National University of Science and

Technology

Lab Report

Error Analysis Lab:

1. Measuring the Length and Width of a rectangular wooden slab, and

calculating the Perimeter, Area and the Error associated with these
measurements.

2. Measuring the Time Period of an oscillating bob, and calculating the

value of ‘g’ and the error associated with it

SCME ME-12
1.Shah Saud 2.Muhammad Arqam Karim 3.Muhammad
Hasham Hussain 4.Muhammad Adil 5.Muhammad Arsalan
Shakoor
GROUP-B1
I. Objectives
When making any measurement it is inevitable to not conclude to some degree of
uncertainties and errors in the recorded value. Whether measuring the radius of a
minute atom, or taking time measurements for oscillations, they are accompanies with
some degree of uncertainty. No measurement can be said to be able to achieve
absolute accuracy. Hence, it is important for engineering students like ourselves to
learn about different methodologies of calculating and propagating errors and
uncertainties in measurements. Hence, this lab should make us able to:

1. Identify the equipment used to make measurements: Length, Width, Time Period;
know their least counts and the units in which measurements are being recorded.

2. Tabulate and formulate tables for the recorded measurements with given
uncertainties

3. Derive further values: Perimeter, Area, value of ‘g’ Constant through operations on
the measurements and also use tools such as Standard Deviation, Standard Error &
Uncertainty Propagation to calculate the errors associated with the derived/desired
values.

4. Learning and understanding how the concepts of precision and accuracy work in
real life while recording different measurements

5. To be able to identify dominant errors and hence provide a guide as to where more
effort is needed to improve an experiment and therefore improving the results of an
experiment.

II. Apparatus

1. Stop Watch
2. Meter Rod
3. Metal Bob and String (for Pendulum)
4. Metal Stand

III. Theory
Error analysis is basically the study and evaluation of uncertainty in measurement. In
all types of measurements, you have to face a lot of errors. Everyone makes
mistakes! But experimental error does not mean mistake. Even though you don’t
make any mistake in experiment, your result still has experimental error which
actually means uncertainty. No measurement is exactly correct. Experimental error is
inevitable!

P a g e 1 | 10
Types of errors:
Measurement errors can be classified into different types, depending on how the
measurement was obtained. Some common errors in measurements are:
1-Systematic errors:
• Errors come due to un calibrated equipment and faulty techniques.
• It is difficult to detect systematic errors but can be eliminated by proper
calibration of equipment.
2-Random errors:
• Errors come due to inability of person.
• Random errors can be determined and minimized by repeated measurements

Precision and accuracy in measurement

Accuracy:
• It is the agreement of how much measured value is closed to actual value.
• Error is amount of inaccuracy in a measurement.
• Accuracy is often reported by using relative error
𝒎𝒆𝒂𝒔𝒖𝒓𝒆−𝒆𝒙𝒑𝒆𝒄𝒕𝒆𝒅
• Relative error =
𝒆𝒙𝒑𝒆𝒄𝒕𝒆𝒅
Precision:
• Precision of measurement is related to measuring instrument
• Smaller is the least count of instrument, more precise in the measurement
• Precision is often reported by using relative or fractional uncertainty
𝒖𝒏𝒄𝒆𝒓𝒕𝒂𝒊𝒏𝒕𝒚
• Relative uncertainty =
𝒎𝒆𝒂𝒔𝒖𝒓𝒆𝒅 𝒗𝒂𝒍𝒖𝒆

Calculation of uncertainty for single measurement:

The uncertainty of single measurement is limited by the precision and accuracy of
measuring instrument and some other factors that might affect the ability of
experimenter to make the measurement

Calculation of uncertainty in repeated measurements:

In order to find uncertainty in repeated measurements, following parameters are
concerned
Average value:
• It is the mean of all the N independent measurements.
𝑠𝑢𝑚 𝑜𝑓 𝑎𝑙𝑙 𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑚𝑒𝑛𝑡𝑠 𝑥1 +𝑥2 +𝑥3 +⋯𝑥𝑁
• Mean = =
𝑛𝑜 𝑜𝑓 𝑚𝑒𝑎𝑠𝑢𝑙𝑒𝑚𝑒𝑛𝑡𝑠 𝑁

Average deviation:
• It is the deviation of every measurement from mean value.
• It tells us on average that how much individual measurement vary from mean
value
|𝑋1− 𝑋̅ |+|𝑋2 −𝑋̅|+⋯+|𝑋𝑛 −𝑋̅|
• Avg deviation =
𝑁

P a g e 2 | 10
Standard deviation:
• Standard deviation is a number used to tell how measurements for a group
are spread out from the average
• A low standard deviation means that most of the numbers are close to the
average.
• A high standard deviation means that the numbers are more spread out.
∑ 𝛿𝑥𝑖2
• Mathematically: s = √
(𝑁−1)

Standard error:
• The standard error is the approximate standard deviation of a statistical
sample population
𝑆
• SE =
√𝑁

Significant figures:
Our experiment involves the measurements of several physical quantities such as
length and time.
The values of these quantities should be presented in term of significant figures
Following rules dictates to the handling of significant figures:
Specify the measured value to be same accuracy as error.
When adding or subtracting, the answer is only good to the least number present
When multiplying or dividing, keep the same number of significant figures as the
factor with fewest numbers of significant figures.

Propagation of Uncertainty:
The simple "significant figure" rules referred to earlier are very approximate attempts
to account for uncertainties in derived quantities. Sometimes a tiny uncertainty in a
measurement will produce a huge uncertainty in a derived quantity—the exponential
function is notorious for this. Clearly, we need a more general approach for
propagating uncertainties. If one assumes that the results of many duplicate
measurements would produce a Normal or Gaussian distribution about the mean,
then statistical theory provides a mechanism for estimating the uncertainty sC in the
derived quantity C = ƒ(A,B). When the uncertainties δA and δB are both small and
uncorrelated, statistical arguments show that the propagated uncertainty δC is given
by

Error Propagation Formulae:

A and B are measurements with associated uncertainties 𝛿𝐴 and 𝛿𝐵, respectively.
C is a derived quantity with associated uncertainty 𝛿𝐶.
1. Addition of an exact (constant) number 𝛽: 𝐶 = 𝛽𝐴 … . 𝛿𝐶 = 𝛿𝐴
2. Multiplication by an exact number 𝛽: 𝐶 = 𝛽𝐴 … . 𝛿𝐶 = 𝛽𝛿𝐴
3. Addition (or subtraction): C = A ± B ± D ± ...𝛿𝐶 = √𝛿𝐴2 + 𝛿𝐵 2 + 𝛿𝐷2
𝐴×𝐵… 𝛿𝐴 2 𝛿𝐵 2 𝛿𝐷 2
4. Multiplication and division: 𝐶 = 𝐷×𝐸… … 𝛿𝐶 = 𝐶 √( 𝐴 ) + ( 𝐵 ) + ( 𝐷 ) + ⋯

P a g e 3 | 10
IV. Data Analysis

Experiment 1: Measuring the Length and Width of a rectangular

wooden slab, and calculating the Perimeter, Area and the Error
associated with these measurements.

1. Procedure of taking Length and Width measurements:

i. Values were recorded by each member of the group member 3 times,

ii. The Values were recorded by placing the meter rod at different positions on the
slab while measuring the Length and the Width.

iii. To minimize error due to rough edges of meter rod, measurements were taken
from 1cm mark, and then 1cm was deducted from each recorded value.

Exp# Student 1 Student 2 Student 3 Student 4 Student 5

𝐋𝟏 𝐖𝟏 𝐋𝟐 𝐖𝟐 𝐋𝟑 𝐖𝟑 𝐋𝟒 𝐖𝟒 𝐋𝟓 𝐖𝟓
1 91.1 31.6 91.8 31.4 91.7 31.5 91.4 30.6 91.4 31.5
2 91.8 31.5 91.7 31.3 91.5 31.5 91.6 31.3 91.7 31.4
3 92.0 31.5 91.6 31.5 91.6 31.4 91.5 31.2 92.0 31.3
AVG 91.63 31.53 91.7 31.4 91.6 31.47 91.5 31.03 91.7 31.4

Table 1.0: Length (L) & Width (W) measurements in cm by 5 group members of G1 repeated 3 times by each
team member (least count of meter rod: ±0.05cm)

2. Combining data to find the mean value

𝐿1 + 𝐿2 + 𝐿3 + 𝐿4 + 𝐿5
̅=
Length, 𝐿
5
91.63 + 91.7 + 91.6 + 91.5 + 91.7
=
5
𝐿̅ = 91.626 𝑐𝑚 ≈ 91.6 𝑐𝑚

𝑊1 +𝑊2 +𝑊3 +𝑊3 +𝑊4 +𝑊4

̅ =
Width, 𝑊
5
31.53 + 31.4 + 31.47 + 31.03 + 31.4
=
5
̅
𝑊 = 31.32 𝑐𝑚 ≈ 31.3 𝑐𝑚

P a g e 4 | 10
 3. Standard Deviation and Standard Error Calculation for Error Propagation

Standard Deviation, Length:

∑𝑛=5
𝑖=1 (𝑥̅ − 𝑥𝑖 )
2
𝑠𝐿 = ±√
𝑛−1

(91.6 − 91.63)2 + (91.6 − 91.7)2 + (91.6 − 91.6)2 + (91.6 − 91.5)2 + (91.6 − 91.7)2
𝑠𝐿 = ±√
5−1

𝑠𝐿 = ±0.088 𝑐𝑚 ≈ ±0.09 𝑐𝑚

Standard Error, Length:

𝑠𝐿
𝜎𝐿 =
√𝑛
0.09
𝜎𝐿 =
√5
𝛿𝐿 = 𝜎𝐿 = ±0.039 𝑐𝑚 ≈ ±0.04 𝑐𝑚 = ±0.0004 𝑚

Standard Deviation, Width:

(31.3 − 31.53)2 + (31.3 − 31.4)2 + (31.3 − 31.47)2 + (31.3 − 31.03)2 + (31.3 − 31.4)2
𝑠𝑊 = ±√
5−1

𝑠𝑊 = ±0.21 𝑐𝑚 ≈ ±0.2 𝑐𝑚

Standard Error, Width:

0.2
𝜎𝑊 =
√5
𝛿𝑊 = 𝜎𝑊 = ±0.089 𝑐𝑚 ≈ ±0.09 𝑐𝑚 = ±0.0009 𝑚

Perimeter:
𝑃 = (2 × 𝐿̅) + (2 × W
̅)

= (2 × 91.6) + (2 × 31.3)
𝑃 = 245.8 𝑐𝑚 = 245.8 × 10−2 𝑚 = 2.458 𝑚 ≈ 2.46 𝑚

P a g e 5 | 10
Uncertainty Propagation, Perimeter

𝛿𝑃 = ±√(2 × 𝛿𝐿)2 + (2 × 𝛿𝑊)2

= ±√(2 × 0.0004)2 + (2 × 0.0009)2

= ±√(0.0008)2 + (0.0018)2
𝛿𝑃 = ±0.00197 𝑚 ≈ ±0.002 𝑚

Hence, the perimeter with error is:

𝑷 ± 𝜹𝑷 = 𝟐. 𝟒𝟔 𝒎 ± 𝟎. 𝟎𝟎𝟐 𝒎

Area:
𝐴 = 𝐿̅ × 𝑊
̅

𝐴 = 91.6 × 31.3
𝐴 = 2867.08 𝑐𝑚2 ≈ 2867.1 𝑐𝑚2 = 2867.1 × 10−4 𝑚2 = 0.28671 𝑚2

Uncertainty Propagation, Area:

2 2
𝐿̅ ̅
𝑊
𝛿𝐴 = ±√( . 𝛿𝐿. 𝛿𝑊) + ( . 𝛿𝑊. 𝛿𝐿)
𝛿𝐿 𝛿𝑊

= ±√(𝐿̅. 𝛿𝑊)2 + (𝑊
̅ . 𝛿𝐿)2

= ±√(91.6 × 0.0009)2 + (31.3 × 0.0004)2

𝛿𝐴 = ±0.0833 𝑐𝑚2 = ±0.08 𝑐𝑚2 = ±0.08 × 10−4 𝑚 = ±0.000008 𝑚2 ≈ 0.00001 𝑚2

Hence, the Area with error is:

𝑨 ± 𝜹𝑨 = 𝟎. 𝟐𝟖𝟔𝟕𝟏 𝒎𝟐 ± 𝟎. 𝟎𝟎𝟎𝟎𝟏 𝒎𝟐

P a g e 6 | 10
 Experiment 2: Measuring the Time Period of an oscillating bob, and
calculating the value of ‘g’ and the error associated with it.

1. Procedure:
i. Firstly. the length of the string (using meter rod) and Diameter of the bob (using
Vernier callipers) were measured.

ii. To reduce error multiple readings were recorded and averaged,

iii. After forming a pendulum by attaching the string to the bob and tying it to the
metal stand,

iv. Time period was recorded by each team member twice and error was recorded to
be the least count of the stop watch.

v. Each turn was recorded by oscillating the pendulum T=10 times and then the
recorded time was divided by 10 to record 1 value for Time period (this was done to
reduce error in individual reading)

2. Data

Exp# Student 1 Student 2 Student 3

String, 𝒍𝒔 Bob, 𝒓 String, 𝒍𝒔 Bob, 𝒓 String, 𝒍𝒔 Bob, 𝒓
1 39.6 0.627 39.8 0.629 39.7 0.628
𝒍𝑷 40.227 40.429 40.328
Table: Measurements for the radius of the bob, r and length of the string, 𝑙𝑠 ; to calculate the length of
the pendulum, 𝑙𝑃 . All measurements are in cm

Mean Value, Pendulum length, 𝒍𝑷 :

40.227 + 40.429 + 40.328
𝒍̅𝑷 =
3

𝒍̅𝑷 = 40.32 𝑐𝑚 = 40.32 × 10−2 𝑚 = 0.4032 𝑚

Standard deviation, Pendulum Length, 𝒔𝒍𝑷 :

(0.4032 − 0.4023)2 + (0.4032 − 0.4043)2 + (0.4032 − 0.4033)2
𝑠𝑙𝑃 = ±√
3−1

𝑠𝑙𝑃 = ±0.001 𝑚

Standard Error, Pendulum Length, 𝝈𝒍𝑷 :

0.001
𝜎𝑙𝑃 =
√3

𝜎𝑙𝑃 = ±0.0006 𝑚

P a g e 7 | 10
Time Period (10 Oscillations) Measurements Table:

Exp# Student 1 Student 2 Student 3

̅𝟏𝟎,𝟏
𝑻 13.10 13.00 12.89
̅𝟏𝟎,𝟐
𝑻 12.86 12.80 13.05
Avg 12.98 12.90 12.97
𝑻 1.298 1.290 1.297
Table: Measurements and Average of time period,
𝑇10 calculated by 3 group members, all readings are in s.
Least count: ±0.01s

̅
Mean Value, Time Period, 𝑇
1.298 + 1.290 + 1.297
𝑇̅ =
3

𝑇̅ = 1.295 𝑠

Standard Deviation, Time Period, 𝒔𝑻 :

(1.295 − 1.298)2 + (1.295 − 1.290)2 + (1.295 − 1.297)2

𝑠 𝑇 = ±√
3−1

(−0.003)2 + (0.005)2 + (−0.002)2

= ±√
2

𝑠𝑇 = ±0.0044 𝑠

Standard Error, Time Period, 𝝈𝑻

0.0044
𝜎𝑇 =
√3

𝜎𝑇 = ±0.0025 𝑠 ≈ ±0.003

𝑻 ± 𝜹𝑻 = 𝟏. 𝟐𝟗𝟓 ± 𝟎. 𝟎𝟎𝟑 𝒔

P a g e 8 | 10
Value of ‘g’:
4𝜋 2 𝑙
𝑔=
𝑇2

4𝜋 2 (0.4032)
=
(1.295)2

𝑔 = 9.49 𝑚/𝑠 2

Uncertainty Propagation in Value of ‘g’:

𝛿𝑙𝑃 2𝑙𝛿𝑇
𝛿𝑔 = 4𝜋 2 ( 2 − 3 )
𝑇 𝑇

𝛿𝑔 𝛿𝑙𝑝 2𝛿𝑇
= −
𝑔 𝑙𝑝 𝑇

𝛿𝑙𝑝 2𝛿𝑇
𝛿𝑔 = 𝑔 ( − )
𝑙𝑝 𝑇

0.0006 2(0.003)
𝛿𝑔 = 9.49 ( − )
0.4032 1.295

𝛿𝑔 = ±0.03

Acceleration due to gravity:

𝒈 ± 𝜹𝒈 = 𝟗. 𝟒𝟗 ± 𝟎. 𝟎𝟑 𝒎/𝒔𝟐

V. Result

Experiment 1:
Perimeter: 2.46 𝑚 ± 0.002 𝑚
0.002
Relative Uncertainty: × 100 = ±0.082%
2.46

Area: 0.28671 ± 0.000008 𝑚2

0.00001
Relative Uncertainty: × 100 = ±3.5%
0.28671

P a g e 9 | 10
 Experiment 2:
Pendulum Length: 0.4032 ± 0.0006 𝑚
0.0006
Relative Uncertainty: × 100 = ±0.15%
0.4032

Time Period: 1.295 ± 0.003 𝑠

0.003
Relative Uncertainty: × 100 = ±0.23%
1.295

Acceleration due to gravity, g: 9.49 ± 0.03 𝑚/𝑠 2

0.03
Relative Uncertainty: × 100 = 0.3%
9.49

VI. Conclusion
In the theory of this lab report we mentioned different types of errors i.e. systematic,
random, human, response. That can generate uncertainty in the measurements.
This lab was a simulation to take measurements and calcite the uncertainty
associated with those measurements. Which tells us that every measurement,
regardless of how carefully gauged will be inevitably accompanied with a certain
degree of uncertainty.

This uncertainty can be represented in the form of Relative Uncertainty which also
indicates how precise the measurement is, or in other words how small/large the
error in the measurements has been.

We learned, to not only take a hands-on approach to theory that we had previously
studied in lectures, but also to apply the calculations of standard deviation and error
in real world applications.

These specific experiments enabled us to measure different measurements for

various types of scenarios with their prospective errors and uncertainties. Using
Standard Deviation, Standard Error and Uncertainty and studying the effect of errors
in a quantity, we drew from this experiment that by using different for measuring the
same quantity can affect the error, depending on the instruments least count. For
example, we measured length for both the diameter of the bob with Vernier calliper
and length of the string with meter rod, but because meter rod is less precise, there
was a greater error in the measurement of the string. Also, we learned how precision
of the measurements of length and width magnifies in the derived quantity. We also
learned to measure acceleration due to gravity g just by recording Time Period of
Oscillations and the length of a simple pendulum.

P a g e 10 | 10