ABHISEKH UMRAO SINGH

7810595
M.Sc BIOTECHNOLOGY

A REPORT
ON
ERROR ANALYSIS

ABHISEKH UMRAO SINGH

UMID: 7810595

M.Sc BIOTECHNOLOGY

Page | 1
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

“As all error is meanness,

it is incumbent on every
man who consults his own
dignity, to retract it as soon
as he discovers it”

SAMUEL JOHNSON

The Purpose of Error Analysis:

Now let’s see an example for instance a particle falling under the influence of gravity is subject to a
constant acceleration of 9.8 m/ . For a scientist performing an experiment this value/specification is not
complete. Does it mean that the acceleration is closer to 9.8 than to 9.9 or 9.7? Does it mean that the
acceleration is closer to 9.80000 than to 9.80001 or 9.79999? Often the answer depends on the context. If
a carpenter says a length is "just 8 inches" that probably means the length is closer to 8 0/16 in. then to 8
1/16 in. or 7 15/16 in. If a machinist says a length is "just 200 millimeters" that probably means it is
closer to 200.00 mm than to 200.05 mm or 199.95 mm. We all know that the acceleration due to gravity
varies from place to place on the earth's surface. It also varies with the height above the surface, and
gravity meters capable of measuring the variation from the floor to a roof are readily available. The value
of g can be determined by an experiment. Since we do not hold a perfect experiment to determine the
value of g it is impossible. But we use this value of g because to give it some meaning it must be changed
to something like a 5 g ball bearing falling under the influence of gravity in Room 126 of Morton
Laboratory of the University of Manchester on March 13, 1995 at a distance of 1.0 ± 0.1 m above the
floor was measured to be subject to a constant acceleration of 9.81 ± 0.03 m/ . Two questions arise
about the measurement. First, is it "accurate?" in other words, did the experiment work properly and were
all the necessary factors taken into account? The answer to this depends on the skill of the person
performing the experiment in identifying and eliminating all systematic errors. The second question
regards the "precision" of the experiment. In this case the precision of the result is given: the experimenter
claims the precision of the result is within 0.03 m/s . However, the following points are important:

1. The person who did the measurement probably had some "gut feeling" for the precision and "hung" an
error on the result primarily to communicate this feeling to other people. Common sense should always
take precedence over mathematical manipulations.

Page | 2
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

2. In complicated experiments, error analysis can identify dominant errors and hence provide a 3. There is
virtually no case in the experimental physical sciences where the correct error analysis is to compare the
result with a number in some book. A correct experiment is one that is performed correctly, not one that
gives a result in agreement with other measurements.

4. The best precision possible for a given experiment is always limited by the apparatus. Polarization
measurements in high-energy physics require tens of thousands of person-hours and cost hundreds of
thousands of dollars to perform, and a good measurement is within a factor of two. Electrodynamics
experiments are considerably cheaper, and often give results to 8 or more significant figures. In both
cases, the person must struggle with the equipments to get the most precise and accurate measurement as
far as possible.

THE UNCERTAINY OF MEASUREMENT:

Some numerical statements are exact like Dixie has 4 burgers or 4+6=10. However all measurements
have some or the other uncertainty originating from different variety of sources. The process of
measurement of these uncertainties associated with experiment/measurement result is called uncertainty
analysis or error analysis.

The complete statement of a measured value should include an estimate of the level of confidence
associated with value. Properly reporting an experiment result along with the experimental result along
with its uncertainty allows other people to make judgment about the quality of the experiment, and it
facilitates meaningful comparison with other similar values or a theoretical prediction. Without an
uncertainty estimate, it is impossible to answer the basic scientific question “does my result agree with a
theoretical prediction or results from other experiments?” This question is the fundamental for deciding if
a scientific hypothesis is confirmed or refuted.

When we make a measurement, we generally assume that some exact or the true value exists based on
how w define what is being measured. While we may never know the true value exactly, we attempt to
make to find this ideal quantity to the best of our ability with time and resources available. As we make
the measurement with different methods, or even making multiple measurements using the same method,
we may obtain slightly different result. So how do we report our findings for the best estimate of this
elusive true value? The most common way to show the range of values that we believe include the true
value is:

Page | 3
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

Measurements = Best Estimate ±Uncertainty (units)

Let’s take an example. Suppose you to find the mass of the gold chain that you would like to sell it to
your friend. You do not want to jeopardize your friendship, so you want to get an accurate mass of the
chain in order to charge a fair market price. By simply examining the ring in your hand, you estimate the
mass to be between 10 and 20 grams, but this is not precise estimate after some searching, you find an
electronic mass balance saying 17.43 grams. While this measurement is much more precise than the
original estimate, how do you know that it is accurate, and how confident are you that this measurement
represent the true value of the chain mass? Since the digital display of the balance is limited to 2 decimal
places, you could report the mass as m=17.49±0.01 g. suppose, you use the same electronic balance and
obtain several more you may feel confident that you may know the mass of the chain to the nearest
hundredth of a gram, but how do you know that the true value definitely lies between 17.43g and 17.45g?
Since you want to be honest, you decide to use another balance which give the reading of 17.22 g. this
value is clearly below the range of values found on the first balance, and under normal circumstances ,
you might not care , but you want to be fair to your friend. So what do you do now?

Now we need some terms to be defined to the answer the question:

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

Precision is a measure of how well result can be determined (without reference to a theoretical or true
value). It is the degree of consistency and agreement amoung independent measurements of the same
quantity; also reliability or reproducibility of the result.

The statement of uncertainty associated with a measurement should include factors that affect both the
accuracy and prediction of the measurement.

But error and uncertainty are often used interchangeably to describe both imprecision and inaccuracy.
This usage is so common that it is impossible to avoid. Whenever you encounter these terms, make sure
you understand they refer to accuracy or precision, or both.

Notice that in order to determine the accuracy of a particular measurement, we have to know the ideal,
true value, which we really never do. Sometimes we have a textbook measured value which is known as
precisely, and we assume that this is our ideal value, and use it to estimate the accuracy of our result.

Page | 4
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

Other times we know a theoretical value, which is calculated from basic principles; this may be taken
from as an ideal value. But physics and other fields are empirical sciences; theory must be validated by
experiments, not the other around. Some fields like biotechnology etc theory can be accidentally validated
by experiments. We can escape the difficulties and retain a useful definitions of accuracy by assuming
that, even when we do not know the true value, we can rely on the best available accepted value with to
compare our experimental value.

For example with the gold chain, there is no accepted value with which to compare, and both, measured
values have the same precision so we have no reason to believe one more than the other. The only way to
assess the accuracy of the measurement is to compare with a known standard. For this situation, it may
be possible to calibrate the balances with a standard mass that is accurate within narrow tolerance and is
traceable to a primary mass standard at the national institute of standard and technology. Calibrating the
balances should eliminate the discrepancy between the readings and provide a more accurate mass
measurement.

Precision is often reported quantitatively by using relative or fractional uncertainty.

 

Relative uncertainty==   

. 
For example, m= 75.5±0.5 has a fractional uncertainty of .  =0.00666=0.66 %.

Accuracy is often quantitatively by using relative error:

  

Relative error=  

..
If the expected value is 80.0 g, then the relative error is: .
=-0.056=-5.6%.

The minus sign indicates that the measurement value is less than expected value.

When analyzing experimental data, it is important that you understand the differences between precision
and accuracy. Precision indicates the quality of the measurement, without guarantee that the measurement
is correct. Accuracy on the other hand, assumes that there is ideal value, and tells us how far your answer
is from the ideal, right answer. These concepts are directly related to random and systematic measurement
errors.

Page | 5
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

Most them may know that most of the physical and other sciences are divided into two branches i.e.
theory and experiment. In general, the theoretical aspect is taught in lectures, tutorials and by doing
problems, while the experimental aspect is learned in the laboratory.

The way these two branches handle numerical data are significantly different. For example, here is a
problem from the end of a chapter of well-known first year university physics textbook:

“A particle falling under the action of gravity is subject to a constant acceleration g of 9.8 m/s2. If the…”

Although this first fragment is perfectly acceptable for doing problems, i.e. for learning theoretical
physical analytical science, in an experimental situation it is incomplete. Does it mean the acceleration is
closer to 9.8 than to 9.9 or 9.7? Does it mean that the acceleration is closer to 9.800000 than to 9.800001
or 9.799999? Often the answer depends on the context. If a carpenter says a length is just 8inches that
probably means the length is closer to 8 0/16 in. then to 8 1/16 in. or 7 15/16 in. if a machinist says the
length is just 200 mm that probably means it is closer to 200.00m than to 200.05 mm or 199.95 mm.

We emphasize that a correct experiment is one that has been correctly performed. Thus: The Error in an
experimentally measurement quantity is never found by comparing it to some number found in a book or
web page.

Also although we will be exploring mathematical and statistical procedures that are used to determine the
error in an experimentally measured quantity, as you will see these are often just rules of thumb and
sometimes good experimenter uses his or her intuition and common sense to simply guess.

MOTIVATION:

A lack of understanding of basic error analysis has led some very bright scientists to make some
incredible blunders. Here are some of the examples:

COLD FUSION

In 1989 two University of Utah researchers, Stanley Pons and Martin Fleischmann, announced that they
had produced nuclear fission with a laboratory bench apparatus consisting of Palladium rods immersed
in a bath of Deuterium, or heavy water. The scientists said their device emitted neutrons and gamma rays,
which are certain signatures of nuclear, as opposed to chemical reactions.

Page | 6
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

This announcement caused a huge reaction in the popular press, and there were many statements being
made this was the beginning of unlimited free energy.

The claim turned out to be wrong: cold fusion in this form does not exist. Amongst other mistakes, Pons
and Fleischmann neglected to do the simple error analysis on their results which would have shown that
they had not achieved any cold fusion in their lab.

HIGH FIBER DIETS

In the early 1970’s researchers reported that a diet which was high in fiber reduced the incidence of
polyps forming in the colon. These polyps are pre-cursor to cancer.

An accident reported at high fiber factory in England

In January 2000 a massive study published in the new England journal of medicine indicated that fiber in
the diet has no effect on the incidence of polyps.

The problem with the earlier studies is that the limited no of people ii the samples meant that the result
were statistically insignificant. Put another way, the error bars of measurements were so huge that the 2
samples, with and without high fiber diets, gave cancer rates that were numerically different but were the
same within errors.

Note the word statistically in the previous paragraph: it indicates correctly that some knowledge of
statistics will be necessary in our study of error analysis.

Page | 7
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

TYPES OF ERROR:

Measurement errors can be classified as either Random or systematic, depending on how the
measurement was obtained (an instrument could cause random error in one situation and a systematic
error in another).

What are random errors?

Random errors are statistically fluctuation (in either direction) in the measured data due to the precision
limitations of the measurement device. Random errors can be evaluated through statistical analysis and
can be reduced by averaging over a large no of observations (see standard error).

Then what are systematic errors?

Systematic errors are reproducible inaccuracies that are consistently in the same direction. These errors
are difficult to detect and cannot be analyzed statistically. If a systematic error is identified when
calibrating against a standard, applying a correction or a correction factor to compensate for the effect can
reduce the bias. Unlike random errors, systematic errors cannot be detected or reduced by increasing the
no of number of observations.

Our main strategy is to reduce as many sources of errors as we can, and then to keep track of those errors
that we can’t eliminate. It is useful to study the types of errors that may occur, so that we may recognize
them when they arise.

COMMON SOURCES OF ERROR IN LABORATORY EXPERIMENTS:

INCOMPLETE DEFINITIONS (may be both type of errors): one reason that it is impossible to
make exact measurements is that the measurement is not always properly defined. Now for example, if
some 2 different people measure the length of the same rope, they would probably get various results
because each people may extend the rope with different errors is to be carefully considered and specify
the conditions that could affect the measurement.

FAILURE TO ACCOUNT FOR A FACTOR (usually systematic): the most challenging part of
designing an experiment is trying to control or account for all possible factors except the one independent
variable that is being analyzed. For instance, you may ignore some air resistance when measuring free-fall
acceleration or you may fail account for the effect of the earth’s magnetic field when measuring the field
of a small magnet. The best way to account for these sources of error is to brainstorm with yours peers

Page | 8
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

about all the factors that could be possibly affect your result. This brainstorm should be done before
beginning the experiment so that arrangements can be made to account for the confounding factors before
taking data. Sometimes a correction can be applied to a result after talking data to account for an error
that was not detected.

ENVIRONMENTAL FACTORS (may be both type): Be aware of errors generated by your immediate
working environment. You may need to take account for or protect your experiment fro vibrations, drafts,
temperature, and electronic noise or any other affects from a nearby instrument.

INSTRUMENT RESOLUTION (mainly random): all instruments have finite precision that limits the
ability to resolve small measurements changes. For instance, meter stick cannot distinguish distances to a
precision much better than half about of its smallest scale division (0.5mm in this case). One of the best
ways to obtain more precise measurements is to be use a null difference method instead of measuring a
quantity directly. Null or balance methods involve using instrumentation to measure the differences
between two similar quantities, one of which is known very accurately and is adjustable. The adjustable
reference quantity is varied until the difference is reduced to zero. The two quantities are then balanced
and the magnitude of the unknown quality can be found by the comparison with a reference sample. With
this method, problems of source instability are eliminated, and the measuring instrument can be very
sensitive and does not even need a scale.

FAILURE TO CALIBRATE OR CHECK ZERO OF INSTRUMENTS (systematic): the calibration

of an instrument should be checked before taking data whenever possible. If a calibration standard is not
available, the accuracy of the instruments should be checked by comparing with other instruments that is
least as precise, or by consulting the technical data provided by the manufacturer. When making a reading
with a micrometer, electronic balances or an electrical meter, always check the zero reading first. Re-zero
the instrument if possible, or measure the displacement of the zero reading from the true zero and correct
any reading accordingly. It is a good idea to check the zero reading from the true zero and correct any
measurement accordingly. It is a good idea to check the zero reading throughout the experiment.

PHYSICAL VARIATIONS (Random): it is always wise to obtain multiple measurements over the
entire range being investigated. Doing so often reveals variations that might otherwise go undetected.
These variations may call for closer examinations, or they may be combined to find an average value.

PARALLAX (may be both): this error can occur whenever there is some distance between the
measuring scale and their indicator used to obtain a measurement. If the observer’s eyes are not

Page | 9
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

squarely aligned with a pointer and scale, the reading may be too high or too low. Some analogs meter
has mirrors to help with this problem.

INSTRUMENT DRIFT (systematic): most electronic instrument has readings that drift over time. The
amount of drift is generally of no concern, but occasionally this source of error can be significant and
should be considered.

LAG TIME AND HYSTERSIS (Systematic): some measuring devices require time to reach
equilibrium, taking the measurement before the instrument is stable will result in a measurement that
is generally too low. The most common example is taking temperature readings with a thermometer that
has not reached thermal equilibrium with its environment. A similar effect is hysteresis where the
instruments readings lag behind and appear to have a memory effect, as data is taken sequentially moving
up or down through a range of values.

PERSONAL ERRORS: personal errors come from carelessness, poor technique, or bias on the part of
the person performing instrument. The experimenter may measure incorrectly, or may use a poor
technique in taking readings or may introduce a bias into the readings by expecting the results to agree
with the expected outcome. Gross errors such as mistakes or blunders should be avoided and corrected if
found. The term human error should be avoided in the error analysis discussions because it is general to
be useful.

ESTIMATING EXPERIMENTAL UNCERTAINTY FOR A SINGLE MEASURMENT: The

uncertainty of the experiment of a single measurement is limited by the precision and the accuracy of the
measurement instrument, along with any other factors that affect the ability of the experimenter to make
the measurement. Unfortunately there is no general rule for determining the uncertainty in all
measurements. The person involved in the experiment is the best to evaluate and quantify the uncertainty
of a measurement based on all the possible factors that affect the results.

ESTIMATING UNCERTAINTY IN REPEATED MEASUREMENT: Suppose you time of

oscillation of a pendulum using digital instrument and find: T=0.44 s. this single measurement of the
period suggests a precision of ±0.005 s, but this instrument precision may not give a complete sense of
the uncertainty. If you repeat the measurement several times and examine the variation among their
measured values, you can get a better idea of the uncertainty in the period. For example, here are the
results of 5 measurements, in seconds: 0.46, 0.44, 0.45, 0.44, and 0.42.

The best of the period is the average, or mean, of these N independent measurements:

Page | 10
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY


⋯..!
Average mean=


Whenever possible, repeat a measurement several times and average the results. This average is the best
estimate of the true value. The more repetitions you make of a measurement, the better this estimate will
be.

One way to express the variation among the measurement is to use the average deviation. This statistics
tell us on average how much individual measurement varies from the mean.

|%& |'|%& |'⋯..|!%& |

Average deviation, "̅=


However, the standard deviation is the most common way to characterize the spread of a data set. The
standard deviation is always slightly greater that the average deviation, and is used of its associated with
the normal distribution that is frequently encountered in statistical analyses.

STANDARD DEVIATION: To calculate the standard deviation we have the formula –

+
)ƍ '⋯'!
S=( ,-.

STANDARD DEVIAITION OF THE MEAN (STANDARD ERROR): when we report the average
value of n measurements, the uncertainty we should associate with this average value is the standard of

the mean, often called the standard error (SE).it is given by the formula /̅ = . The standard error is
√

smaller than the standard deviation by a factor of 1/√1.

FRACTIONAL UNCERTAINTY: when a reported value is determined by taking the average of set
independent readings, the fractional uncertainty is given by the ratio of the uncertainty divided by the
average value. note that the uncertainty is dimensionless and it is important because it is used in
propagating uncertainty in calculations using the result of a measurement.

PROPAGATION OF UNCERTAINTY: for a single variable function f(x), the deviation in f can be
related to the deviation in x using calculus-

Page | 11
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

By using the definition of σ we have

23
The partial derivative 2 mean differentiating f with respect to x holding the other variables fixed. Taking
the square and the average, we get the law of propagation of uncertainty:

If the uncertainty terms are more than 3 times greater than the other terms, the root squares formula can
be skipped, and the combined uncertainty is simply the largest uncertainty. This shortcut can save a lot of
time without losing accuracy in the estimate of the overall uncertainty.

THE UPPER-LOWER BOUND METHOD OF UNCERTAINTY PROPAGATION: is an

alternative and sometimes simpler procedure, to the tedious propagation of uncertainty law. This
alternative method does not yield a standard uncertainty estimate, but it does give a reasonable estimate of
the uncertainty for practically any situation. The basic idea of using this method is to use the uncertainty
ranges of each variable to calculate maximum and minimum values of the function. The upper lower
bound method is especially useful when the functional is not clear or is incomplete.

SIGNIFICANT FIGURES: the no of significant figures in a value can be defined between and including
the first non-zero digit from the left, through the last digit. To avoid this ambiguity, such number should
be expressed in scientific notation. In fact, the number of significant figures suggests a rough estimate of
relative uncertainty.

UNCERTAINTY AND SIGNIFICANT FIGURES: for the same reason that it is dishonest to report a
result with more significant figures than are reliably known, the uncertainty value should also not be
reported with excessive precision experimental results should be rounded off in one or two significant
figures. When conducting experiment, it is important to keep in mind that precision is expensive. The
cost increases exponentially with the amount of precision required, so the potential benefit of this
precision must be weighed against extra cost.

COMBINING AND REPORTING UNCERTAINTIES: in 1993, the international standard

organization published the first official worldwide guide to the expression of uncertainty in measurement.
Before this time, uncertainty estimates are evaluated and reported according to different convention

Page | 12
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

depending upon the context of the measurement or the scientific discipline. Here are few of the lines in
the book by ISO.

When reporting a measurement, the measured value should be reported along with an estimate of total
combined standard uncertainty of the value. The total uncertainty is found by combining the uncertainty
components based open two types of uncertainty analysis: 1) TYPE A EVALUATION OF STANDARD
UNCERTAINTY- method of evaluation of uncertainty statistical analysis of a series of ob observations.
This method primarily includes random errors. 2) TYPE B EVALUATION OF STANDARD
UNCERTAINTY- method of evaluation of uncertainty by means other than the statistical analysis of
series of observations. This method includes systematic errors and any other uncertainty factors that the
experiment believes are important.

The individual uncertainty components should be combined using the law of propagation of uncertainties,
commonly called the root-sum-of-squares or RSS method.

AGREEMENT WITH DIFFERENT MEASUREMENTS: we have question that arises in the report
i.e. “does my result agree with a theoretical prediction or results from the experiments?”

Generally speaking, measured result agrees with a theoretical prediction if the prediction lies within the
range of experimental uncertainty. Similarly, if the two measured values have standard uncertainty ranges
that overlap, then measurement are said to be consistent. If the uncertainty ranges do not overlap, then the
measurement are said to be discrepant. However, you should recognize that these overlap criteria can give
two opposite answers depending on the evaluation and confidence level of the uncertainty. It would be
unethical to arbitrarily inflate the uncertainty range just to make measurement agree with expected value.
A better procedure would be to discuss the size of the differences between measured and expected values
within the context of the uncertainty, and try to discover the source of discrepancy if the different is truly
significant.

Error analysis is usually a big field where there are advanced studies also, but in this report I have written
some insights and significance of using error analysis.

REFERENCES:

1) John Taylor. An Introduction to Error Analysis, 2nd .ed. university science books.1997
2) ISO. Guide to the expression of uncertainty in measurement. International organization for
standardization and the international committee on weights and measures, Switzerland 1993.

Page | 13
 ABHISEKH UMRAO SINGH
7810595
M.Sc BIOTECHNOLOGY

3) Introduction to measurement and error analysis, dept of astronomy and physics, university of
North Carolina. 1998
4) David M. Harrison. Error analysis in experimental physical science, dept of physics, University
of Toronto.2010

ONLINE REFERENCES: www.scribd.com

Thank you for your patience in reading the report.

ABHISEKH UMRAO SINGH
7810595
University of Manchester

Page | 14