Unit 1: Physical Quantities and Measurements

Lesson 1.4
Sources and Types of Error
Contents
Introduction 1

Learning Objectives 2

Warm Up 2

Learn about It! 3

Errors 3
Random Errors 4
Systematic Errors 6
Mistakes 7
Least Count 8
Error Bars 9
Methods in Reducing Errors 11

Key Points 13

Check Your Understanding 14

Challenge Yourself 16

Bibliography 17
Unit 1: Physical Quantities and Measurements

Lesson 1.4

Sources and Types of Error

Introduction
Have you encountered measuring the length of an object only to be confused when the
length varied by a millimeter or two when you measured it again? Experience has shown
that no measurement is free from uncertainty or errors. There are numerous factors that
can affect the result of your measurement—it can be a small offset in the zero mark of the
device or its incorrect calibration or the angle from which you read the measurement.
Identifying the uncertainty and evaluating its source is important to correct and to address
it. This is essential, especially in science, where the whole field is dependent on
measurement. What are the types of errors? How can the smallest divisions in a device
affect and contribute to it? In this lesson, we will discuss the types of errors in
measurement and explain the concept of the least count.

1.4. Sources and Types of Error 1

Unit 1: Physical Quantities and Measurements

Learning Objectives DepEd Competencies

● Differentiate random errors from
In this lesson, you should be able to do the systematic errors
(STEM_GP12EU-Ia-3).
following:
● Use the least count concept to
● Differentiate random errors from estimate errors associated with
single measurements
systematic errors.
(STEM_GP12EU-Ia-4).
● Identify the different sources of
errors.
● Use the least count concept to
estimate errors associated with
single measurements.
● Recognize the use of error bars.

● Identify techniques in reducing

errors in measurement.

Warm Up
Spot the Error! 10 minutes
In this activity, you will be identifying differences in measurements when certain
instruments or equipment are used.

Materials
● printed photos of measuring devices (e.g., thermometer, ruler, meter stick, digital
balance, and analog voltmeter/ammeter)
● blank sheet of paper

1.4. Sources and Types of Error 2

Unit 1: Physical Quantities and Measurements

Procedure
1. Divide the class into five groups.
2. Each group will receive a photo of a measuring device.
3. In a blank sheet of paper, list down all possible scenarios where the measured value
can deviate from your expected value if the assigned instrument is used. For
example, a measured value from a ruler may be different from the expected value if
the first 10 cm has been cut off.
4. After five minutes, exchange paper with other groups.
5. This time, identify whether the listed possible scenarios can be addressed or
corrected. A maximum of three minutes will be given for each group.
6. After the activity, all papers should be returned to the original group.
7. Representatives from each group should write their answers on the board for the
discussion.

Guide Questions
1. What are the possible factors that can affect the difference between the measured
and expected value?
2. Can all the listed scenarios be categorized into different types? What are your
suggested categories for these scenarios/factors?
3. Which scenarios can produce random fluctuations or variations in results? Which
scenarios will remain constant even if you repeat the measurements?

Learn about It!

Errors
As mentioned in the previous lesson, error is defined as the difference between the true
value and the measured value and in most sources, it is used interchangeably with
uncertainty. It is important to remember that in measurement, error is different from a
mistake or a blunder. An error signifies an inevitable uncertainty that is present in all types
of measurement. It cannot be completely eliminated, even if one is conducting an
experiment carefully. However, it is still the main goal to reduce the number of errors and to
estimate or quantify the error present.

1.4. Sources and Types of Error 3

Unit 1: Physical Quantities and Measurements

There are two main types of errors—systematic and random errors. Each type will be
discussed in detail below. We will also briefly discuss common mistakes that occur in
measurements and experiments.

What is the difference between a random error and

a systematic error?

Random Errors
Random errors occur when repeated measurements produce randomly different results.
This is a short-term variation of values around a mean value. It is caused by unknown and
unpredictable changes in the environment. The differences in the result cannot be predicted
and are only observed after repeated results. It is important to highlight that they can only
be discovered when the same quantity is measured in the same condition or setup. The
smaller the value of the random uncertainty, the more precise the measurement.

Random errors can be quantified and processed statistically using the calculations
introduced in the previous lesson, the arithmetic mean, and the standard deviation.
Arithmetic mean or average can provide an estimate of the measurement values. On the
other hand, the distribution or spread of the data with respect to the mean is defined by the
standard deviation.

The electrical noise in the circuit of an electrical instrument is a cause of random errors. It
appears regardless of the quality of the instrument used. Another example of a cause of
random errors is the irregular changes in the heat loss rate of a solar panel due to the
changes in the wind. The resolution of the instrument is another source of random error.
Each instrument has limited precision and can only resolve measurement differences based
on this limitation. For example, a meter stick cannot be used to distinguish distances smaller
than half of its smallest scale, which is 0.5 mm. A more precise instrument is needed to
accurately measure the said distance.

1.4. Sources and Types of Error 4

Unit 1: Physical Quantities and Measurements

What are examples of random errors?

Once the uncertainty was identified, it is also important to know the origin of the variation in
the values, for example, in an experiment. There are two types of scattering in experiments:
(1) technical, and (2) fundamental noise. Any apparatus has a fundamental noise, which is
affected by the laws of physics. However, it can also operate with a higher than average
noise known as technical noise. Technical noise can be reduced or lessened. For example,
in one of the physics experiments, you are required to launch a ball from a launcher. The
behavior of the ball as it moved from the launcher up to the air is governed by physical laws
and cannot be changed no matter how many times you repeat the experiment or the
measurement. Technical noise comes in when the data you collected vary based on how the
launcher recoils or from a slightly different angle by which you launch the ball. Perhaps
reloading the launcher with the same ball might load the spring at a different tension
affecting the result of the experiment.

Did You Know?

Johnson noise (Johnson-Nyquist noise or thermal noise) is the

random variation of voltage due to the thermal agitation of
electrons within an electrical conductor. It is an example of a
fundamental noise. It was first detected by John B. Johnson in 1926
and was later explained by Harry Nyquist. It is present in all
electrical circuits and can limit the sensitivity of electronic
equipment.

The vibration of the electrons is dependent on the temperature -

the higher the temperature, the higher the agitation, which results
in higher thermal noise. It is random in nature and is unpredictable.
It is the reason why it is not possible to reduce its effects even if the
components used are already of high quality.

1.4. Sources and Types of Error 5

Unit 1: Physical Quantities and Measurements

A good experimenter tries his or her best to reduce technical noise. However, if the
fundamental noise is the main reason why the data is scattered, it is suggested to change
the method of experiment or to change the measuring device.

Systematic Errors
A measurement error is categorized as systematic if it remains constant or changes in
regular fashion even if measurements are repeated. It causes the measured values to veer
away from the accepted or predicted value. If the shift is small, the measurement is
considered accurate. It can still be addressed by introducing corrections, but it is impossible
to remove completely and, in some instances, is challenging to detect.

A systematic error is related to a particular measuring instrument or an experimental

technique. For example, when you measure using a ruler where the first 5 cm is cut off,
even if you repeatedly measure the length of an object, you will still get the same error.
Another example is when you perform an experiment, and the environment needs a
background temperature of 20°C, but the room temperature is relatively higher. All these
values will vary the results away from the expected or true value.

Common sources of systematic errors are faulty calibrations of instruments, poorly

maintained instruments, or incorrect reading by the user. An example of a systematic error
is the parallax error. It occurs when the user reads an instrument at an angle that results in
a slightly lower or slightly higher measurement compared to the expected value, as shown
in Fig. 1.4.1. Zero offset or zero error occurs when there is a measured value, even if the
true reading at that specific time should be zero. It is very important to account for it in any
measuring device such as a ruler, a micrometer caliper, or an electronic balance. If possible,
re-zero the instrument or record the zero error and account for it in your final
measurements.

Unlike random errors, there are no statistical techniques that can quantify systematic
errors. The experimenter needs to decide the set of measurements to investigate the
source of systematic errors.

1.4. Sources and Types of Error 6

Unit 1: Physical Quantities and Measurements

Fig. 1.4.1. Parallax error occurs when the measuring device is viewed along different lines of
sights.

Mistakes
As mentioned in the introduction, mistakes are different from errors. However, it is still a
common occurrence in measurements and laboratory activities. Mistakes are similar in
nature to systematic errors and are also difficult to detect. There are also no statistical
techniques available to quantify it. Fortunately, it can be avoided and can be reduced by
being extra careful in any measurement activity. For example, writing 3.43 mm instead of
3.34 mm is a mistake, and if not corrected early, can affect the whole results. Mistakes, in
general, produce bad data points.

There are well-known mistakes that can affect the precision and accuracy of experimental
results and sometimes can result in potentially harmful consequences. When scales are
misread in an analog device, especially if there are two scales near each other with different
gradations. Malfunction of the apparatus is also another source of mistake and might be
difficult to detect. The presence of erroneous data can sometimes be detected only when
graphed. Confusion of units is also a mistake, such as what happened in the 1999 NASA
Mars Climate Orbiter. Its mission was unsuccessful because of the failure to convert English
to metric units, which resulted in its navigation error. Air Canada’s Boeing 767 (Gimli Glider)
aircraft also ran out of fuel mid-flight due to the misunderstanding between metric and
imperial units of volume.

1.4. Sources and Types of Error 7

Unit 1: Physical Quantities and Measurements

How can the least count concept be used to

estimate the error in a measurement?

Least Count
The least count is the smallest division found on an instrument. If you need a more precise
measurement, use a measuring device with finer gradations. For example, a meter stick with
gradations in millimeters has a least count of 1 mm. A micrometer, which is a device that
can precisely measure thicknesses and short lengths, has a least count of 0.01 mm. A digital
stopwatch may have a least count of 0.01 s. Least count is essential in estimating the errors
associated with the single measurements.

The instrument limit of error is generally taken to be the least count or some fraction of
the least count. There are rules of thumb available, but there is no specific rule whether to
choose the least count, half the least count, or a fraction of the least count. This depends on
the instrument you are using, the scale division in the instrument being used, and your
judgment as an experimenter. If the scale division is significantly large, you can be
comfortable in estimating 1/5 or 1/10 of the least count. If the scale divisions are slightly
smaller, you might want to only estimate ½ of the least count. Lastly, if the division is much
smaller, then you can only estimate up to the least count.

Let us take the following as an example. Suppose that you are required to measure the
length of the rod using the ruler below in cm. What is the length of the rod?

Fig. 1.4.2. Measuring the length of the rod using a ruler with coarse gradations.

1.4. Sources and Types of Error 8

Unit 1: Physical Quantities and Measurements

The least count of the instrument is 1 cm. Based on Fig. 1.4.2, you are certain that the rod is
between 3 cm and 4 cm. Let us say that we estimate it as 3.4 cm since it is already close to
the midpoint. However, to increase the precision of the measurement, we can use a ruler
that has finer gradations, as shown below.

Fig. 1.4.3. Measuring the length of the rod using a ruler with finer gradations.

In Fig. 1.4.3, the ruler has a least count of 0.1 cm. This time, we can be certain that the
length of the rod lies between 3.3 cm and 3.4 cm. From the previous lesson, we mentioned
that as a rule of thumb, the highest precision achievable in an analog device is half a division
or half the least count. Applying it in this example, we can have an estimate of 3.35 cm,
while half the division is 0.05 cm. Therefore, the length of the rod can be reported as 3.35
cm ± 0.05 cm. It means that the true length of the rod lies between 3.30 cm to 3.40 cm.

Remember
There is no simple rule in estimating the error of any measurement
in an analog device. There are instances when you have to estimate
it using your best judgment, especially if the scale divisions are
significantly large. For digital devices, always check the
manufacturer’s specifications for its uncertainty.

Error Bars
Error bars are graphical representations that indicate the range by which each point
probably lies in the graph. It is a line passing through a data point that extends from the
smallest value up to the largest value possible for that specific data point. It shows the
uncertainty of the values as what was displayed in Fig. 1.4.4. Measurement A is 15 ± 1 ohms,
while measurement B is 25 ± 2 ohms. An error bar on the vertical axis shows the uncertainty
of the value in that axis, while an error bar in the horizontal axis determines the range of
uncertainty in the said axis.

1.4. Sources and Types of Error 9

Unit 1: Physical Quantities and Measurements

Fig. 1.4.4. Two measurements of resistance plotted to show the range of uncertainty. Each
measurement is characterized by the best estimate (black dot) and a range of probable
values (vertical error bar).

Many physical laws require that one quantity is proportional to another. One example is
Hooke’s law, which states that the extension of the spring is proportional to the force
stretching it. Newton’s law is another example, which requires that the acceleration of a
body is proportional to the applied force. The relationship between these quantities can be
presented in a graph.

Suppose you conducted an experiment and measured the quantities x and y. If a quantity y
is directly proportional to quantity x, a graph of y vs x will produce a straight line that passes
through the origin, as shown in Fig. 1.4.5 (a). Therefore, to test the proportionality between
y and x, plot all the points in the graph and observe if all the points lie in a straight line
through the origin.

Error bars are included to indicate the uncertainties for each data point, as shown in Fig.
1.4.5 (b). If, for example, the best straight line in a graph misses a high proportion of the
error bars or it misses any by a significant amount, compared to the length of error bar,
then that specific data point may be problematic or inconsistent. An example is shown in
Fig. 1.4.5 (c). Once this is observed, there is a need to recheck the measurement or the

1.4. Sources and Types of Error 10

Unit 1: Physical Quantities and Measurements

uncertainties. This can also be used as a reference, whether it is really an incorrect

measurement or there is a scientific principle behind why that specific point behaves as
such.

If there is a measurement with uncertainties for both variables, then crosses are seen in a
graph to illustrate the error bars for each variable, as shown in Fig. 1.4.6. It is important to
note that error bars are not limited to graphs with straight lines but are also found in other
graphs such as parabola. However, it is a bit challenging to judge whether a set of points fit
a parabola or any other curve, in that case.

What is the significance of using error bars?

Methods in Reducing Errors

While there are errors that cannot be avoided, there are techniques that can be used to
reduce the error in experiments.
1. Mass is commonly measured in several experiments and it is very essential to reduce
the errors when measuring it. Systematic error in electronic balances during
laboratory activities can be reduced by weighing masses by difference. Begin by
measuring the mass of both the material and the container holding it. Then, transfer
the material to another container. Next, measure the mass of the container. Lastly,
calculate the mass of the removed sample by getting the difference between the
initial and the final weights of the original container.
2. Random errors can be reduced by getting the average of the results. This is to
ensure that any error that falls below or slightly above the actual value can be
accounted for. The more trials, the more accurate the average result is.
3. Systematic error can be reduced by calibrating the equipment. This requires the
comparison between a standard device with a known accuracy with a device that
needs calibration.

1.4. Sources and Types of Error 11

Unit 1: Physical Quantities and Measurements

Fig. 1.4.5. Example of graphs showing the relationship between the length of the spring x
and the mass of the load m: (a) A graph without any error bars; (b) A similar graph with an
error bar to show the uncertainties in x while the uncertainties in m was considered
negligible; (c) A graph showing a different set of data displaying an inconsistent relationship
between the variables.

What are some techniques that can be used to

reduce errors in measurement?

1.4. Sources and Types of Error 12

Unit 1: Physical Quantities and Measurements

Fig. 1.4.6. Crosses are used to indicate uncertainties for both variables in the vertical and
horizontal axes.

Key Points
___________________________________________________________________________________________

● Error is defined as the difference between the true value and the measured value.
● Random errors occur when repeated measurements produce varying results. It
influences the precision of the result.
● Errors are categorized as systematic if it remains constant or changes in a regular
fashion even if measurements are repeated. It influences the accuracy of the result.
● The least count is the smallest division found in a measuring device.
● Error bars are graphical representations indicating the range of uncertainty of a
specific value or data.
● Errors cannot be 100% eliminated in measurement but there are techniques that
can be used to lessen its occurrence.
___________________________________________________________________________________________

1.4. Sources and Types of Error 13

Unit 1: Physical Quantities and Measurements

Check Your Understanding

A. Fill in the missing information in each statement.

1. Error is defined as the difference between the __________ and the __________.
2. __________ is a short-term variation of values around a mean value and influences
the accuracy of the result.
3. Electrical noise in a circuit is an example of a __________ error.
4. Errors that remain constant or changes in a regular manner even if repeated
measurements are conducted is called __________.
5. __________ occurs when a user reads an instrument at a different angle or line of
sight.
6. Zero error is an example of __________ error.
7. A common source of systematic error is __________.
8. Incorrect conversion from one unit to another is an example of a __________.
9. Least count is the __________ division found on a measuring device.
10. __________ is used to represent the range of uncertainty in the data points when
plotted on a graph.

B. Write true if the statement is correct. Otherwise, write false.

_________________ 1. The use of calibrated instruments can reduce random

errors.
_________________ 2. Limited precision of an instrument can be a source of
systematic error.
_________________ 3. “Human error” or mistake is not a source of experimental
error.

_________________ 4. A meterstick with gradations in centimeter has a least

count of 1 cm.
5. A larger error bar indicates large uncertainty.
_________________

1.4. Sources and Types of Error 14

Unit 1: Physical Quantities and Measurements

C. Read the following measurements along with their corresponding

errors using the following format: best estimate ±
error/uncertainty. Items 1-3 are in centimeters, while items 4-5 are
in volts.

1.

2.

3.

4.

5.

Challenge Yourself

A. Answer the questions briefly while referring to the accompanying

figures.

1.4. Sources and Types of Error 15

Unit 1: Physical Quantities and Measurements

1. How will you describe the measured value if an error bar is observed below?

2. How will you read the voltmeter with a reading below? Defend your answer.

3. Data from an experiment were plotted in a graph and error bars were included to
provide the uncertainty of each data point. You are expecting that the data points
will lie in a straight diagonal line from the origin. Will it agree with your
expectations? What can you conclude about the set of data points shown in the
graph below?

1.4. Sources and Types of Error 16

Unit 1: Physical Quantities and Measurements

B. Answer the questions in 2-3 sentences only.

1. Why are mistakes or blunders not categorized as either random or systematic

errors?
2. Explain how changes in the environment can result in both systematic and random
errors.

Bibliography
Bell, Stephanie. A Beginner’s Guide to Uncertainty of Measurement (Issue 2). UK: Crown House,
2001.

Hughes, Ifan G. and Thomas P.A. Hase. Measurements and their Uncertainties: A Practical
Guide to Modern Error Analysis. UK: Oxford University Press, 2010.

Kirkup, L. and R. B. Frenkel. An Introduction to Uncertainty in Measurement. UK: Cambridge

University Press, 2006.

Taylor, John R. An Introduction to Error Analysis: The Study of Uncertainties in Physical

Measurements (2nd Edition). USA: University Science Books, 1997.

Young, Hugh D., Roger A. Freedman, and A. Lewis Ford. Sears and Zemansky’s University
Physics with Modern Physics (13th ed). USA: Pearson Education, 2012.

1.4. Sources and Types of Error 17