuracy.
Conceptual
Q2. Define precision.
Conceptual
Q3. Differentiate between precision and resolution.
Conceptual
Q4. Can measurements be precise but inaccurate? Explain.
Conceptual
Q5. Why are repeated measurements taken?
Conceptual
Q6. What is measurement trueness?
Conceptual
Q7. How does systematic error affect accuracy?
Conceptual
Q8. How does random variation affect precision?
Conceptual
Q9. Why do additional decimal places not guarantee accuracy?
Conceptual
Q10. How can calibration improve measurements?
Numerical
Q11. A reference length is 50.0 cm, and the measured length is 49.5 cm. Find the error and percentage error.
Numerical
Q12. Find the mean of 12.1, 12.2, 12.0 s.
Numerical
Q13. Find the range of 8.02, 8.05, 8.01 cm.
Numerical
Q14. A reference mass is 100.0 g, while the mean measured mass is 100.8 g. Find the bias.
Exam Practice
Important Short Question
Q1. Differentiate between accuracy and precision.
Long Question Preparation
Q2. Explain accuracy, precision, and their relationship with measurement errors.