ERROR IN MEASUREMENT

A. ABSOLUTE ERROR
Any measurement made with a measuring device is approximate. If you measure the
same object two different times, the two measurements may not be exactly the same. 
The difference between two measurements is called a variation in the measurements.
Another word for this variation - or uncertainty in measurement - is "error."  This
"error" is not the same as a "mistake."  It does not mean that you got the wrong
answer.  The error in measurement is a mathematical way to show the uncertainty in
the measurement.  It is the difference between the result of the measurement and the
true value of what you were measuring. 
Error in measurement may be represented by the actual amount of error, or by a ratio
comparing the error to the size of the measurement. The absolute error of the
measurement shows how large the error actually is, while the relative error of the
measurement shows how large the error is in relation to the correct value.
Absolute errors do not always give an indication of how important the error may be. 
If you are measuring a football field and the absolute error is 1 cm, the error is
virtually irrelevant.  But, if you are measuring a small machine part (< 3cm), an
absolute error of 1 cm is very significant.  While both situations show an absolute
error of 1 cm., the relevance of the error is very different.  For this reason, it is more
useful to express error as a relative error.  We will be working with relative error.

B. PERCENTAGE OF ERROR/ RELATIVE ERROR

Relative error gives an indication of how good a measurement is relative to the size of the
thing being measured. Let’s say that two students measure two 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). Clearly, the overall accuracy of the ceiling height is much better than that of the 7.5
cm cylinder. The comparative accuracy of these measurements can be determined by looking
at their relative errors.

Or in plain English:

Error in measurement may also be expressed as a percent of error. The

percent of error is found by multiplying the relative error by 100%.
C. STATIC ERROR
The static error of a measuring instrument is the numerical difference between the true value
of a quantity and its value as obtained by measurement, i.e. repeated measurement of the
same quantity gives different indications. Types of Static error are categorized as gross errors
or human errors, systematic errors, and random errors.
 Gross Error
These errors are mainly due to human mistakes in reading or in using
instruments or errors in recording observations. Errors may also occur due to
incorrect adjustment of instruments and computational mistakes. These errors
cannot be treated mathematically.
The complete elimination of gross errors is not possible, but one can minimize
them. Some errors are easily detected while others may be elusive. One of the
basic gross errors that occurs frequently is the improper use of an instrument.
The error can be minimized by taking proper care in reading and recording the
measurement parameter.
In general, indicating instruments change ambient conditions to some extent
when connected into a complete circuit. (Refer Examples 1.3(a) and (b)). One
should therefore not be completely dependent on one reading only; at least
three separate readings should be taken, preferably under conditions in which
instruments are switched off and on.)
 Systematic Error
These errors occur due to shortcomings of the instrument, such as defective or
worn parts, or ageing or effects of the environment on the instrument.
These errors are sometimes referred to as bias, and they influence all
measurements of a quantity alike. A constant uniform deviation of the
operation of an instrument is known as a systematic error. There are basically
three types of systematic errors:
1. Instrumental error
Instrumental errors are inherent in measuring instruments, because of their
mechanical structure. For example, in the D’Arsonval movement, friction in
the bearings of various moving components, irregular spring tensions,
stretching of the spring, or reduction in tension due to improper handling or
overloading of the instrument.
Instrumental errors can be avoided by
(a) selecting a suitable instrument for the particular measurement applications.
(b) applying correction factors after determining the amount of instrumental
error.
(c) calibrating the instrument against a standard.
2. Environmental Error
Environmental errors are due to conditions external to the measuring
device, including conditions in the area surrounding the instrument, such
as the effects of change in temperature, humidity, barometric pressure or
of magnetic or electrostatic fields.
 These errors can also be avoided by (i) air conditioning, (ii) hermetically
sealing certain components in the instruments, and (iii) using magnetic
shields.
3. Observational Errors
Observational errors are errors introduced by the observer. The most
common error is the parallax error introduced in reading a meter scale, and
the error of estimation when obtaining a reading from a meter scale.
These errors are caused by the habits of individual observers. For example,
an observer may always introduce an error by consistently holding his
head too far to the left while reading a needle and scale reading.
In general, systematic errors can also be subdivided into static and
dynamic errors. Static errors are caused by limitations of the measuring
device or the physical laws governing its behavior. Dynamic errors are
caused by the instrument not responding fast enough to follow the changes
in a measured variable.
 Random Error
A random error, as the name suggests, is random in nature and very difficult to
predict. It occurs because there are a very large number of parameters beyond
the control of the experimenter that may interfere with the results of the
experiment. Random errors are caused by sources that are not immediately
obvious and it may take a long time trying to figure out the source.
Random error is also called as statistical error because it can be gotten rid of in
a measurement by statistical means because it is random in nature. Unlike in
the case of systematic errors, simple averaging out of various measurements of
the same quantity can help offset random errors. Random errors can seldom be
understood and are never fixed in nature - like being proportional to the
measured quantity or being constant over many measurements.
The reason why random errors can be taken care of by averaging is that they
have a zero-expected value, which means they are truly random and scattered
around the mean value. This also means that the arithmetic mean of the errors
is expected to be zero. There can be a number of possible sources of random
errors and their source depends on the type of experiment and the types of
measuring instruments being used.
A random error can also occur due to the measuring instrument and the way it
is affected by changes in the surroundings. For example, a spring balance
might show some variation in measurement due to fluctuations in temperature,
conditions of loading and unloading, etc. A measuring instrument with a
higher precision means there will be lesser fluctuations in its measurement.
Random errors are present in all experiments and therefore the researcher
should be prepared for them. Unlike systematic errors, random errors are not
predictable, which makes them difficult to detect but easier to remove since
they are statistical errors and can be removed by statistical methods like
averaging.
D. Based on the quantification
Based on the quantification, error can be separate into 3 types of error :
 Limiting Error
The deviations from the nominal value of a particular quantity is the Limiting
error or Guarantee error. This error is specified by the manufacturer. The
nominal value, As, and maximum or limiting error is ±δA Actual value, Aa =
As ±δA Ex: The nominal magnitude of resistor is 100 Ω and limiting error
±10Ω Actual value is 90 6 Aa 6 110Ω. Relative (Fractional) Limiting error is
defined as the ratio of the error to the specified (nominal) magnitude of a
quantity. Therefore, relative limiting error.
 Parallax error
A parallax error is an error in reading an instrument due to the eye of the
observer and pointer are not in a line perpendicular to the plane of the scale.

 Quantifization Error
Error resulting from trying to represent a continuous analog signal with discrete,
stepped digital data. The problem arises when the analog value being sampled falls
between two digital “steps.” When this happens, the analog value must be represented
by the nearest digital value, resulting in a very slight error. In other words, the
difference between the continuous analog waveform, and the stair-stepped digital
representation is quantization error. For a sine wave, quantization error will appear as
extra harmonics in the signal. For music or program material, the signal is constantly
changing and quantization error appears as wideband noise, cleverly referred to as
“quantization noise.” It is extremely difficult to measure or spec quantization noise,
since it only exists when a signal is present.
 Quantization error is one reason higher digital resolutions (longer word lengths) and
higher sample rates sound better to our ears; the “steps” become finer, reducing
quantization errors.

E. THE WAY TO IMPROVE ACCURACY IN MEASUREMENTS

1. Make the measurement with an instrument that has the highest level of precision. The
smaller the unit, or fraction of a unit, on the measuring device, the more precisely the device
can measure. The precision of a measuring instrument is determined by the smallest unit to
which it can measure.

2. Know your tools! Apply correct techniques when using the measuring instrument and
reading the value measured. Avoid the error called "parallax" -- always take readings by
looking straight down (or ahead) at the measuring device. Looking at the measuring device
from a left or right angle will give an incorrect value.

3. Repeat the same measure several times to get a good average value.

4. Measure under controlled conditions. If the object you are measuring could change size
depending upon climatic conditions (swell or shrink), be sure to measure it under the same
conditions each time. This may apply to your measuring instruments as well.