Chapter 5

Result and Data Analysis

Dr. Anita Ahmad
Faculty of Electrical Engineering
Universiti Teknologi Malaysia
81310 UTM Skudai, Johor
Malaysia

innovative • entrepreneurial • global

 DATA AND ANALYSIS
Definition

Data:
• refers to a collection of organized information,

• results of experience, observation or experiment, or a set of

premises

• consist of numbers, words, or images, particularly as

measurements or observations of a set of variables

Data analysis:
• process of looking at and summarizing data with the intent to
extract useful information and develop conclusions
2

www.utm.my innovative • entrepreneurial • global

 DATA ANALYSIS
Data of a result of measurement can be classified:

a) Qualitative Data - descriptive, wording, non-numbers

- the iron is red hot
- the fuse blow when the switch turned ‘on’

b) Quantitative Data - numerical value of the data

- the iron temperature is at 1000C
- the fuse blew when 5A current flow

Data analysis to include:

• error • precision
• uncertainty • deviation
• accuracy
3

www.utm.my innovative • entrepreneurial • global

 ERROR

• refers to a difference between

actual behavior or measurement &
the norms or expectations for the
behavior or measurement

• classify into 3 categories

- gross (human)
- systematic
- random

www.utm.my innovative • entrepreneurial • global

 GROSS ERROR

Gross / Human error

• undetected mistakes that cause a measurement to be

very much farther from the mean measurement than
other measurements

• example: reading digital & analog meter by human

- digital shows value of measurement
- analog show value of measurement but range selection must
be correct

www.utm.my innovative • entrepreneurial • global

 GROSS ERROR

Gross / Human error due to:

• misreading instruments

• mistakes in calculation

• improper application of
equipment

• incorrect adjustment

• neglecting loading effects

www.utm.my innovative • entrepreneurial • global

 SYSTEMATIC ERROR

Systematic error (error in measurement)

• is any biasing effect,

in the environment,
methods of observation
or instruments used,
which introduces error
into an experiment and
is such that it always
affects the results of an
experiment in the same direction

• categorized by Instrumentation error & Environmental error

7

www.utm.my innovative • entrepreneurial • global

 SYSTEMATIC ERROR
Systematic error (error in measurement)
Instrumentation Errors
• Friction in moving parts
• Faulty or non-linear components
• Out of Calibration equipment
• Faulty equipment

Estimation:
•Comparison with a known standard
•Application of correction factors

Reduced/ avoided by:

•Calibrating instruments
•Selecting a suitable instrument for the particular measurement
•Inspecting equipment for faulty operation or defective parts
•Using more than one method for high accuracy measurement

www.utm.my innovative • entrepreneurial • global

 SYSTEMATIC ERROR
Systematic error (error in measurement)

Environmental Errors
1. Due to external conditions to the measurement system:
- Temperature
- Pressure
- Humidity
- Magnetic fields
2. Can be detected by careful monitoring of measured
values and environmental conditions.

3. Can be minimized or eliminated by controlling the

environment conditions:
- Air conditioning
- Shielding
- Sealing

www.utm.my innovative • entrepreneurial • global

 RANDOM ERROR
Random error

• Deviations from the "true value" can be equally likely to be

higher or lower than the true value

• factors that randomly

affect measurement

• For instance, each

person's mood can
inflate or deflate their
performance on any
occasion
10

www.utm.my innovative • entrepreneurial • global

 RANDOM ERROR
Random error

-Unexplainable and random in occurrence.

-Many readings need to be taken and statistical techniques are
applied to obtain the best estimate of the true quantity

-Where
i. E= the measurement error
ii. Xmeasured = the value obtained by a measurement
iii. Xtrue = the value obtained by a measurement

11

www.utm.my innovative • entrepreneurial • global

 ABSOLUTE ERROR
Absolute & Relative Error

Absolute error

• is the amount of physical

error in a measurement

• for instance, say a meter

stick is used to measure a given distance, error is ±1mm, the
given error of the measurement is the absolute error

• The actual error in a quantity, having the same units as the

quantity. Thus if c = (2.95 ± 0.07) m/s, the absolute error is 0.07
m/s.
12

www.utm.my innovative • entrepreneurial • global

 RELATIVE ERROR
Absolute & Relative Error

Relative error

• the ratio of absolute error to the average, may also be called

percentage error or fractional uncertainty

• gives an indication of how good a measurement is relative to

the size of the thing being measured

• Let’s say that 2 students measure 2 objects with a meter Stick

• One student measures the height of a room and gets a
value of 3.215 meters ±1mm (0.001m)
• Another student measures the height of a small cylinder and
measures 0.075 meters ±1mm (0.001m)
13

www.utm.my innovative • entrepreneurial • global

 RELATIVE ERROR
Absolute & Relative Error

Relative error

14

www.utm.my innovative • entrepreneurial • global

 RELATIVE ERROR
Absolute & Relative Error

• Example 2

A resistor labeled as 240 Ω is

actually 243.32753 Ω. What are
the absolute, eabs and
relative, erel errors of the labeled
value?

eabs = |240 - 243.32753| ≈ 3.3 Ω

erel = |240 - 243.32753|/|243.32753| ≈ 0.014

15

www.utm.my innovative • entrepreneurial • global

 SUM OF QUANTITIES
Absolute & Relative Error

Sum of quantities
• error in sum of quantities equals sum of errors

E = V1 + V2
= (V1 ± ΔV1) + (V2 ± ΔV2)
= (V1 + V2) ± (ΔV1 + ΔV2)

16

www.utm.my innovative • entrepreneurial • global

 DIFFERENCE OF QUANTITIES
Absolute & Relative Error

Difference of quantities
• error in difference of quantities equals sum of errors

E = V1 - V2
= (V1 ± ΔV1) -(V2 ± ΔV2)
= (V1 - V2) ± (ΔV1 + ΔV2)

17

www.utm.my innovative • entrepreneurial • global

 PRODUCT OF QUANTITIES
Absolute & Relative Error

Product of quantities
• percentage error in product or quotient of quantities
equals sum of percentage errors

P = EI
= (E ± ΔE) × (I ± ΔI)
= (E × I) ± (ΔE + ΔI)

R = (E ± ΔE) /(I ± ΔI)

= E/I ± (ΔE + ΔI)
18

www.utm.my innovative • entrepreneurial • global

 EXAMPLES
Absolute & Relative Error,

Example: A resistor with 820 ohm,10% accuracy & carries current of

10mA. An analog ammeter on a 25mA range measured the current with an
accuracy of 2% of full scale. Calculate the power dissipated in the resistor &
determine the accuracy of the result (% error).

P = I2R
= (I ± ΔI) × (I ± ΔI) × (R ± ΔR)
= (I2 × R) ± (ΔI + ΔI + ΔR)

|P|= (10 × 10-3)2 × 820

= 82 mW
19

www.utm.my innovative • entrepreneurial • global

 EXAMPLES
Absolute & Relative Error,

example: Calculate the power dissipated in the resistor & determine the
accuracy of the result (% error).

ErrorAbs from the Ammeter = ± 2% of 25 mA

= ± 0.5 mA

% Error in I10mA = (± 0.5/10 ) × 100 % = ± 5%

P = (I2 × R) ± (2 x ΔI + ΔR)
% Error in P = (2ΔI + ΔR)
= ± 10% + ±10 %
= ± 20% 20

www.utm.my innovative • entrepreneurial • global

 UNCERTAINTY
Uncertainty: measuring error exceeds some specific value
• uncertainty measurement
should include factors that
affect both the accuracy and
precision of the measurement

• accuracy is the closeness of

agreement between a measured
value and a true or accepted
value. Measurement error is the
amount of inaccuracy.

• precision is a measure of how well a result can be determined

(without reference to a theoretical or true value)
21

www.utm.my innovative • entrepreneurial • global

 UNCERTAINTY
Uncertainty for product function:

22

www.utm.my innovative • entrepreneurial • global

 UNCERTAINTY
Uncertainty: Example

Two resistors R and Rs are connected in series. The voltage

drops across each resistor are measured respectively as,

E = 10 V ± 0.1 V (1 %)
Es = 1.2 V ± 0.005 V (0.467%)

Along with a value of Rs = 0.0066Ω ± 0.25%

Determine the power dissipated in resistor R and its

uncertainty

23

www.utm.my innovative • entrepreneurial • global

 UNCERTAINTY
Uncertainty:
Example:

24

www.utm.my innovative • entrepreneurial • global

 UNCERTAINTY
Uncertainty: to consider/include

• Uncertainty propagation
or (propagation of error) is the effect of variables'
uncertainties (or errors) on the uncertainty of a function
based on them
i.e y = f(x, x1, x2, x3, ……..xn)
where y is calculated result, x is independent measured value

Examples:
f = x.y ( Area of a rectangle )

f = p.cosq ( x-component of momentum )

f = x / t ( velocity )
25

www.utm.my innovative • entrepreneurial • global

 UNCERTAINTY
Uncertainty: to consider/include

• partial derivative of differentiating,

• to represent uncertainties

26

www.utm.my innovative • entrepreneurial • global

 Uncertainty for additive functions:

www.utm.my innovative • entrepreneurial • global

 Example:

Find RMS uncertainty for

Where
Ro = 10.0 ± 0.5 % is the resistance at 20οC
α = 0.002 C-1 ± 1% is the temperature coefficient of resistance
T = 40 ± 1οC

www.utm.my innovative • entrepreneurial • global

 RESULT / DATA ANALYSIS
Uncertainty: accuracy & precision

29

www.utm.my innovative • entrepreneurial • global

 ACCURACY AND PRECISION
Uncertainty: accuracy & precision

30

www.utm.my innovative • entrepreneurial • global

 DEVIATION
Uncertainty: accuracy, precision, deviation

31

www.utm.my innovative • entrepreneurial • global

 EXAMPLES
Uncertainty: accuracy, precision, deviation
Example: Calculate the standard deviation of a population, use
the ages of four young children: { 5, 6, 8, 9 }.

32

www.utm.my innovative • entrepreneurial • global

 EXAMPLES
Uncertainty: accuracy, precision, deviation – example cont….

33

www.utm.my innovative • entrepreneurial • global

 CHAUVENET CRITERION
Chauvenet’s criterion (analysis for data rejection)-
improve standard deviation

Chauvenet’s criterion is a means of assessing one piece

of experimental data/from a set of observations. To apply
Chauvenet’s criterion,
• First calculate the mean and standard deviation,σ
• Calculate the ratio of di/σ for each data
• Refer to Chauvenet's criteria for data rejection.
• Re-calculate mean and standard deviation after data
rejection

34

www.utm.my innovative • entrepreneurial • global

 CHAUVENET CRITERION
Chauvenet’s criterion: example

35

www.utm.my innovative • entrepreneurial • global

 CHAUVENET CRITERION
Chauvenet’s criterion: example

36

www.utm.my innovative • entrepreneurial • global

 CHAUVENET CRITERION
Chauvenet’s criterion: example

37

www.utm.my innovative • entrepreneurial • global

 Chauvenet’s criterion: example
reading x di=dx-mean di/σ reading x di=dx-mean di/σ
1 5.3 -0.313 -0.678110918
1 5.3 -0.313 -0.499538051

2 5.73 0.117 0.186728281 2 5.73 0.117 0.253479161

3 6.77 1.157 1.846535225 3 6.77 1.157 2.506627261

4 5.26 -0.353 -0.764770461

4 5.26 -0.353 -0.56337678

5 4.33 -1.283 -2.04762722

6 5.45 -0.163 -0.260142819 6 5.45 -0.163 -0.353137635

7 6.09 0.477 0.761276839

7 6.09 0.477 1.033415042
8 5.64 0.027 0.043091142
8 5.64 0.027 0.058495191
9 5.81 0.197 0.314405738
9 5.81 0.197 0.426798246

10 5.75 0.137 0.218647645 10 5.75 0.137 0.296808932

mean stddev mean stddev

5.61 0.626578895 5.76 0.461576405

www.utm.my innovative • entrepreneurial • global

 Chauvenet's criterion for rejecting a reading
Number of readings, n ration of maximum acceptable
deviation to standard deviation
3 1.38
4 1.54
5 1.65
6 1.73
7 1.8
10 1.96
15 2.13
25 2.33
50 2.57
100 2.81
300 3.14
500 3.29
1000 3.48
www.utm.my innovative • entrepreneurial • global
 DATA PLOTTING
Plot / Plotting

• efficient ways to analyse data

• can directly see/check the

consistency of data

• type of function to represent

relationship between x-y axis

40

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting: presenting tabular data

41

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting: presenting tabular data

• include title presenting data

• each column in table refer to

measurement of one quantity

• each column be titled that identify value

• unit stated on top of column

• horizontal or vertical to separate headings & column

• associated error with data included

42

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting: rules to present graphical data

• to include title presenting data

• x-y axis to be labeled by variable

& units

• reasonable number of point

marked on axis

• data within the boundaries of

min & max graph

43

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting:

linear graph semi-log graph

44

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting:

log-log graph polar graph

45

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting: straight lines

46

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting: straight lines

47

www.utm.my innovative • entrepreneurial • global

 DATA PLOTTING
Plot / Plotting: Identifying data

48

www.utm.my innovative • entrepreneurial • global

 LEAST SQUARE METHOD
Plot / Plotting: Least Square method

49

www.utm.my innovative • entrepreneurial • global

 LEAST SQUARE METHOD
Plot / Plotting: Least Square method - example

50

www.utm.my innovative • entrepreneurial • global

 LEAST SQUARE METHOD
Plot / Plotting: Least Square method - example

Working solution steps:

• define equation of linear function (y = mx + b)

• sum of and

• define n (no. of data meas.)

• calculate m (slope) and

• complete equation parameters

51

www.utm.my innovative • entrepreneurial • global

 LEAST SQUARE METHOD
Plot / Plotting: Least Square method – example

n=5

52

www.utm.my innovative • entrepreneurial • global

 LEAST SQUARE METHOD
Plot / Plotting: Least Square method – example

53

www.utm.my innovative • entrepreneurial • global

 Correlation coefficient

r = ±1 (correlated)
r = 0 (not correlated)

www.utm.my innovative • entrepreneurial • global

 CORRELLATION

• Correlation exists when the data being measured is

dependent on some other quantity and is directly
related to that quantity.
– E.g: temperature increases and the output from a DC
power supply also increases, thus the output is
dependent on the ambient temperature.
• Signal-to-noise ratio (SNR) is a measurement of the
amount of true signal is transmitted compared to the
amount of garbage that is transmitted. The goal is to
have infinite SNR.

55

www.utm.my innovative • entrepreneurial • global