Uncertainties and Error Propagation

Contents
1. Systematic versus Random Errors
2. Determining Random Errors (a) Instrument Limit of Error, least count
(b) Estimation (c) Average Deviation
(d) Conflicts (e) Standard Error in the Mean
3. What does uncertainty tell me? Range of possible values
4. Relative and Absolute error
5. Propagation of errors (a) add/subtract (b) multiply/divide (c) powers
(d) mixtures of +-*/ (e) other functions
6. Rounding answers properly
7. Significant figures
8. Problems to try
9. Glossary of terms (all terms that are bold face and underlined)

Part III The Vernier

Part II Graphing
Caliper
In this manual there will be problems for you to try. They
are highlighted in yellow, and have answers.
There are also examples highlighted in green.

1. Systematic and random errors.

2. Determining random errors.
3. What is the range of possible values?
4. Relative and Absolute Errors
5. Propagation of Errors, Basic Rules
Suppose two measured quantities x and y have uncertainties, x and y, determined
by procedures described in previous sections: we would report (x ± x), and (y ± y).
From the measured quantities a new quantity, z, is calculated from x and y. What is
the uncertainty, z, in z? For the purposes of this course we will use a simplified
version of the proper statistical treatment. The formulas for a full statistical treatment
(using standard deviations) will also be given. The guiding principle in all cases is to
consider the most pessimistic situation. Full explanations are covered in statistics
courses.

The examples included in this section also show the proper rounding of answers,
which is covered in more detail in Section 6. The examples use the propagation of
errors using average deviations.

(a) Addition and Subtraction: z = x + y or z = x - y

Derivation: We will assume that the uncertainties are arranged so as to make z as far from
its true value as possible.

Average deviations z = |x| + |y| in both cases

With more than two numbers added or subtracted we continue to add the uncertainties.

Using simpler average errors Using standard deviations

Eq. 1a Eq. 1b

Example: w = (4.52 ± 0.02) cm, x = ( 2.0 ± 0.2) cm, y = (3.0 ± 0.6) cm. Find
z = x + y - w and its uncertainty.

z = x + y - w = 2.0 + 3.0 - 4.5 = 0.5 cm

z = x + y + w = 0.2 + 0.6 + Solution with standard

0.02 = 0.82 rounding to 0.8 cm deviations, Eq. 1b, z = 0.633
cm
So z = (0.5 ± 0.8) cm z = (0.5 ± 0.6) cm
 Notice that we round the uncertainty to one significant figure and round the
answer to match.

For multiplication by an exact number, multiply the uncertainty by the same exact
number.

Example: The radius of a circle is x = (3.0 ± 0.2) cm. Find the

circumference and its uncertainty.

C = 2  x = 18.850 cm

C = 2  x = 1.257 cm (The factors of 2 and  are exact)

C = (18.8 ± 1.3) cm

We round the uncertainty to two figures since it starts with a 1, and round the
answer to match.

Example: x = (2.0 ± 0.2) cm, y = (3.0 ± 0.6) cm. Find z = x - 2y and its
uncertainty.

z = x - 2y = 2.0 - 2(3.0) = -4.0

cm

z = x + 2 y = 0.2 + 1.2 =
Using Eq 1b, z = (-4.0 ± 0.9)
1.4 cm
cm.
So z = (-4.0 ± 1.4) cm.

The 0 after the decimal point in 4.0 is significant and must be written in the
answer. The uncertainty in this case starts with a 1 and is kept to two
significant figures. (More on rounding in Section 7.)
(b) Multiplication and Division: z = x y or z = x/y

Derivation: We can derive the relation for multiplication easily. Take the largest values for
x and y, that is

z + z = (x + x)(y + y) = xy + x y + y x + x y

Usually x << x and y << y so that the last term is much smaller than the other terms and
can be neglected. Since z = xy,

z = y x + x y

which we write more compactly by forming the relative error, that is the ratio of z/z,
namely

The same rule holds for multiplication, division, or combinations,

namely add all the relative errors to get the relative error in the result.

Using simpler average errors Using standard deviations

Eq. 2a Eq.2b

Example: w = (4.52 ± 0.02) cm, x = (2.0 ± 0.2) cm. Find z = w x and its
uncertainty.

z = w x = (4.52) (2.0) = 9.04

Using Eq.
2b we get
z =
 So z = 0.1044 (9.04 ) = 0.944 which 0.905
we round to 0.9 , and
z = (9.0 ±
z = (9.0 ± 0.9) .
0.9) .

The uncertainty is rounded to one significant figure and the result is rounded
to match. We write 9.0 rather than 9 since the 0 is significant.

Example: x = ( 2.0 ± 0.2) cm, y = (3.0 ± 0.6) sec Find z = x/y.

z = 2.0/3.0 = 0.6667 cm/s.

So z = 0.3 (0.6667 Using Eq. 2b we get z =
cm/sec) = 0.2 cm/sec (0.67 ± 0.15) cm/sec
z = (0.7 ± 0.2) cm/sec

Note that in this case we round off our answer to have no more decimal
places than our uncertainty.

(c) Products of powers: .

The results in this case are

Using simpler average errors Using standard deviations

Eq. 3a Eq.3b

Example: w = (4.52 ± 0.02) cm, A = (2.0 ± 0.2) , y = (3.0 ± 0.6) cm. Find
.

Using
Eq. 3b,
z=(29 ±
12)
 The second relative error, (y/y), is multiplied by 2 because
the power of y is 2.
The third relative error, (A/A), is multiplied by 0.5 since a
square root is a power of one half.

So z = 0.49 (28.638 ) = 14.03 which we round to

14

z = (29 ± 14)

Because the uncertainty begins with a 1, we keep two significant figures and
round the answer to match.

(d) Mixtures of multiplication, division, addition, subtraction, and powers.

If z is a function which involves several terms added or subtracted we must apply the
above rules carefully. This is best explained by means of an example.

Example: w = (4.52 ± 0.02) cm, x = (2.0 ± 0.2) cm, y = (3.0 ± 0.6) cm. Find
z = w x +y^2

z = wx +y^2 = 18.0

First we compute v = wx as in the example in (b) We have v = wx = (9.0 ±

to get v = (9.0 ± 0.9) . 0.9) cm.
Next we compute The calculation of the
uncertainty in is the
same as that shown to the
left. Then from Eq. 1b

Finally, we compute z = v + (y^2) = 0.9 + z = 3.7

3.6 = 4.5 rounding to 4 z = (18 ± 4) .
Hence z = (18 ± 4) .
(e) Other Functions: e.g.. z = sin x. The simple approach.

For other functions of our variables such as sin(x) we will not give formulae.
However you can estimate the error in z = sin(x) as being the difference between the
largest possible value and the average value. and use similar techniques for other
functions.
Thus

(sin x) = sin(x + x) - sin(x)

Example: Consider S = x cos () for x = (2.0 ± 0.2) cm,  = 53 ± 2 °. Find S

and its uncertainty.

S = (2.0 cm) cos 53° = 1.204 cm

To get the largest possible value of S we would make x larger, (x + x) = 2.2
cm, and  smaller, ( - ) = 51°. The largest value of S, namely (S + S), is
(S + S) = (2.2 cm) cos 51° = 1.385 cm.

The difference between these numbers is S = 1.385 - 1.204 = 0.181 cm

which we round to 0.18 cm.

Then S = (1.20 ± 0.18) cm.

(f) Other Functions: Getting formulas using partial derivatives

The general method of getting formulas for propagating errors involves the total
differential of a function. Suppose that z = f(w, x, y, ...) where the variables w, x, y,
etc. must beindependent variables!

The total differential is then

We treat the dw = w as the error in w, and likewise for the other differentials, dz, dx,
dy, etc. The numerical values of the partial derivatives are evaluated by using the
average values of w, x, y, etc. The general results are
 Using simpler average errors

Eq. 4a.

Using standard deviations

Eq. 4b

Example: Consider S = x cos () for x = (2.0 ± 0.2) cm,  = (53 ± 2) °=

(0.9250 ± 0.0035) rad. Find S and its uncertainty. Note: the uncertainty in
angle must be in radians!

S = 2.0 cm cos 53° = 1.204 cm

Hence S = (1.20 ± 0.13) cm (using average deviation approach) or S = (1.20

± 0.12) cm (using standard deviation approach.)

6. Rounding off answers in regular and scientific notation.

In the above examples we were careful to round the answers to an appropriate number
of significant figures. The uncertainty should be rounded off to one or two significant
figures. If the leading figure in the uncertainty is a 1, we use two significant figures,
otherwise we use one significant figure. Then the answer should be rounded to match.
Example Round off z = 12.0349 cm and z = 0.153 cm.

Since z begins with a 1, we round off z to two significant figures:

z = 0.15 cm. Hence, round z to have the same number of decimal places:

z = (12.03 ± 0.15) cm.

When the answer is given in scientific notation, the uncertainty should be given in
scientific notation with the same power of ten. Thus, if

z = 1.43 x s and z = 2 x s,

we should write our answer as

z = (1.43± 0.02) x s.

This notation makes the range of values most easily understood. The following is
technically correct, but is hard to understand at a glance.

z = (1.43 x ±2x ) s. Don't write like this!

Problem: Express the following results in proper rounded form, x ± x.

(i) m = 14.34506 grams, m = 0.04251 grams.

(ii) t = 0.02346 sec, t = 1.623 x 10-3sec.
(iii) M = 7.35 x kg M = 2.6 x kg.
(iv) m = 9.11 x kg m = 2.2345 x kg Answer

7. Significant Figures
The rules for propagation of errors hold true for cases when we are in the lab, but
doing propagation of errors is time consuming. The rules for significant figures allow
a much quicker method to get results that are approximately correct even when we
have no uncertainty values.

A significant figure is any digit 1 to 9 and any zero which is not a place holder. Thus,
in 1.350 there are 4 significant figures since the zero is not needed to make sense of
the number. In a number like 0.00320 there are 3 significant figures --the first three
zeros are just place holders. However the number 1350 is ambiguous. You cannot tell
if there are 3 significant figures --the 0 is only used to hold the units place --or if there
are 4 significant figures and the zero in the units place was actually measured to be
zero.

How do we resolve ambiguities that arise with zeros when we need to use zero as a
place holder as well as a significant figure? Suppose we measure a length to three
significant figures as 8000 cm. Written this way we cannot tell if there are 1, 2, 3, or 4
significant figures. To make the number of significant figures apparent we use
scientific notation, 8 x cm (which has one significant figure), or 8.00 x cm
(which has three significant figures), or whatever is correct under the circumstances.

We start then with numbers each with their own number of significant figures and
compute a new quantity. How many significant figures should be in the final answer?
In doing running computations we maintain numbers to many figures, but we must
report the answer only to the proper number of significant figures.

In the case of addition and subtraction we can best explain with an example. Suppose
one object is measured to have a mass of 9.9 gm and a second object is measured on a
different balance to have a mass of 0.3163 gm. What is the total mass? We write the
numbers with question marks at places where we lack information. Thus 9.9???? gm
and 0.3163? gm. Adding them with the decimal points lined up we see

09.9????
00.3163?
10.2???? = 10.2 gm.

In the case of multiplication or division we can use the same idea of unknown digits.
Thus the product of 3.413? and 2.3? can be written in long hand as

3.413?
2.3?
?????
10239?
6826?
7.8????? = 7.8

The short rule for multiplication and division is that the answer will contain a number
of significant figures equal to the number of significant figures in the entering number
having the least number of significant figures. In the above example 2.3 had 2
significant figures while 3.413 had 4, so the answer is given to 2 significant figures.
It is important to keep these concepts in mind as you use calculators with 8 or 10 digit
displays if you are to avoid mistakes in your answers and to avoid the wrath of
physics instructors everywhere. A good procedure to use is to use use all digits
(significant or not) throughout calculations, and only round off the answers to
appropriate "sig fig."

Problem: How many significant figures are there in each of the

following? Answer

(i) 0.00042 (ii) 0.14700 (ii) 4.2 x (iv) -154.090 x

8. Problems on Uncertainties and Error Propagation.

Try the following problems to see if you understand the details of this part . The
answers are at the end.

(a) Find the average and the average deviation of the following measurements of a
mass.

4.32, 4.35, 4.31, 4.36, 4.37, 4.34 grams.

(b) Express the following results in proper rounded form, x ± x.

(i) m = 14.34506 grams, m = 0.04251 grams.

(ii) t = 0.02346 sec, t = 1.623 x sec.
(iii) M = 7.35 x kg M = 2.6 x kg.
(iv) m = 9.11 x kg m = 2.2345 x kg

(c) Are the following numbers equal within the expected range of values?

(i) (3.42 ± 0.04) m/s and 3.48 m/s?

(ii) (13.106 ± 0.014) grams and 13.206 grams?
(iii) (2.95 ± 0.03) x m/s and 3.00 x m/s

(d) Calculate z and z for each of the following cases.

(i) z = (x - 2.5 y + w) for x = (4.72 ± 0.12) m, y = (4.4 ± 0.2) m, w = (15.63 ± 0.16)

m.
(ii) z = (w x/y) for w = (14.42 ± 0.03) m/ , x = (3.61 ± 0.18) m, y = (650 ± 20) m/s.
(iii) z = for x = (3.55 ± 0.15) m.
(iv) z = v (xy + w) with v = (0.644 ± 0.004) m, x = (3.42 ± 0.06) m, y = (5.00 ± 0.12)
m, w = (12.13 ± 0.08) .
(v) z = A sin y for A = (1.602 ± 0.007) m/s, y = (0.774 ± 0.003) rad.

(e) How many significant figures are there in each of the following?

(i) 0.00042 (ii) 0.14700 (ii) 4.2 x (iv) -154.090 x 10-27

(f) I measure a length with a meter stick which has a least count of 1 mm I measure
the length 5 times with results in mm of 123, 123, 124, 123, 123 mm. What is the
average length and the uncertainty in length?

Answers for Section 8:

(a) (4.342 ± 0.018) grams

(b) i) (14.34 ± 0.04) grams ii) (0.0235 ± 0.0016) sec or (2.35 ± 0.16) x sec
iii) (7.35 ± 0.03) x kg iv) (9.11 ± 0.02) x kg

(c) Yes for (i) and (iii), no for (ii)

(d) i) (9.4 ± 0.8) m ii) (0.080 ± 0.007) m/s iii) (45 ± 6) iv) 18.8 ± 0.6) v)
(1.120 ± 0.008 m/s

(e) i) 2 ii) 5 iii) 2 iv) 6

(f) (123 ± 1) mm (I used the ILE = least count since it is larger than the average
deviation.)

9. Glossary of Important Terms

Term Brief Definition

 The actual error in a quantity, having the same units as the
Absolute quantity. Thus if
error c = (2.95 ± 0.07) m/s, the absolute error is 0.07 m/s.
See Relative Error.
How close a measurement is to being correct. For gravitational
Accuracy acceleration near the earth, g = 9.7 m/s2 is more accurate than g
= 9.532706 m/s2. SeePrecision.
When several measurements of a quantity are made, the sum of
Average
the measurements divided by the number of measurements.
Average The average of the absolute value of the differences between
Deviation each measurement and the average. See Standard Deviation.
The fraction of measurements that can be expected to lie
within a given range. Thus if m = (15.34 ± 0.18) g, at 67%
Confidence
confidence level, 67% of the measurements lie within (15.34 -
Level
0.18) g and (15.34 + 0.18) g. If we use 2 deviations (±0.36
here) we have a 95% confidence level.
A measure of range of measurements from the average. Also
Deviation
called error oruncertainty.
A measure of range of measurements from the average. Also
Error
called deviation or uncertainty.
An uncertainty estimated by the observer based on his or her
Estimated
knowledge of the experiment and the equipment. This is in
Uncertainty
contrast to ILE, standard deviation or average deviation.
The familiar bell-shaped distribution. Simple statistics assumes
Gaussian
that random errors are distributed in this distribution. Also
Distribution
called Normal Distribution.
Changing the value of one variable has no effect on any of the
Independent
other variables. Propagation of errors assumes that all
Variables
variables are independent.
Instrument
Limit The smallest reading that an observer can make from an
of Error instrument. This is generally smaller than the Least Count.
(ILE)
The size of the smallest division on a scale. Typically
Least Count
the ILE equals the least count or 1/2 or 1/5 of the least count.
The familiar bell-shaped distribution. Simple statistics assumes
Normal
that random errors are distributed in this distribution. Also
Distribution
called Gaussian Distribution.
The number of significant figures in a measurement. For
gravitational acceleration near the earth, g = 9.532706 m/s2 is
Precision
more precise than g = 9.7 m/s2. Greater precision does not
mean greater accuracy! See Accuracy.
Propagation Given independent variables each with an uncertainty, the
of Errors method of determining an uncertainty in a function of these
variables.
Random Deviations from the "true value" can be equally likely to be
Error higher or lower than the true value. See Systematic Error.
Measurements give an average value, <x> and an
Range of
uncertainty, x. At the 67% confidence level the range of
Possible
True Values possible true values is from <x> - x to <x> + x. See
Confidence Level .
The ratio of absolute error to the average, x/x. This may also
Relative
be called percentage error or fractional uncertainty.
Error
See Absolute Error.
Significant All non-zero digits plus zeros that do not just hold a place
Figures before or after a decimal point.
Standard
The statistical measure of uncertainty. See Average Deviation.
Deviation
Standard An advanced statistical measure of the effect of large numbers
Error of measurements on the range of values expected for the
in the Mean average (or mean).
A situation where all measurements fall above or below the
Systematic
"true value". Recognizing and correcting systematic errors is
Error
very difficult.
A measure of range of measurements from the average. Also
Uncertainty
called deviation or error.