CLASSIFICATION OF ERRORS

Classification of Error
All measurement can be made without perfect accuracy (degree of error must
always be assumed). In reality, no measurement can ever made with 100%
accuracy. It is important to find that actual accuracy and different types of errors
can be occurred in measuring instruments.
a) Gross Errors
b) Systematic Errors
i. Instrumental errors
 Inherent shortcomings of instruments
 Misuse of instruments
 Loading effects
ii. Environmental errors
iii. Observational errors
c) Random Errors
 MEASUREMENT SYSTEM
1. Gross Errors
The main source of Gross errors is human mistakes in reading or using instruments and
in recording and calculating measured quantity. As long as human beings are involved
and they may grossly misread the scale reading, then definitely some gross errors will
be occurred in measured value.

Example, Due to an oversight, Experimenter may read the temperature as 22.7℃ while
the actual reading may be 32.7℃. He may transpose the reading while recording. For
example, he may read 16.7℃ and record 27.6℃ as an alternative.

The complete elimination of gross errors is maybe impossible, one should try to predict
and correct them.
Some gross errors are easily identified while others may be very difficult to detect.
Gross errors can be avoided by using the following two ways. Great care should be
taken in reading and recording the data.
 MEASUREMENT SYSTEM
2. Systematic Errors
Systematic errors are divided into following three categories.
i. Instrumental Errors
ii. Environmental Errors
iii. Observational Errors

These errors are arises due to following three reasons (sources of error).
a) Due to inherent shortcoming of instrument
b) Due to misuse of the instruments,
c) Due to loading effects of instruments
 MEASUREMENT SYSTEM
a) Inherent Shortcomings of instruments
These errors are inherent in instruments because of their mechanical structure due
to construction, calibration or operation of the instruments or measuring devices.
These errors may cause the instrument to read too low or too high. Example, if the
spring (used for producing controlling torque) of a permanent magnet instrument
has become weak, so the instrument will always read high. Errors may be caused
because of friction, hysteresis or even gear backlash.

Elimination or reduction methods of these errors,

o The instrument may be re-calibrated carefully.
o The procedure of measurement must be carefully planned. Substitution methods
or calibration against standards may be used for the purpose.
o Correction factors should be applied after determining the instrumental errors.
 MEASUREMENT SYSTEM
b) Misuse of Instruments
In some cases the errors are occurred in measurement due to the fault of the
operator than that of the instrument. A good instrument used in an unintelligent way
may give wrong results.
Examples, Misuse of instruments may be failure to do zero adjustment of instrument,
poor initial adjustments, using leads of too high a resistance and ill practices of
instrument beyond the manufacturer’s instruction and specifications

c) Loading Effects
The errors committed by loading effects due to improper use of an instrument for
measurement work. In measurement system, loading effects are identified and
corrections should be made or more suitable instruments can be used.
Example, a well calibrated voltmeter may give a misleading (may be false) voltage
reading when connected across a high resistance circuit. The same voltmeter, when
connected across a low resistance circuit may give a more reliable reading
(dependable or steady or true value)
 MEASUREMENT SYSTEM
b) Environmental Error
Environmental error occurs due to external environmental conditions of the
instrument, such as effects of temperature, pressure, humidity, dust, vibration or
external magnetic or electrostatic fields.
Elimination or reduction methods of these undesirable errors are
• Arrangements should be made to keep the conditions as nearly as constant as
possible. Example, temperature can be kept constant by keeping the instrument
in the temperature controlled region.
• The device which is used against these environmental effects. Example, variations
in resistance with temperature can be minimized by using very low resistance
temperature co-efficient of resistive material.
• Employing techniques which eliminate the effects of these disturbances. For
example, the effect of humidity dust etc., can be entirely eliminated by tightly
sealing the equipment.
 MEASUREMENT SYSTEM
b) Environmental Error
• The external or electrostatic effects can be eliminated by using magnetic or
electrostatic shield on the instrument.
• Applying computed corrections: Efforts are normally made to avoid the use of
application of computed corrections, but where these corrections are needed and are
necessary, they are incorporated for the computations of the results.

Observational Errors
There are many sources of observational errors. As an example, the pointer of a
voltmeter rests slightly above the surface of the scale. Thus an error on account of
PARALLAX will be acquired unless the line of vision of the observer is exactly above the
pointer.
 MEASUREMENT SYSTEM
Observational Errors

When the pointer’s image appears hidden by the pointer, observer’s eye is directly in
line with the pointer. Although a mirrored scale minimizes parallax error.
 MEASUREMENT SYSTEM
Random Errors
These errors are occurred due to unknown causes and are observed when the
magnitude and polarity of a measurement fluctuate in changeable (random) manner.
The quantity being measure is affected by many happenings or disturbances and
ambient influence about which we are unaware are lumped together and called as
Random or Residual. The errors caused by these disturbances are called Random
Errors. Since the errors remain even after the systematic errors have been taken care,
those errors are called as Residual (Random) Errors. Random errors cannot normally
be predicted or corrected, but they can be minimized by skilled observer and using a
well maintained quality instrument.
 MEASUREMENT SYSTEM
SOURCES OF ERRORS
The sources of error, other than the inability of a piece of hardware to provide a true
measurement are listed below
• Insufficient knowledge of process parameters and design conditions.
• Poor design
• Change in process parameters, irregularities, upsets (disturbances) etc.
• Poor maintenance
• Errors caused by people who operate the instrument or equipment. Certain design
limitations.
 MEASUREMENT SYSTEM
Errors in Measuring Instruments
• True value
The true value of quantity being measured is defined as the average of an infinite
number of measured values when the average deviation due to the various
contributing factors tends to zero.In ideal situation is not possible to determine the
True value of a quantity by experimental way. Normally an experimenter would never
know that the quantity being measured by experimental way is the True value of the
quantity or not.In practice the true value would be determined by a “standard
method”, that is a method agreed by experts with sufficient accurate.

• Static Error
Static error 𝛿𝐴 is defined as a difference between the measured value 𝐴𝑚 and the
true value 𝐴𝑚 of the quantity being measured. It is expressed as follows.
𝛿𝐴 = 𝐴𝑚 − 𝐴𝑡
Where 𝜖0 = 𝛿𝐴 is also called as absolute static error of quantity A
 MEASUREMENT SYSTEM
Errors in Measuring Instruments
• Relative static error
Relative static error is defined as the ratio between the absolute static errors and true
value of quantity being measured.
 MEASUREMENT SYSTEM
Errors in Measuring Instruments
• Relative static error
𝜖0 = 𝛿𝐴 is small, which means that the difference between measured value and true
values is very small, 𝛿𝐴 = 𝐴𝑚 − 𝐴𝑡 Negligible or small. So Almost
 MEASUREMENT SYSTEM
Errors in Measuring Instruments

• ERROR CORRECTION OR METHOD OF CORRECTION

It is the difference between the true value and the measured value of quantity
𝛿𝐶 = 𝐴𝑡 − 𝐴𝑚 = −𝛿𝐴

Note: Rectification or Elimination or Reduction) methods of all categories of errors are

dependent on the correction/reduction and elimination of errors as classified.
 ANALYSIS OF DATA
Data analysis

• Analysis of data is a process of inspecting, cleaning, transforming, and modeling

data with the goal of discovering useful information, suggesting conclusions, and
supporting decision-making.
• Analysis refers to breaking a whole into its separate components for individual
examination. Data analysis is a process for obtaining raw data and converting it into
information useful for decision-making by users.

Statistical Analysis of measurement data

Statistical Evaluation of measured data is obtained in two methods of tests
A. Multi Sample Test: In multi sample test, repeated measured data have been
acquired by different instruments, different methods of measurement and different
observer.
B. Single Sample Test: measured data have been acquired by identical conditions (same
instrument, methods and observer) at different times.
 ANALYSIS OF DATA
Statistical Evaluation methods will give the most probable true value of measured
quantity.
The mathematical background statistical evaluation methods are
• Arithmetic Mean
• Deviation
• Average Deviation
• Standard Deviation
• Variance.
1. Arithmetic Mean
Arithmetic mean or average of measured variables X is calculated by taking the sum of
all readings and dividing by the number of reading.
𝑥1 + 𝑥2 + 𝑥3 + ⋯ + 𝑥𝑛 𝑥𝑛
𝐴𝑟𝑖𝑡ℎ𝑒𝑚𝑒𝑡𝑖𝑐 𝑀𝑒𝑎𝑛 = =
𝑛 𝑛
𝑤ℎ𝑒𝑟𝑒 𝑥𝑖 = 𝑚𝑒𝑎𝑠𝑢𝑟𝑒𝑑 𝑣𝑎𝑙𝑢𝑒𝑠
 ANALYSIS OF DATA
2. Deviation (Deviation from the Average value)
The Deviation is departure of the observed reading from the arithmetic mean of the
group of reading. Let the deviation of reading 𝑥1 be 𝑑1 and that of 𝑥2 be 𝑑2 etc.,

3. Average Deviation:
Average deviation defined as the average of the modulus (without respect to its sign) of
the individual deviations.
The average deviation is used to identify precision of the instruments which is used in
making measurements.
Highly precise instruments will give a low average deviation between readings.
 ANALYSIS OF DATA
4. Standard Deviation
Standard deviation is used to analyze random errors occurred in measurement.
The standard Deviation of an infinite number of data is defined as the square root of the
sum of individual deviations squared, divided by the number of readings

𝑑1 2 + 𝑑2 2 + 𝑑3 2 … + 𝑑𝑛 2 𝑑𝑛 2
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛, 𝜎 = = ; 𝑓𝑜𝑟 𝑛 > 20
𝑛 𝑛

𝑑1 2 + 𝑑2 2 + 𝑑3 2 … + 𝑑𝑛 2 𝑑𝑛 2
𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑑𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛, 𝜎 = = ; 𝑓𝑜𝑟 𝑛 < 20
𝑛−1 𝑛−1
 ANALYSIS OF DATA
4. variance
The variance is the mean square deviation, which is the same as S.D except Square root.
Variance is Just the squared standard deviation.
𝑑1 2 +𝑑2 2 +𝑑3 2 …+𝑑𝑛 2 𝑑𝑛 2
𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑉 = = ; 𝑓𝑜𝑟 𝑛 > 20
𝑛 𝑛

𝑑1 2 + 𝑑2 2 + 𝑑3 2 … + 𝑑𝑛 2 𝑑𝑛 2
𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑉 = = ; 𝑓𝑜𝑟 𝑛 < 20
𝑛−1 𝑛−1
 ANALYSIS OF DATA
5. Histogram:
A histogram is also called a frequency distribution curve.
Example: Following table shows a set of 50 readings of length measurement. The most
probable or central value of length is 100mm represented as shown.

This histogram indicates the number of occurrence of particular

value. At the central value of 100mm is occurred 19 times and
recorded to the nearest 0.1mm as shown in figure 3.3. Here bell
shape dotted line curve is called as normal or Gaussian curve.
 ANALYSIS OF DATA
6. Measure of Dispersion from the Mean:
The property which denotes the extent to which the values are dispersed about the
central value is termed as dispersion. The other name of dispersion is spread or scatter.
Measure of dispersion from central value is an indication of the degree of consistency
(precision) and regularity of the data.
Example: the Figure shows the two sets of data and curve 1 vary from 𝑥1 to 𝑥2 and
curve 2 vary from 𝑥3 to 𝑥4 . Curve 1 is having smaller dispersion from central value than
the curve 2. Therefore curve 1 is having greater precision than the curve 2.