Measurement, Accuracy and Errors

Medir es comparar una magnitud desconocida con un patrón que tomamos como referencia y que llamamos unidad. El resultado de la medida es el producto de un número real por la unidad de referencia.

Hay unidades fundamentales que se eligen arbitrariamente y derivadas que se definen a partir de las fundamentales.

El sistema internacional de unidades (SI) se elabora por la Conferencia General de Pesas y Medidas, que se reune para corregirlo cada cuatro años.

El SI consta de 7 unidades fundamentales, las unidades 2 suplementarias y las unidades derivadas.

Copyright© by TutorVista.com.

[Open questions]

[Edit this page]


In no subject does measurement play as important a role as in science. Real science cannot exist without measurement. According to Lord Kelvin, one of the greatest scientists, unless you can measure what you are speaking about and express it in numbers you have not started 'Exact Science'.

Thus there are two aspects regarding measurements. One is actual MEASUREMENT of objects or happenings and the second is expressing that in terms of NUMBERS.

Physics is an exact science. It deals with accurate measurements. But the question comes up 'How much accuracy do we need to make our measurements accurate?'

The answer depends upon the purpose for which measurements are made. To know exact time we take to reach the school we may need to know the number of minutes we take to reach the school. But Olympic records need timings in fractions of seconds! For us the year is equal to 365 or 365.25 days. And it is acceptable. But ask an astronomer. He will say that it is equal to 365.242195 days! But apart from such astronomers, for whom is such accuracy important?

Everything that we use in our daily life is ultimately governed by principles of physics. All gadgets we use everyday at home, bicycles and cars, all different types of machinery and instruments, work on principles of physics. Hence to understand even the elementary working of these things, the study of physics is essential.

For understanding the relationships between matter and energy, measuring them is very essential. There are thousands and thousands of different things around us. There are different kinds of forces around us and there are different kinds of energies that we come across every day. Thus there would be millions and millions of physical quantities and energies that could be measured. Yet you would be surprised to know that there are only six basic units from which all other units are derived. These basic units are the units of length, mass, time, electric current, temperature and luminous intensity. Out of these only three fundamental or basic units of measurements are used in mechanics.

These are the units length, mass and time. All the measurements in physics are related to these three fundamental units. All other units of measurements can be derived from these three fundamental units.

While discussing these units of measurement we must be clear about two terms - unit and measure number. For example, when we say that the length of a line is five centimeters, the unit we have chosen is centimeter and the measure number is five as the line is five times the unit length.

Quantitative vs. Qualitative

Most experiments in physics require the observations made to be quantitative rather than qualitative. If observations are only descriptive or qualitative, they are likely to be imprecise and could cause disagreements between experimenters. For example, scientists cannot merely say that an object is large or small. Instead they have to specify its size as a quantity, that is, with a number and using a standard unit such as kilogram. This is called a quantitative observation.

As science advanced through the centuries, the importance of accurate and uniform measurements was realized all over the world. In order to enable scientists working in different parts of the world to compare their measurements, a need for certain basic unit of measurement was felt and hence later a basic unit was defined.

Unit is a standard for comparison. In earlier times the measurement of quantity of things was quite arbitrary. In many cases it was related to the dimension of different parts of the human body. These parts were chosen as "units" to measure these quantities. For example, for measuring length, distance between the nose and the fingers or outstretched hand was used as a unit. Can you imagine the confusion caused when different countries used different units or measures! This also caused a lot of inconvenience.

Estimation

It is easy to measure the length of your page but if you are asked to find the height of a high rise building how would you do it? It is of course possible to do that by going to the terrace of the building, hanging a rope… etc. Not very convenient!

On the other hand one can estimate approximate height of a floor and then do some simple arithmetic to arrive at a figure which, though not accurate from the point of view of an architect, may serve our purpose. Thus even if accurate measurement is necessary it is always advisable to estimate. This would help; you to avoid silly mistakes that frequently take place while calculating.

For better estimations, specially for large numbers, comparison is easier to make. For example in the above calculations it would be easier to compare the height of the high rise building in terms of a building with two or three storeys.

Approximation

In measurements approximation also plays an important role. In day to day practice we always use approximation. You must have heard people saying "It is approximately five minutes walk from the station" or "Both of them are more or less of the same height" etc. Approximations also play an important role in science. To give you some idea about the size of an atom your chemistry teacher would say, "In a centimetre, there might be approximately 100 million atoms lying side by side!"

Accuracy

It is the smallest reading that can be accurately measured while using an instrument or a device. For example the least count of various measuring devices are listed below:

Device Least count
protractor  1°
clock 1 second
Ruler 1 mm
Thermometer
Spring Balance 1 gram

Significant Figures along with Examples

These express the degree of accuracy of measurements. It is a statement which gives number of digits up to which we are sure about their accuracy. It gives the degree of accuracy or precision made with the instrument. In practical life we depend only on approximate measurements. We ignore small measurements when we are computing large measurements. For example, we may measure the length of a wall as 10 meters and 57 centimeters or 10.57 meters. The actual length of the wall is between 10 meters 57.5 cm and 10 meters 56.5 cm. Now we can say that the length 10.57 meters is correct up to four significant figures.

Measuring instruments have a limit upto which measurements can be made, called as the least count of the instrument. Errors are seen when measuring through various instruments. For e.g., accuracy of the temperature measured by thermometer depends on the thermometer and the measuring person. If the temperature of hot water is 99.9o C and if the thermometer gives exactly 99.9o C then the measured value is accurate, or else it is an error.

Accuracy depends on

Sub Topics

 

nt o

 

Hence accuracy is the extent to which the measured value agrees to the true value for measurement. Precision is the extent to which a given set of measurements of the same quantity agree with their mean value. Precision measurement need not be accurate. Also accuracy depends on systematic errors while precision depends on random errors.

 

Significant Figures is the expression of accuracy of a physical quantity. The value in digits, accurately known in a measurement plus one digit that is not certain, is called significant figures. Significant figures is directly proportional to the accuracy of measurement. Eg., a reading of 6.34 is accurate to 3 significant figures, while a figure of 6.342 is accurate to 4 significant figures. Thus, significant figures are all digits about which we are determining significant figures.

Eg., 256, 34 has 5 significant figures.

Eg., 567.003 has 6 significant figures.

Eg., In 1995.00 there are 6 significant figures

In 0.019940, there are 5 significant figures.

Eg., In 199500, the significant figure is 4.

Sub Topics

 

Rounding off

The omission of the last digit, which is not required for correctness of a physical quantity, with least deviation from its original value, is called rounding off.

One or two rules has to be followed in rounding off, like

Eg.,

a) When 6.83 m is rounded off to first decimal point then 3 is dropped. As 3<5 and 8 is retained, i.e., 6.83 m when rounded off, becomes 6.8 m.

b) When 0.009037 m is rounded off to 2 significant figures, we get 0.0090 m.

Eg., 5.76kg when rounded off to first decimal place, becomes 5.8 kg.

Eg., When 8.745 kg is rounded off to 3 significant figures, we get 8.74 g.

 

 

Example 1:

(a) 8.88 correct to two significant figures is 8.9, because 8.88 is nearer to 8.9 than to 8.8.

(b) On the other hand 8.82 correct to two significant figures is 8.8. This is because 8.82 is nearer to 8.8 than to 8.9.

Example 2:

Correct the number 8.5775

(a) up to 2 significant figures = 8.6

(b) up to 3 significant figures = 8.58

(c) up to 4 significant figures - 8.578

Example 3:

Suppose you are measuring the diameter of a cylinder using a vernier calliper as 2.38 cm. The accurate value may lie between 2.375 cm and 2.385 cm. In this case figures 2 and 3 are absolutely correct while 8 is reasonably correct. This measurement is said to be accurate up to 3 significant figures.

Rules for Determining Significant Digits

 

Arithmetic Operations With Significant Figures

The accuracy of a result is taken to be equal to the least accurate among the numbers, when 2 or more numbers are used to add, subtract, multiply or divide. The number of significant figures in the result, is equal to the number of significant digits in the least accurate one among them.

Sub Topics

 

Errors in measurement

When we make use of various measuring instruments, we encounter various types of errors. They are:

Constant errors

If the same error is repeated every time in a series of observations, the error is said to be constant error.

Constant error is due to faulty calibration of the scale of a measuring instrument.

In order to minimise the constant error, measurements are made with all possible methods.

Relative and percentage errors

Relative error is the ratio of absolute error to true value given by

Absolute errors

It is the magnitude of difference between the true value and the measured value.

Absolute error = True value - measured value

If arithmetic mean is regarded as true value, absolute error in ith measurement is given by

It is expressed in the units of measured value. Final absolute error is taken as the arithmetic mean of the absolute error in various measurements.

Considering a single measurement, the result of measurement will lie between

Gross errors

This error occurs due to a person's mistake while performing the experiment. It cannot be corrected. It is of three types:-

  • Neglect of the source of error: This is a result of negligence of sources of errors.

E.g., plotting field of a magnet, improper setting of magnet along NS line.

  • Reading the instrument incorrectly: Sometimes 1 ampere might have been divided in 20 or 5 parts instead of 10. The experimenter may read the instrument without paying attention to the value of 1 division.
  • Improper recording of reading: This mistake is done by the person performing the experiment when he records the reading incorrectly. Eg., He may record 21.3 in place of 23.1.

Random errors

These are irregularly occurring errors, which are at random, in magnitude and direction. They are not due to any definite cause and hence, are called accidental errors. A person may get different readings while experimenting, say, for E.g., while measuring the diameter of a wire with a screw gauge. This happens due to many reasons. Therefore, if the observation is repeated a number of times, the arithmetic mean of all readings is found to be very close to the most accurate reading. Therefore, in an experiment, it is advisable to take readings, a number of times and then take their arithmetic mean.

If a1, a2 ………….. an are n different readings in an experiment, their arithmetic mean is given by

Systematic errors

These errors always have the same sign. These can be eliminated by detecting the source of error and the rule governing this error. They are of 4 types:

Instrumental errors

These are inherent errors of the measuring instruments and apparatus. E.g., zero error of a measuring instrument. All instrumental errors come under this category. Instrumental errors, if any, can be detected by interchanging two similar instruments or by using different methods for measuring the same physical quantity.

Errors due to external causes

These are caused by external conditions (wind, temperature, etc.,). These errors can be prevented by applying suitable corrections.

Errors due to imperfection

Sometimes, even if the error is known, it cannot be corrected due to imperfection in the experimental arrangement.

e.g., Loss of heat due to radiation in calorimetry.

These errors always exist, but observations can be corrected.

Observational errors

This arises from the mode of observation of the person, taking the reading. e.g., Parallax error. These errors can be minimised by obtaining several readings and taking their arithmetic mean.

 

 

Errors

No measurement is ever perfectly accurate. Even with high precision instruments some error is inevitable.

There are two main types of errors:

Random errors

occur in all measurements. They arise when observers estimate the last figure of the reading on an instrument. These include the noise in the room or the mechanical vibrations in the room. These are called random, because they cannot be predicted. The best way of minimizing the error is to take the average of many readings.

Systematic errors

Such mistakes are not random, but constant. They may cause an experimenter to under estimate or over estimate a reading. Systematic errors may be due to defective equipment - for instance, an incorrectly marked ruler; or they may be due to environmental factors - for instance, the weather conditions on a particular day. While recording time using a stop-watch, your reaction time in starting or stopping the stop-watch will certainly vary at times significantly if you are tired or distracted. At times the variation will be more than a few hundredth of a second.

Percentage Error

While reading the length of a simple pendulum or the length of a resistance wire or while finding the weight of a body using spring balance, mass by a beam balance etc., we are likely to make mistakes. The percentage error can be calculated by using the formula.

Percentage error

For example, if the length of an object (100cm long) is measured as 99.8 cm, then

% error

Presentation of data in tabular and graphical form

Results and observations of scientific experiments should be properly recorded under headed columns and numbered rows. The headings must include the units in which each quantity is measured.

Graphs are very helpful in comparing measurements. The general rule in plotting the graph of a given data is to plot the independent variable on the horizontal axis (X-axis). The data, which changes, is plotted on the vertical axis (Y-axis). It is possible to use the graph to predict the reading of measurements that lie between those actually made. This is known as interpolation. If the pattern of the graph is extended beyond the observed data in either direction, a prediction may also be made of readings lying outside the observed data. Predicting readings by this method is called extrapolation.

Every graph should have the following:

  • a title (e.g., Load against extension)
  • both axes labeled with units
  • scales being used on each axis
  • points plotted with crosses or with dot and circle.

The plotted points must be joined with a single straight line or a continuous curve. The graph should cover as much of the area of the graph paper as possible. This requires a sensible choice of scale for each axis. It is unlikely that points plotted from real experimental results will all lie on a straight line or smooth curve due to errors. Hence try to produce a straight line or smooth curve which passes through as many of the plotted points as possible or which leaves an equal distribution of points on either side. Refer the graph shown in figure (a) above which shows a best fitting line.

Propagation of Errors

Final result of an experiment is calculated from a number of observations taken from different instruments, connected through a formula.

Sub Topics

 

Maximum permissible error in different cases is calculated as follows

Result involving sum of two observed quantities

X is the sum of 2 observed quantities a and b.

X = a + b

Maximum absolute error in X = Maximum absolute error in a + Maximum absolute error in b

Result involving difference of two observed quantities

Suppose X = a - b

Let Da and Db be absolute errors in measurements of quantities a and b, values of a and b and DX be maximum error in X.

Maximum absolute error in X = Maximum absolute error in a + Maximum absolute error in b

From equations (1) and (2) it is evident that, when result involves sum or difference of 2 observed quantities, absolute error is the sum of absolute errors in the observed quantities.

Result involving the product of two observed quantities

Suppose X = ab

Let Da and Db be absolute errors in measurements of quantities a and b, values of a and b and DX be the maximum possible error in X.

Dividing both sides by X = ab, we get

are relative errors of fractional errors in values of a, b and x. Neglecting  as its product is very small.

The above result is obtained by logarithmic differentiation.

Take log on both sides,

Log X = log a + log b

Differentiating, we get ,

Thus, maximum relative error in X = maximum relative error in a x maximum relative error in b

Maximum absolute error in X = Maximum absolute error in a + Maximum absolute error in b

Result involving quotient of 2 observed quantities

Let Da and Db be absolute errors in measurement of quantities a and b and DX be maximum possible error in X.

Maximum possible relative error in X,

Maximum relative error in X = maximum relative error in a + maximum relative error in b

Maximum percentage error in X,

i.e., Maximum percentage error in X = maximum percentage error in a + maximum percentage in b. From equations 3, 4, 5 and 6, it is seen that when the result involves the multiplication or quotient of 2 observed quantities, the maximum possible relative error in the result is equal to the sum of the relative errors in the observed quantities.

Result involving product of powers of observed quantities

Relative error in an is n times the relative error a

It can be proved that maximum relative error in X,

Also, maximum percentage error in X,

Maximum percentage error in X = l times maximum percentage error in a + m times maximum percentage error in b + n times maximum percentage error in c.

 

 

 

 

 

 

 

 

Summary

Accuracy

By using an appropriate device, we can know the measure of a physical quantity.The accuracy of the reading depends on the device and the person taking the measurement. Therefore, accuracy means the extent upto which a measured value agrees with the standard or true value. For example, if the temperature of water is 211.82oC and the thermometer reads exacts 211.82oC, then the measured value is accurate.

Least count

The accuracy of a measuring instrument depends upon the least count of the measuring instrument.

For example, length of a body that can be measured using a tailor's tape, cannot be beyond the minimum and maximum length of the measuring tape.

Least count of this tape is the least distance that can be accurately measured using this tape. Various types of measuring instruments like steel tape, ruler, vernier callipers, micrometer, screw gauge, etc., are more accurate compared to each other, i.e., the least count of a screw gauge is 0.01mm, the least count of a vernier callipers is 0.1mm. When we measure the diameter of the wire that is 1mm thick nearest to 0.01mm, using the screw gauge, the accuracy is about 0.01mm in 1.00mm. If this measurement is taken by using the vernier callipers, then the accuracy would have been 1 out of 10. Hence, the accuracy depends on the least count of the measuring instrument.

[Note: However, error cannot be avoided, in spite of using the right instrument]

Significant figures

The significant figures are all those digits which we are absolutely sure of plus those digits that we are not sure of. However, the digits have a meaningful value. This also provides us with information about the extent of uncertainty in measurement.

E.g.,

All the experiments give answer upto 3.14 and the rest of the digits vary due to change in the value adopted for circumference and diameter and hence, it is rounded off to 3 digits.

Sensitivity of the instrument

This is the instantaneous reaction of the instrument, which can indicate the measurement at the instance of use.

Hysteresis

The instrument must be free from any previous recording or reading, and should give the same value for ascending or descending readings.

Precision

The extent to which a given set of measurements of the same quantity, agree with their mean value, which will not be true value.

Errors

All physical quantities can be measured. In spite of using the best and most appropriate instrument or device to measure, we may not get the true value. Hence, the difference between the true value and the measured value is called 'error'. There are different types of errors that are encountered while taking measurements. They are

Systematic errors which are caused due

  • instrument error (manufacturing or inbuilt defect or limitation)
  • zero error of devices.
  • faulty calibration of meters
  • inequality of balance arms of physical balance
  • backlash error
  • end error

Environmental error

Change in temperature, pressure, humidity, magnetic field, wind, etc. Eg., Metal tape expands in summer and at noon and contracts in winter and during night.

Observation or human error

Parallax

Due to imperfection

Conducting experiments imperfectly, without observing all the conditions stipulated for the purpose. E.g., if the experiment is to be conducted in an air-conditioned atmosphere, if the air-conditioning system does not exist during the experiment, this error occurs.

Random errors

These are called accidental errors, which occur irregularly and at random, in magnitude and direction.

(When the experiment is repeated, one may obtain different results each time.)

Gross errors

This is caused due to the carelessness of the person taking measurement.

Absolute error

Absolute error = true value - measured value

Mean absolute error

The arithmetic mean of the absolute error of the different measurements taken is called MAE.

Where is the true value and is the absolute error.

Relative and percentage errors

The ratio of the maximum error to time value of the measured quantity.

Where arithmetic mean is taken as time value.


 


References and Suggestions for Further Reading

  1.  

Open Questions

  1.