DP IB Physics: SL Your notes

Processing Uncertainties
Contents
Random & Systematic Errors
Calculating Uncertainties
Determining Uncertainties from Graphs

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 1
 Random & Systematic Errors
Your notes
Random & Systematic Errors
Measurements of quantities are made with the aim of finding the true value of that
quantity
In reality, it is impossible to obtain the true value of any quantity as there will always
be a degree of uncertainty
The uncertainty is an estimate of the difference between a measurement reading and
the true value
The two types of measurement errors that lead to uncertainty are:
Random errors
Systematic errors

Random Errors
Random errors cause unpredictable fluctuations in an instrument’s readings as a result
of uncontrollable factors, such as environmental conditions
This affects the precision of the measurements taken, causing a wider spread of results
about the mean value
To reduce random error:
Repeat measurements several times and calculate an average from them

Reading Errors
When measuring a quantity using an analogue device such as a ruler, the uncertainty in
that measured quantity is ±0.5 the smallest measuring interval
When measuring a quantity using a digital device such as a digital scale or stopwatch,
the uncertainty in that measured quantity is ±1 the smallest measuring interval
To reduce reading errors:
Use a more precise device with smaller measuring intervals and therefore less
uncertainty

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 2
Both rulers measure the same candy cane, yet Ruler B is more precise than Ruler A due to a
smaller interval size
Your notes
Systematic Errors
Systematic errors arise from the use of faulty instruments or from flaws in the
experimental method
This type of error is repeated consistently every time the instrument is used or the
method is followed, which affects the accuracy of all readings obtained
To reduce systematic errors:
Instruments should be recalibrated, or different instruments should be used
Corrections or adjustments should be made to the technique

Systematic errors on graphs are shown by the offset of the line from the origin

Zero Errors
This is a type of systematic error which occurs when an instrument gives a reading when
the true reading is zero
For example, a top-ban balance that starts at 2 g instead of 0 g
To account for zero errors
Take the difference of the offset from each value
For example, if a scale starts at 2 g instead of 0 g, a measurement of 50 g would
actually be 50 – 2 = 48 g
The offset could be positive or negative

Precision
Precise measurements are ones in which there is very little spread about the mean
value, in other words, how close the measured values are to each other

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 3
 If a measurement is repeated several times, it can be described as precise when the
values are very similar to, or the same as, each other
Your notes
Another way to describe this concept is if the random uncertainty of a
measurement is small, then that measurement can be said to be precise
The precision of a measurement is reflected in the values recorded – measurements to a
greater number of decimal places are said to be more precise than those to a whole
number
Accuracy
A measurement is considered accurate if it is close to the true value
Another way to describe this concept is if the systematic error of a measurement is
small, then that measurement can be said to be accurate
The accuracy can be increased by repeating measurements and finding a mean of the
results
Repeating measurements also helps to identify anomalies that can be omitted from the
final results

The difference between precise and accurate results

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 4
 Representing precision and accuracy on a graph

Reliability Your notes

Reliability is defined as
A measure of the ability of an experimental procedure to produce the expected
results when using the same method and equipment
A reliable experiment is one which produces consistent results when repeated many
times
Similarly, a reliable measurement is one which can be reproduced consistently when
measured repeatedly
When thinking about the reliability of an experiment, a good question to ask is
Would similar conclusions be reached if someone repeated this experiment?

Validity
The validity of an experiment relates to the experimental method and the appropriate
choice of variables
Validity is defined as
A measure of the suitability of an experimental procedure to measure what it is
intended to measure
It is essential that any variables that may affect the outcome of an experiment are
identified and controlled in order for the results to be valid
For example, when using Charles’ law to determine absolute zero, pressure must be kept
constant
When thinking about the validity of an experiment, a good question to ask is
How relevant is this experiment to my original research question?

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 5
 Calculating Uncertainties
Your notes
Calculating Uncertainties
There is always a degree of uncertainty when measurements are taken; the uncertainty
can be thought of as the difference between the actual reading taken (caused by the
equipment or techniques used) and the true value
Uncertainties are not the same as errors
Errors can be thought of as issues with equipment or methodology that cause a
reading to be different from the true value
The uncertainty is a range of values around a measurement within which the true
value is expected to lie, and is an estimate
For example, if the true value of the mass of a box is 950 g, but a systematic error with a
balance gives an actual reading of 952 g, the uncertainty is ±2 g
These uncertainties can be represented in a number of ways:
Absolute Uncertainty: where uncertainty is given as a fixed quantity
Fractional Uncertainty: where uncertainty is given as a fraction of the measurement
Percentage Uncertainty: where uncertainty is given as a percentage of the
measurement

To find uncertainties in different situations:

The uncertainty in a reading: ± half the smallest division
The uncertainty in a measurement: at least ±1 smallest division
The uncertainty in repeated data: half the range i.e. ± ½ (largest - smallest value)
The uncertainty in digital readings: ± the last significant digit unless otherwise
quoted
The uncertainty in the natural log of a value: absolute uncertainty in ln(x) =
uncertainty in x
x

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 6
 Your notes

How to calculate absolute, fractional and percentage uncertainty

Always make sure your absolute or percentage uncertainty is to the same number of
significant figures as the reading

Combining Uncertainties
When combining uncertainties, the rules are as follows:

Operation Example Propagation Rule

Addition & y = a ±b Δy = Δa + Δb
Subtraction
The sum of the absolute uncertainties

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 7
 Multiplication & y = a × b or Δy Δa Δb
Division a y
= a
+ b Your notes
y=
b The sum of the fractional uncertainties

Power y = a ±n Δy Δa
y
=n ⎛
⎜
⎜
a
⎞
⎟
⎟

⎝ ⎠

The magnitude of n times the fractional

uncertainty

Adding / Subtracting Data

Add together the absolute uncertainties

Multiplying / Dividing Data

Add the percentage or fractional uncertainties

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 8
 Your notes

Raising to a Power
Multiply the percentage uncertainty by the power

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 9
 Your notes

Examiner Tips and Tricks

Remember:
Absolute uncertainties (denoted by Δ) have the same units as the quantity
Percentage uncertainties have no units
The uncertainty in constants, such as π, is taken to be zero
Uncertainties in trigonometric and logarithmic functions will not be tested in the
exam, so just remember these rules and you’ll be fine!

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 10
 Determining Uncertainties from Graphs
Your notes
Determining Uncertainties from Graphs
The uncertainty in a measurement can be shown on a graph as an error bar
This bar is drawn above and below the point (or from side to side) and shows
the uncertainty in that measurement
Error bars are plotted on graphs to show the absolute uncertainty of values plotted

Representing error bars on a graph

To calculate the uncertainty in a gradient, two lines of best fit should be drawn on the
graph:
The ‘best’ line of best fit, which passes as close to the points as possible
The ‘worst’ line of best fit, either the steepest possible or the shallowest possible
line which fits within all the error bars

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 11
 Your notes

The line of best fit passes as close as possible to all the points. The steepest and
shallowest lines are known as the worst fit
The percentage uncertainty in the gradient can be found using the magnitude of the
'best' and 'worst' gradients:

best gradient − worst gradient

percentage uncertainty =
best gradient
× 100%
Either the steepest or shallowest line of best fit may have the 'worst' gradient on a case-
by-case basis.
The 'worst' gradient will be the one with the greatest difference in magnitude from
the 'best' line of best fit.
The equation above is for the case where the 'worst' gradient is the shallowest.
If the 'worst' gradient is the steepest, then the 'worst' gradient should be
subtracted from the 'best' gradient and then divided by the best gradient and
multiplied by 100
Alternatively, the average of the two maximum and minimum lines can be used to
calculate the percentage uncertainty:

max . gradient − min . gradient

percentage uncertainty =
2
× 100%
The percentage uncertainty in the y-intercept can be found using:

best y intercept − worst y intercept

percentage uncertainty =
best y intercept
× 100%

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 12
 max . y intercept − min . y intercept
percentage uncertainty =
2
× 100%
Your notes
Percentage Difference
The percentage difference gives an indication of how close the experimental value
achieved from an experiment is to the accepted value
It is not a percentage uncertainty
The percentage difference is defined by the equation:

experimental value − accepted value

percentage difference =
accepted value
× 100%
The experimental value is sometimes referred to as the 'measured' value
The accepted value is sometimes referred to as the 'true' value
This may be labelled on a component such as the capacitance of a capacitor or the
resistance of a resistor
Or, from a reputable source such as a peer-reviewed data booklet
For example, the acceleration due to gravity g is known to be 9.81 m s–2. This is its
accepted value
From an experiment, the value of g may be found to be 10.35 m s–2
Its percentage difference would therefore be 5.5 %
The smaller the percentage difference, the more accurate the results of the experiment

Worked Example
On the axes provided, plot the graph for the following data and draw error bars and
lines of best and worst fit.

Find the percentage uncertainty in the gradient from your graph.

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 13
 Your notes

Answer:
Step 1: Draw sensible scales on the axes and plot the data

Step 2: Draw the errors bars for each point

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 14
 Your notes

Step 3: Draw the line of best fit

Step 4: Draw the line of worst fit

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 15
 Your notes

Step 5: Work out the gradient of each line and calculate the percentage uncertainty

∆y 26 − 6
best gradient =
∆x = 80 − 0
= 0 . 25

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 16
 ∆ y 27 − 4 . 7
worst gradient =
∆ x = 80 − 0 = 0 . 28 Your notes
0 . 28 − 0 . 25
% uncertainty =
0 . 25
× 100% = 12%

Examiner Tips and Tricks

A common misconception is that error bars need to all be the same size. In physics,
this is not the case and each data point can have different error bar sizes as they have
different uncertainties.

© 2025 Save My Exams, Ltd. Get more and ace your exams at savemyexams.com 17