1 MEASUREMENTS AND ERROR

1.1 Objectives:
1. To understand the form of measurements in the laboratory, including the measured
values and their uncertainties.
2. To learn the types of errors, understand the difference between precision and accuracy.
3. To know how to obtain the uncertainties for measurements made using laboratory
instruments.
4. To learn when to discriminate between measurements that agree and those that are
discrepant.
5. To investigate reaction time and the "error" involved in repeating a measurement a
number of times.
6. To learn some basic operations in Excel.

1.2 Apparatus:
Meter stick.

1.3 Error as Uncertainties

1.3.1 Background

Measurement is one of the fundamental activities which occur in the physical sciences.
Measurement allows us to put a value on quantities such as mass, time and length etc. All
measurements, however careful and scientific, are subject to some errors. This is because we live
in a real world and measurements are never perfect due to limitations with our measuring device.
The process of evaluating the uncertainties associated with a measurement result is often called
uncertainty analysis or error analysis. The term error or uncertainty does not imply a mistake or
carelessness on the part of the experimenter but its analysis rather allows the experimenter to
estimate how large their uncertainties are and help them to reduce the uncertainties to a level
that allows proper conclusions to be drawn. As such, you cannot eliminate error by being very
careful.

When a measurement is made, it is generally assumed that some exact or true value exists based
on how we define what is being measured. What is perceived as the exact or true value may be
theoretical and while we may never know this true value exactly, we attempt to find this ideal

9
quantity to the best of our ability with the time and resources available. Sometimes, we make
measurements by different methods, or make multiple measurements using the same method
and we may obtain slightly different results. Suppose we measure a quantity 𝑥, the usual way to
display the results or show the range of values that we believe includes the true value is:

𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑥 = 𝑏𝑒𝑠𝑡 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒 ± 𝑢𝑛𝑐𝑒𝑟𝑡𝑎𝑖𝑛𝑡𝑦 1.1

In general, the result of any measured quantity 𝑥 can be written as:

𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑥 = 𝑥 ± 𝛿𝑥 1.2

The number 𝑥 is the best estimate, and the part 𝛿𝑥 which is always defined to be positive is
called the error or margin of error or the uncertainty. As an example, suppose we measure the
acceleration of gravity and obtain the results 9.7m/s2, 9.9m/s2, and 10.1m/s2, then our best
estimate will be the average of these three values which is 9.9m/s 2. The probable range is
9.7m/s2 to 10.1m/s2. In the context of equation 1.1 or 1.2, the measured value will be reported
as:
𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 = 9.9 ± 0.2 𝑚/𝑠 1.3

The form expressed in equation 1.3 means 𝑥 = 9.9𝑚/𝑠 and 𝛿𝑥 = 0.2 𝑚/𝑠 . Thus, the
measured value has a minimum of 9.9 - 0.2 = 9.7m/s 2 and a maximum of 9.9 + 0.2 = 10.1m/s2.

Consider this example: 𝑡𝑖𝑚𝑒 = 3.12 ± 0.03 𝑠. A close inspection reveals that when the best
estimate of the time result is good to the hundredth place, so is the uncertainty. Suppose we
measure the acceleration of gravity and obtain the best estimate to be 9.85m/s 2 with uncertainty
of 0.0275985m/s2, it would be out of place to state the result as:

𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑔 = 9.85 ± 0.0275985 𝑚/𝑠 1.4

It is reasonable to state the result in terms of the best estimate and uncertainty as:

𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑔 = 9.85 ± 0.03𝑚/𝑠 1.5

Thus, we conclude that in a measurement

(i) the measured value and its uncertainty must always have the same number of digits
after the decimal place.
(ii) Experimental uncertainties should almost always be rounded to one significant figure.

10
1.3.2 Types of Errors
Measurement errors may be classified as either random or systematic, depending on how the
measurement was obtained. An instrument could cause a random error in one situation and a
systematic error in another. Experimental uncertainties that can be revealed by repeating the
measurements are called random errors; those that cannot be revealed this way are called
systematic errors.

Random errors are statistical fluctuations (in either direction) in the measured data due to the
precision limitations of the measuring device. Random errors can be evaluated through statistical
analysis and can be reduced by averaging over many observations. For example, you may want
to weigh a small piece of wood very accurately. You weigh it on five different balances and get a
slightly different value each time. The reasons for the variation could be that each balance is
calibrated somewhat differently; slightly different amounts of dust on the piece of wood each
time; different moisture content due to the humidity being higher during some of the readings,
and so on.

Systematic Error is due to some sort of bias in a measuring system, which causes the
measurements to be consistently too high or too low. Systematic Errors are reproducible
inaccuracies that consistently affect results in the same direction. For example, "if the needle on
an instrument is slightly bent, it will consistently read too high. Another example is if the observer
has some sort of bias (he is part of the measuring system, too). Suppose you're trying to read a
gauge and the needle is not steady, but rather moves back and forth. Your opinion of the average
reading may be influenced by what you expect it to be. If a systematic error is identified when
calibrating against a standard, the bias can be reduced by applying a correction or correction
factor to compensate for the effect. Unlike random errors, systematic errors cannot be detected
or reduced by increasing the number of observations.

There is no such thing as Human Error. Though human error can simply be described as an error
made by human and does not point to any particular type of error described above, this phrase
is vague and does not describe the source of error clearly. Its use is highly discouraged in this
course. gives some of the common sources of error in the laboratory and their associated
classification.

11
 Table 1.1: Some common sources of error in the laboratory and their classification.

Source of Error Classification Remarks

Failure to account for a factor e.g. air Systematic Look for ways to reduce these factors, e.g. by
resistance, effect of friction using air supply to reduce friction on air track

Environmental factors Systematic or Beware of your environment and look for ways to
e.g. errors introduced by your immediate random eliminate these errors.
working environment. These errors may
include changes in temperature,
electronic noise or other effects from
nearby apparatus.
Instrument resolution e.g. Instruments Random A meter stick can only measure one digit after the
have finite precision that limits the ability decimal place. Measuring length between say 3.4
to resolve small measurement differences. and 3.5 cm can be estimated for instance as
3.45cm.
Failure to calibrate or check zero of Systematic It is a good practice to always check for the
instrument calibration of an instrument before taking data. If
a calibration standard option is not available, the
accuracy of the instrument should be checked by
comparing with another instrument that is at least
as precise, or you may check with the technical
data provided by the manufacturer.
Physical variations Random It is always advisable to obtain multiple
measurements over the entire range being
investigated. Doing so often reveals variations
that might otherwise go undetected. If desired,
these variations may be cause for closer
examination, or they may be combined to find an
average value.
Parallax Systematic or This error occurs whenever there is some distance
random between the measuring scale and the indicator
used to obtain a measurement or if the observer's
eye is not squarely aligned with the pointer and
scale, the reading may be too high or low (some
analog meters have mirrors to help with this
alignment).
Instrument drift Systematic This occurs with most electronic instruments.
They have readings that drift over time. The
amount of drift is generally not a concern, but
occasionally this source of error can be significant
and should be considered.

12
1.3.3 Precision, Accuracy and Discrepancy
There are two terms often associated with experimental error: accuracy and precision. The
accuracy of an experiment is a measure of how close the experimental result is to the true value.
This, of course, presupposes that a true value exists and is known before the experiment is
performed.

Figure 1.1. Random and systematic errors in target practice. (a) Since all shots in and around the target are close
to one another and evenly distributed, we conclude both random and systematic errors are small. Result is
described as precise and accurate (b) Since shots are close to one another, random error is still small. Systematic
error is large because shots shift to the right of the target and result is described as precise but not accurate. (c)
Shots are not close to one another so random error is large. Shots are evenly scattered so systematic error is
small. Result is described as accurate but not precise. (d) Both random and systematic errors are large and result
is neither accurate nor precise.

Al and Bertha, two Stockton students, make independent measurements of the length of a metal
rod, obtaining the values 5.03 cm and 4.99 cm, respectively. If the rod is known to be 5.00 cm
long to two decimal places, Bertha's result of 4.99 cm is more accurate than Al's value of 5.03 cm.

The precision of an experiment, measurement, or instrument is a measure of how reproducible

the result is. This is often measured by the smallness of the uncertainty in the measurement. Let
us again consider Al's and Bertha's measurements. Al reports a value of 5.03 ± 0.05 cm, which
means that his measurement indicates that the rod is probably between 4.98 and 5.08 cm long

13
and this is Al’s range of values or margin of error. Bertha reports a value of 4.99 ± 0.03 cm, which
means her measurement indicates that the rod's length is probably between 4.96 and 5.02 cm.
Bertha's value is more precise because her uncertainty or margin of error is smaller. Note in this
case that most precise result is also more accurate. However, it is possible for the more precise
result to be less accurate simply due to random fluctuations or due to some unaccounted
systematic error present in the experiment (see Figure 1.1). Also, note that both values (Al's and
Bertha's) are consistent with the accepted value of the length of the rod, since 5.00 cm lies within
both error ranges.

In Physics labs, sometimes we compare the value obtained in a measurement to a standard or

theoretical one, or we compare two measurements of the same quantity. When the two
measurements being compared do not agree, we want to know by how much they do not agree.
The difference between the two measurements of the same quantity is the discrepancy.
Discrepancy may or may not be significant.

Using the results obtained by Al and Bertha, their discrepancy is 5.03 – 4.99 which is
0.04cm. This result is insignificant since the discrepancy is smaller than the combined
uncertainty of 0.08cm, and their margins of error overlap. There’s no reason to doubt
either measurements.

Suppose this time, Al reports 5.03 ± 0.02 cm and Bertha also reports 4.99 ± 0.01 cm. Al’s margins
of error is 5.01 and 5.05 cm and Bertha’s is 4.98 and 5.00 cm. Their combined uncertainty is
0.03cm, and the discrepancy is still 0.04cm. The discrepancy is significant because their margins
of error do not coincide and is also larger than the combined uncertainty. One of them must be
doing something wrong!

1.4 Comparison of Values

1.4.1 Comparison of Measured and Accepted Values
In introductory physics laboratory, you may be asked to measure a quantity and compare to that
which has been measured carefully many times before by scientist, and for which an accurate
accepted value has been established. The accepted value may have an uncertainty but, in many
cases, this uncertainty is extremely small, and the value is much more accurate. Discussing the
difference between a measured value and a related accepted value is useful and helps to validate
your method.

14
As an illustration, suppose in an experiment to measure the speed of sound in air (at standard
temperature and pressure, stp) John obtains 330 ± 4 𝑚/𝑠, Bryan arrives at 326 ± 4 𝑚/𝑠 and
Sam measures 341 ± 4 𝑚/𝑠. The accepted value of the speed of sound at stp is 331m/s.

All the three values are compared graphically in Figure 1.2. Because the accepted value of
331m/s falls in John’s margins of error, we conclude that John’s result is satisfactory. The
accepted value is just outside of Bryan’s margin of error, but his measurement is nevertheless
acceptable. The accepted value is far outside Sam’s margins of error and his result is certainly not
satisfactory.

Figure 1.2. Comparison of measured and accepted values

1.4.2 Comparison of Two Measured Values

Sometimes, a laboratory exercise will involve measuring two quantities predicted to be equal by
theory. For example, the law of conservation of momentum states that for a collision occurring
between two objects in an isolated system (i.e. free of external forces), the total momentum of

15
the two objects before the collision is equal to the total momentum of the two objects after the
collision.

Suppose two carts collide on a frictionless dynamics track and for a single pair of collision, we
measure the initial total momentum as 𝐴 = 2.50 ± 0.04 𝑘𝑔 ∙ 𝑚/𝑠 and the final total momentum
as 𝐵 = 2.57 ± 0.06 𝑘𝑔 ∙ 𝑚/𝑠. According to theory, if total the momentum is conserved, then
𝐴 = 𝐵 within experimental uncertainties. Figure 1.3 shows a graphical representation of the
results. Because the ranges of 𝐴 and 𝐵 overlap, these measurements are absolutely consistent
with the conservation of momentum.

Figure 1.3. Comparison of two measured quantities using conservation of momentum

1.4.3 Acceptability of a Measured Result

In the example to illustrate whether discrepancy is significant or not, we argued that, if the
discrepancy (that is 𝑥 −𝑥 ) between an expected value 𝑥 and the best result 𝑥 from
experiment is less than (or equal to) the combined uncertainty 𝜎, then the result is satisfactory;
but if the discrepancy is greater than the combined uncertainty, then the result is unsatisfactory.
A more quantitative description of this statement is1

𝑥 −𝑥 1.6
𝑡=
𝜎

1
In t-test statistics suppose 𝑥 = 𝑥 ± 𝛿𝑥 and 𝑦 = 𝑦 ± 𝛿𝑦 are two measured quantities, then to study the
| |
discrepancy between 𝑥 and 𝑦, we modify equation 1.6 to obtain 𝑡 = . This is better than equation 1.6.

16
where 𝑡 is the number of standard deviations by which 𝑥 differs from 𝑥 . Clearly if 𝑥 =𝑥 ,
then 𝑡 = 0 and implies perfect agreement. In this laboratory exercise, usually results for which
𝑡 ≤ 1 i.e. less than or equal to one standard deviation shows somewhat perfect agreement
between 𝑥 and 𝑥 .

As an illustration, suppose we want to compare how acceptable John’s result (330 ± 4 𝑚/𝑠) is
to the acceptable value (331 𝑚/𝑠) as in section 1.4.1, then 𝑥 = 330 m/s, 𝑥 =331 m/s and 𝜎 =
4 m/s. Using equation 1.6,

|330 − 331|
𝑡= = 0.25
4

Since 𝑡 = 0.25 is less than 1, we conclude that John’s result is acceptable. On the other hand,
Sam’s result is unacceptable based on the standard set in this laboratory course because his 𝑡 =
2.5 is greater than 1.

Example 1
Two groups A and B of different age category and each consisting of ten members measured the
length of a room. The results of the length in meters are shown below:

Group A Group B
12.1 11.8
12.3 12.0
12.8 13.0
12.9 13.2
12.4 13.5
12.6 11.5
12.7 12.0
12.2 13.2
12.0 13.9
12.5 12.5

(a) What type of error (e.g. random or systematic) does each group exhibit?
(b) Write the result of each group in the form of equation 1.2.
(c) Discuss the precision of the two groups.
(d) Discuss the discrepancy of the two groups.

17
 (e) The nominal value of the length of the room is (12.4 ± 0.1)𝑚. How do the results of
the two groups compare with the nominal value?

Solution
(a) Since the distribution appears to center around a mean value, both exhibit a random
error.
(b) Here, the mean value is the best estimate and the standard deviation characterizes the
average uncertainty of the measurements. Using MS Excel, we have:
𝐿 = (12.5 ± 0.3)𝑚, 𝐿 = (12.7 ± 0.8)𝑚.
(c) The uncertainty of group A is smaller than that of group B. As a result, group A is more
precise compared to group B.
(d) The dicrepancy of the two groups is 𝐿 − 𝐿 = 12.7 − 12.5 = 0.2𝑚. The combined
uncertainty of the two groups is 𝛿 = 𝛿 + 𝛿 = 0.3 + 0.8 = 1.1𝑚. The discrepancy is
insignificant since it is smaller than the combined uncertainty. Thus the margins
of error of the two groups overlap and there is no reason to doubt either
measurements.

(e) Since the margin of error for the nominal value coincides with that of A and B we conclude
that both results are satisfactory.

Alternatively, using equation 1.6, we can analyze the results as follows:

| . . | . | . . | .
𝑡 = ( .
= = 0.25 and 𝑡 = = = 0.33. Since both 𝑡 and 𝑡 are
. ) . ( . . ) .
less than 1, there is an agreement between the measured and nominal values.

18
1.5 Propagation of Uncertainties
In science, most physical quantities cannot be measured in a single direct measurement but
instead in two or more separate steps. For instance, in order to find the area of a rectangle, the
length and width will be measured separately. Suppose the length of the rectangle is 𝑙 ± 𝛿𝑙 and
the width is 𝑤 ± 𝛿𝑤, then the area is (𝑙 ± 𝛿𝑙 )(𝑤 ± 𝛿𝑤). The uncertainties in length and width
will “propagate” through the calculations to produce an uncertainty in the final answer for the
area. This propagation of uncertainties is the main focus of this section.

1.5.1 Estimating Uncertainties in the Lab

In this laboratory course, you will be required most of the time (unless directed otherwise) to
record measurements and their uncertainties. Recording measurements may be an easy task but
figuring out the error associated with the measuring device is a bit challenging.

Figure 1.4. Uncertainty in measuring with a ruler.

As an example, let’s look at . It is a portion of a meter stick that is used to determine the length
of a pencil. Here, we want to assign a reasonable value to the uncertainty in the measurement.
Since the least count of the ruler is 0.1 cm the length of the pencil will lie in the range 3.4cm -
3.5cm. The process of estimating positions between the scale markings of an instrument is called
interpolation. In writing the length of the pencil in the form of equation 1.2, the best value is the
midpoint between 3.4 cm and 3.5cm which is 3.45cm, and hence the length becomes 3.45 ±
0.05 𝑐𝑚.

In order to determine the voltage in Figure 1.5 we have to decide where the needle points on the
voltmeter scale. The least count of the voltage scale is 1 volt and so one will resort to guessing
the value. Any guess between 5 – 6 volts is accepted due to this limitation. Thus, on the basis of
equation 1.2, the voltage is 5.5 ± 0.5 𝑣𝑜𝑙𝑡𝑠. Since the uncertainty in the ruler is smaller than the
voltmeter, we can say confidently that the ruler scale has a better precision than the voltmeter
scale.

19
 Figure 1.5. Uncertainty in a voltmeter

1.5.2 Examples on adding, subtracting, multiplying and dividing measured

quantities and their uncertainties

Example 2
Find the sum of 3.4 ± 0.1 and 6.8 ± 0.5

Answer
3.4 ± 0.1
+6.8 ± 0.5
10.2 ± 0.6

Example 3
The length of a rectangle is 𝐿 = (5.04 ± 0.05) cm and the width is 𝑊 = (3.50 ± 0.02) cm. Find
(a) The perimeter 𝑃
(b) Area 𝐴
Write your answer in a form consistent with equation 1.2.

Solution
(a) Since 𝑃 = 2(𝐿 + 𝑊), we need 𝑃 and 𝑃 and from there find the 𝑃 as
𝑃 = , and the uncertainty is 𝛿𝑃 = 𝑃 −𝑃 or 𝛿𝑃 = 𝑃 −𝑃 .
𝑃 = 2(𝐿 +𝑊 ) and 𝑃 = 2(𝐿 +𝑊 ).

𝑃 𝑃 𝑃

𝛿𝑃 𝛿𝑃

Thus, 𝑃 = 𝑃 ± 𝛿𝑃. Now 𝑃 = 2[(5.04 ± 0.05) + (3.50 ± 0.02)].

𝑃 = 2[(5.04 + 0.05) + (3.50 + 0.02)] = 17.22𝑐𝑚

20
 = 2[(5.04 − 0.05) + (3.50 − 0.02)] = 16.94𝑐𝑚
𝑃
17.22𝑐𝑚 + 16.94𝑐𝑚
𝑃 = = 17.08𝑐𝑚
2
The uncertainty is 𝛿𝑃 = 17.22𝑐𝑚 − 17.08𝑐𝑚 = 0.14𝑐𝑚. It follows that the perimeter is
𝑃 = (17.08 ± 0.14)𝑐𝑚

(b) Area 𝐴 = 𝐿𝑊. 𝐴 =𝐿 𝑊 = (5.04 + 0.05)(3.50 + 0.02) = 17.92𝑐𝑚 .

𝐴 =𝐿 𝑊 = (5.04 − 0.05)(3.50 − 0.02) = 17.37𝑐𝑚 .
. .
𝐴 = = 17.64𝑐𝑚 and 𝛿𝐴 = 17.92𝑐𝑚 − 17.64𝑐𝑚 = 0.28𝑐𝑚
𝐴 = (17.64 ± 0.28)𝑐𝑚

Example 4
The density 𝜌 of a substance is defined as the ratio of its mass 𝑚 to the volume 𝑉. If the mass of
a substance is 𝑚 = (2.1 ± 0.3)𝑘𝑔 and the volume is 𝑉 = (0.9 ± 0.1)𝑚 , calculate the density.

Solution
( . . )
𝜌= ,𝜌 = and 𝜌 = . Now 𝜌 = =( = 3.0𝑘𝑔/𝑚 .
. . )
( . . ) ( . . )
𝜌 = =( = 1.8𝑘𝑔/𝑚 . 𝜌 = = 2.4𝑘𝑔/𝑚 and 𝛿𝜌 = 3.0 − 2.4 =
. . )
0.6𝑘𝑔/𝑚 . Thus 𝜌 = (2.4 ± 0.6)𝑘𝑔/𝑚 .

Example 5
The quantity 𝐶 relates 𝐴 and 𝐵 through the equation 𝐶 = 𝐴 – 𝐵. If 𝐴 = 69.7 ± 0.5 and
𝐵 = 2.30 ± 0.02, Find 𝐶 and 𝐶 and hence write 𝐶 in the form of equation 1.2.

Solution
𝐶 =𝐴 −𝐵 and 𝐶 =𝐴 −𝐵 .
𝐶 = (69.7 + 0.5) − (2.30 − 0.02) = 70.2 − 2.28 = 67.92. Similarly,
𝐶 = (69.7 − 0.5) − (2.30 + 0.02) = 69.2 − 2.32 = 66.88. The range is 66.88 − 67.92.
. .
𝐶 = = 67.40 and 𝛿𝐶 = 67.92 − 67.40 = 0.52. Hence 𝐶 = 67.40 ± 0.52.

Example 6
If 𝜃 = 30° ± 2°, find sin 𝜃.

21
Solution
𝜃 = 30° + 2° = 32° and 𝜃 = 30° − 2° = 28°.
sin 𝜃 = sin 32° = 0.53, sin 𝜃 = sin 28° = 0.47. The mean is 0.50 and the uncertainty is
0.03. Thus sin(30° ± 2°) = 0.50 ± 0.03.

1.5.3 Questions
1. Find the sum of the following: 6.7 ± 0.2 cm, 2.5 ± 0.1 cm, and 5.2 ± 0.2 cm.

2. In a one dimensional motion along a dynamics track, a student measured the position,
velocity, and acceleration at one instant. The results are shown in Table 1.2. Rewrite these
results in the standard form 𝑥 ± 𝛿𝑥.

Table 1.2: Measurements of position, velocity, and acceleration.

Variable Best estimate Probable range

Position, x 9.3 cm 9.1 to 9.5 cm
Velocity, v -10.5 cm/s -11.0 cm/s to -10.0cm/s
Acceleration, a 41 cm/s2 37 cm/s2 to 45 cm/s2

3. The height of a building is measured as 6.02 ± 0.03529 meters. Rewrite this result in the
clearest form, with suitable numbers of significant figures.

4. A student measured the acceleration of gravity, g (in m/s 2) ten times and obtained the
results 9.8, 9.7, 9.6, 10.1, 9.9, 9.9, 9.8, 9.7, 9.8, and 10.2. (a) Find the mean and the
standard deviation and write the result to be consistent with the form 𝑥 ± 𝛿𝑥. (b)
What type of error did the student make? (c) Compare and comment the student’s result
with the accepted value of 9.80 ± 0.05 m/s2 using equation 1.6.

5. Mary and Debi calculated v using the equation

𝑣 = 𝑓𝜆

Their measurement of f was 1000 ± 1 (don't worry about the units), and λ was 0.34 ± 0.01.
What was their maximum and minimum calculated v?

22
6. George made several measurements of the time it took a glider to travel down an inclined
dynamics track. He found that the time was (1.77 ± 0.02) seconds. The distance traveled
was (1.60 ± 0.01) meters. Give the maximum and minimum average velocity (i.e.
distance/time).

7. Body Mass Index (BMI) provides a reliable indicator of body fatness for most people and
is used to screen for weight categories that may lead to health problems (see Table 1.3).
BMI is a number which takes into account your weight in kilograms and height in meters.
It is defined as:

𝑚𝑎𝑠𝑠 𝑖𝑛 𝑘𝑖𝑙𝑜𝑔𝑟𝑎𝑚𝑠 1.7

𝐵𝑀𝐼 =
(ℎ𝑒𝑖𝑔ℎ𝑡 𝑖𝑛 𝑚𝑒𝑡𝑒𝑟𝑠)

Sharon weighs 71.6kg and is 1.8m tall. Assuming an uncertainty of 0.2kg in mass and 0.1m
in height, find her maximum and minimum BMI and use Table 1.3 to comment on your
result.
Table 1.3. The Body Mass Index

BMI (kg/m2) Evaluation

Below 18.5 Underweight
18.5 – 24.9 Normal
25.0 – 29.9 Overweight
30.0 – 39.9 Obese
40 and above Morbidly obese

8. Eileen, Jackie and Kathryn needed to find the difference between two angles, A 1 and A2,
that they had drawn on a diagram:

𝐵 =𝐴 −𝐴

They then had to find the cosine of B. Jackie measured A1 to be 50° and A2 to be 3°. But
their pencil was kind of dull, so the lines they had drawn were about 0.5° thick. How much
uncertainty in each angle was caused by the thick pencil lines (remember this affects both
sides of the angle)? How much uncertainty is there in B? How about the cosine of B?

23
1.6 Measuring Reaction Time
In this section, we shall employ the lessons learned so far about uncertainties to study the
reaction time of a member of your lab group.

1.6.1 Background
A child darts out in front of your car, or you accidentally put your hand on a hot surface. You
cannot react instantaneously. When your senses perceive something, they send a message which
travels very rapidly, but not instantaneously, to your brain. The brain must then interpret the
message and decide upon a course of action. The brain transmits a message of its decision, and
when this reaches the proper muscles, you react. The time it takes for all this to happen is known
as reaction time. Reaction times vary from person to person, and under different conditions.

Figure 1.6: Measuring reaction time. Here, y is the reaction time in centimeters.

You will indirectly measure reaction time by measuring how far a meter stick will drop through
your hand before you can catch it as illustrated in Figure 1.6. This method takes a little practice.
Your partner has to hold the meter stick just above your waiting hand and release it without

24
signaling that moment. For example, if your partner unconsciously lifts his index finger before
releasing the meter stick, this could signal you, resulting in a superhuman reaction time.

The reaction "time" will actually be the measurement of distance (in centimeters) by which the
stick drops. Since all meter sticks fall with the same acceleration, this is a good measurement for
the purpose of comparison. Measuring time by using distance seems strange at first, but is not
so unusual. For example, astronomers measure great distances in light years (i.e. distance which
light would travel in one year). The reverse is also used. How often have you heard the
expression, "15 minutes from Atlantic City?"

How will you find the error in your reaction time? A good method for finding errors is to make
the measurement several times, find the average, and then find out how much the individual
measurements differ from this average. The difference between a measurement and the average
is called the deviation. Obviously, if there's a lot of variation, your error is large. If the
measurements are pretty consistent, the error is small. Since there is a deviation for each
measurement, it is often useful to talk about the standard deviation. The standard deviation of
the measured points (called the population) is the uncertainty.

Table 1.4. Sample table showing reaction time in cm.

Trial Reaction Times 𝑥 (cm) Deviations 𝑥 − 𝑥̅ (cm) Square of Deviation (𝑥 − 𝑥̅ ) (cm2)

1 16 2 4
2 18 4 16
3 12 -2 4
4 10 -4 16
Sum = 56 Sum = 40

As an illustration, consider which gives the reaction time for four trials. The average reaction
time 𝑥̅ in cm is given by:

𝑆𝑢𝑚 𝑜𝑓 𝑟𝑒𝑎𝑐𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒 (16 + 18 + 12 + 10) 56

𝑥̅ = = = = 14 𝑐𝑚
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑟𝑖𝑎𝑙𝑠 4 4

The deviation from the average value is given by:

Deviation (𝑥 − 𝑥̅ ) = Reaction time – Average value

25
For instance, the reaction time of 16cm produces a deviation equal to: 16 – 14 = 2cm. The sum
of the deviation squared is ∑(𝑥 − 𝑥̅ ) = 4 + 16 + 4 + 16 = 40 𝑐𝑚 . The equation for
calculating the standard deviation (SD) of a population is:

∑(𝑥 − 𝑥̅ )
𝑆𝐷 = 1.8
𝑁

Thus,

40
𝑆𝐷 = = 3.16 𝑐𝑚
4

The measured reaction "time" is then y = (14 ± 3) cm. From the average reaction time in distance,
we can proceed to find the actual reaction time in seconds. Suppose the meter stick in Figure 1.6,
is dropped from rest, then the magnitude of the reaction time obeys the equation:

1 1.9
𝑦= 𝑔𝑡
2

where y is the reaction time in distance, g = 980 cm/sec2 is the acceleration due to gravity and t
is the reaction time in seconds. Solving for t, we have

2𝑦
𝑡= 1.10
𝑔

Using equation

1.10, the reaction time can be found as follows:

2𝑦 2(14 ± 3) 𝑐𝑚
𝑡= =
𝑔 980 𝑐𝑚/𝑠

2(14 + 3) 𝑐𝑚
𝑡 = = 0.19 𝑠
980 𝑐𝑚/𝑠

26
 2(14 − 3) 𝑐𝑚
𝑡 = = 0.15 𝑠
980 𝑐𝑚/𝑠

The reaction time in seconds is thus,

𝑡 = (0.17 ± 0.02) 𝑠

1.7 Procedure:
The reaction time exercise in Figure 1.6 for a student is repeated 20 times. The results in 𝑐𝑚 are
as follows: 9, 13, 15.4, 10, 16, 24, 6, 19, 21, 22, 7, 17.7, 18, 11, 22.5, 15, 7, 12, 21.3, and 16.8.

(i) Set up a spreadsheet in Excel similar to Table 1.4 to calculate the average reaction
time and the standard deviation in 𝑐𝑚.
(ii) Find the maximum and minimum reaction times in seconds.

Questions
9. Write the result of the student’s reaction time (in cm) in the form of equation 1.2.

10. Give the range of the student’s time and then write your answer in the form of equation
1.2.

11. List at least two factors that could explain why different people have different reaction
times of the same event.

12. Discuss whether the reaction time measured in the lab will be higher or lower than the
reaction time measured while driving a car? Note that in the lab experiment, the student
may be more anticipating than when we drive a car.

1.8 Real World Applications

Reaction times are used in developmental psychology to study the neural maturity of children’s
information processing system. Slow and apparently delayed reactions from infants and children
often try the patience of parents and teachers alike, and can be interpreted as non-compliance
already before the child gets the chance to respond [1].

27
In sports, reaction time often determines who wins and who loses a race or event. Even more
importantly, in real-life situations, like when driving a car, it can mean the difference between
life and death. For Olympic swimmers and runners, a fraction of a second can determine the
difference between winning a gold medal or a bronze. Furthermore, reaction time could dictate
whether one is winning any medal or returning home with nothing but hopes at another chance
in four more years.

1.9 References
[1] C. Lange-Küttner, The Importance of Reaction Times for Developmental Science: What a
Difference Milliseconds Make, International Journal of Developmental Science 6 (2012), 51–55
DOI 10.3233/DEV-2012-11089IOS Press.

28