Statistical Treatment of Analytical Data I Semester Unit-I

ANALYTICAL CHEMISTRY
Introduction
The Scope of Analytical Chemistry
 Analytical chemistry has bounds which are amongst the widest of any technological
discipline.
 An analyst must be able to design, carry out, and interpret measurements within the
context of the fundamental technological problem with which he or she is presented.
 The selection and utilization of suitable chemical procedures requires a wide knowledge
of chemistry, whilst familiarity with and the ability to operate a varied range of
instruments is essential.
 Finally, analysts must have a sound knowledge of the statistical treatment of experimental
data to enable them to gauge the meaning and reliability of the results that they obtain.
 When an examination is restricted to the identification of one or more constituents of a
sample, it is known as qualitative analysis, while an examination to determine how much
of a particular species is present constitutes a quantitative analysis.
 Sometimes information concerning the spatial arrangement of atoms in a molecule or
crystalline compound is required or confirmation of the presence or position of certain
organic functional groups is sought. Such examinations are described as structural
analysis and they may be considered as more detailed forms of analysis.
 Any species that are the subjects of either qualitative or quantitative analysis are known
as analytes.

The Function of Analytical Chemistry

(a) Fundamental Research
 The first steps in unravelling the details of an unknown system frequently involve the
identification of its constituents by qualitative chemical analysis.
 Follow-up investigations usually require structural information and quantitative
measurements.
 This pattern appears in such diverse areas as the formulation of new drugs, the
examination of meteorites, and studies on the results of heavy ion bombardment by
nuclear physicists.

1
Statistical Treatment of Analytical Data I Semester Unit-I

(b)Product Development
 The design and development of a new product will often depend upon establishing a
link between its chemical composition and its physical properties or performance.
 Typical examples are the development of alloys and of polymer composites.
(c)Product Quality Control
 Most manufacturing industries require a uniform product quality.
 To ensure that this requirement is met, both raw materials and finished products are
subjected to extensive chemical analysis.
 On the one hand, the necessary constituents must be kept at the optimum levels, while
on the other impurities such as poisons in foodstuffs must be kept below the
maximum allowed by law.
(d)Monitoring and Control of Pollutants
 Residual heavy metals and organo-chlorine pesticides represent two well-known
pollution problems.
 Sensitive and accurate analysis is required to enable the distribution and level of a
pollutant in the environment to be assessed and routine chemical analysis is important
in the control of industrial effluents.
(e)Assay
 In commercial dealings with raw materials such as ores, the value of the ore is set by
its metal content.
 Large amounts of material are often involved, so that taken overall small differences
in concentration can be of considerable commercial significance.
(f)Medical and Clinical Studies
 The levels of various elements and compounds in body fluids are important indicators
of physiological disorders.
 A high sugar content in urine indicating a diabetic condition and lead in blood are
probably the most well-known examples.
Analytical Problems and Their Solution
The solutions of all analytical problems, both qualitative and quantitative, follow the same
basic pattern. This may be described under seven general headings.
(1)Choice of Method
 The selection of the method of analysis is a vital step in the solution of an analytical
problem.

2
Statistical Treatment of Analytical Data I Semester Unit-I

 A choice cannot be made until the overall problem is defined, and where possible a
decision should be taken by the client and the analyst in consultation.
 Inevitably, in the method selected, a compromise has to be reached between the
sensitivity, precision and accuracy desired of the results and the costs involved.
 For example, X-ray fluorescence spectrometry may provide rapid but rather imprecise
quantitative results in a trace element problem. Atomic absorption spectrophotometry,
on the other hand, will supply more precise data, but at the expense of more time-
consuming chemical manipulations.
(2)Sampling
 Correct sampling is the corner stone of reliable analysis.
 The analyst must decide in conjunction with technological colleagues how, where,
and when a sample should be taken so as to be truly representative of the parameter
that is to be measured.
(3)Preliminary Sample Treatment
 For quantitative analysis, the amount of sample taken is usually measured by mass or
volume.
 Where a homogeneous sample already exists, it may be subdivided without further
treatment.
 With many solids such as ores, however, crushing and mixing are prior requirements.
 The sample often needs additional preparation for analysis, such as drying, ignition
and dissolution.
(4)Separations
 A large proportion of analytical measurements is subject to interference from other
constituents of the sample.
 Newer methods increasingly employ instrumental techniques to distinguish between
analyte and interference signals.
(5)Final Measurement
 The fundamental necessity is a known proportionality between the magnitude of the
measurement and the amount of analyte present.
(6)Method Validation
 It is pointless carrying out the analysis unless the results obtained are known to be
meaningful.
 This can only be ensured by proper validation of the method before use and
subsequent monitoring of its performance.

3
Statistical Treatment of Analytical Data I Semester Unit-I

 The analysis of validated standards is the most satisfactory approach. Validated

standards have been extensively analysed by a variety of methods, and an accepted
value for the appropriate analyte obtained.
 A standard should be selected with a matrix similar to that of the sample. In order to
ensure continued accurate analysis, standards must be re-analysed at regular intervals.
(7)The Assessment of Results
 Results obtained from an analysis must be assessed by the appropriate statistical
methods and their meaning considered in the light of the original problem.

The Nature of Analytical Methods

It is common to find analytical methods classified as classical or instrumental, the
former comprising 'wet chemical' methods such as gravimetry and titrimetry. Such a
classification is historically derived andlargely artificial as there is no fundamental difference
between the methods in the two groups. All involve the correlation of a physical
measurement with the analyte concentration. Indeed, very few analytical methods are entirely
instrumental, and most involve chemical manipulations prior to the instrumental
measurement. A more satisfactory general classification is achieved in terms of the physical
parameter that is
measured
A general classification of important analytical techniques
Group : Property measured
Gravimetric weight of pure analyte or of a stoichiometric compound
containing it
volumetric : volume of standard reagent solution reacting with the analyte
Spectrometric : intensity of electromagnetic radiation emitted or absorbed by
the analyte.
Electrochemical : electrical properties of analyte solutions
Radiochemical : intensity of nuclear radiations emitted by the analyte
Mass spectrometric : abundance of molecular fragments derived from the analyte
Chromatographic : physico-chemical properties of individual analytes after
separation
Thermal : physico-chemical properties of the sample as it is heated and
cooled

4
Statistical Treatment of Analytical Data I Semester Unit-I

Trends in Analytical Methods and Procedures

 Better instrument design and a fuller understanding of the mechanics of analytical
processes enable steady improvements to be made in sensitivity, precision, and
accuracy.
 These same changes contribute to more economic analysis as they frequently lead to
the elimination of time-consuming separation steps.
 The ultimate development in this direction is a non-destructive method, which not
only saves time but leaves the sample unchanged for further examination or
processing.
The Language of Analytical Chemistry
 Analysis
An analysis provides chemical or physical information about a sample. The components
of interest in the sample are called analytes, and the remainder of the sample is the matrix.
 Determination
In an analysis we determine the identity, concentration, or properties of the analytes. To
make this determination we measure one or more of the analyte’s chemical or physical
properties.
 Measurement
An experimental determination of an analyte’s chemical or physical properties.B
 Techniques
A technique is any chemical or physical principle that can be used to study an
analyte. Many techniques have been used to determine lead levels. For example, in
graphite furnace atomic absorption spectroscopy lead is atomized, and the ability of the free
atoms to absorb light is measured; thus, both a chemical principle (atomization) and a
physical principle (absorption of light) are used in this technique.
 Methods
A method is the application of a technique for the determination of a specific analyte in a
specific matrix. As shown in (Figure 2), the graphite furnace atomic absorption spectroscopic
method for determining lead levels in water is different from that for the determination of
lead in soil or blood. Choosing a method for determining lead in water depends on how the
information is to be used and the established design criteria (Figure 3). For some analytical
problems the best method might use graphite furnace atomic absorption spectroscopy,

5
Statistical Treatment of Analytical Data I Semester Unit-I

whereas other problems might be more easily solved by using another technique, such as
anodic stripping voltammetry or potentiometry with a lead ion-selective electrode.

Techniques Graphite furnace atomic absorption spectroscopy

Methods Pb in Soil Pb in Water Pb in Blood

Procedures APHA (American Public Health Association)

ASTM (American Society for Testing Materials)

Protocols EPA (Environmental Protection Agency)

(Figure 3)

1. Identify the problem: Determine type of information needed (qualitative, quantitative, or

characterization) Identify context of the problem
2. Design the experimental procedure: Establish design criteria (accuracy, precision, scale of
operation, sensitivity, selectivity, cost, speed)
 Identify interferents
 Select method
 Establish validation criteria
 Establish sampling strategy
 Procedures
A procedure is a set of written directions detailing how to apply a method to a particular
sample, including information on proper sampling, handling of interferents, and validating
results. A method does not necessarily lead to a single procedure, as different analysts or
agencies will adapt the method to their specific needs. As shown in Figure 3.2, the American
Public Health Agency and the American Society for Testing Materials publish separate
procedures for the determination of lead levels in water.
 Protocols
A protocol is a set of stringent written guidelines detailing the procedure that must be
followed if the agency specifying the protocol is to accept the results of the analysis.

6
Statistical Treatment of Analytical Data I Semester Unit-I

Protocols are commonly encountered when analytical chemistry is used to support or define
public policy. For purposes of determining lead levels in water under the Safe Drinking
Water Act, labs follow a protocol specified by the Environmental Protection Agency.

Classifying Analytical Techniques

 Analyzing a sample generates a chemical or physical signal whose magnitude is
proportional to the amount of analyte in the sample.
 The signal may be anything we can measure; common examples are mass, volume, and
absorbance.
 For our purposes it is convenient to divide analytical techniques into two general classes
based on whether this signal is proportional to an absolute amount of analyte or a relative
amount of analyte.
Consider two graduated cylinders, each containing 0.01 M Cu(NO3)2 (Figure 4).
Cylinder 1 contains 10 mL, or 0.0001 mol, of Cu2+; cylinder 2 contains 20 mL, or 0.0002
mol, of Cu2+.
 If a technique responds to the absolute amount of analyte in the sample, then the signal
due to the analyte, SA, can be expressed as
SA = knA ------------1
where nA is the moles or grams of analyte in the sample, and k is a proportionality constant.
Since cylinder 2 contains twice as many moles of Cu2+ as cylinder 1, analyzing the contents
of cylinder 2 gives a signal that is twice that of cylinder 1.

Graduated cylinders containing 0.01 M Cu(NO3)2. (a) Cylinder 1

contains 10 mL, or 0.0001 mol, of Cu2+. (b) Cylinder 2 contains 20 mL, or 0.0002 mol, of
Cu2+.
(Figure 4)
 A second class of analytical techniques are those that respond to the relative amount of
analyte; thus
SA = kCA ------------2

7
Statistical Treatment of Analytical Data I Semester Unit-I

where CA is the concentration of analyte in the sample. Since the solutions in both cylinders
have the same concentration of Cu2+, their analysis yields identical signals.

Selecting an Analytical Method

A method is the application of a technique to a specific analyte in a specific matrix.
Methods for determining the concentration of lead in drinking water can be developed using
any of the techniques mentioned in the previous section. Insoluble lead salts such as PbSO4
and PbCrO4 can form the basis for a gravimetric method.
1. Lead forms several soluble complexes that can be used in a complexation
titrimetric method or, if the complexes are highly absorbing, in a
spectrophotometric method.
2. Lead in the gaseous free-atom state can be measured by an atomic absorption
spectroscopic method.
3. Finally, the availability of multiple oxidation states (Pb, Pb2+, Pb4+) makes
coulometric, potentiometric, and voltammetric methods feasible.
The requirements of the analysis determine the best method. In choosing a method,
consideration is given to some or all the following design criteria: accuracy, precision,
sensitivity, selectivity, robustness, ruggedness, scale of operation, analysis time, availability
of equipment, and cost. Each of these criteria is considered in more detail in the following
sections.
Accuracy
 Accuracy is a measure of how closely the result of an experiment agrees with the
expected result.
 The difference between the obtained result and the expected result is usually divided by
the expected result and reported as a percent relative error

 Analytical methods may be divided into three groups based on the magnitude of their
relative errors.
 When an experimental result is within 1% of the correct result, the analytical method is
highly accurate. Methods resulting in relative errors between 1% and 5% are moderately
accurate, but methods of low accuracy produce relative errors greater than 5%.

8
Statistical Treatment of Analytical Data I Semester Unit-I

The magnitude of a method’s relative error depends on how accurately the signal is
measured, how accurately the value of k in equations 1 or 2 is known, and the ease of
handling the sample without loss or contamination.
Precision
 When a sample is analyzed several times, the individual results are rarely the same.
Instead, the results are randomly scattered.
 Precision is a measure of this variability. The closer the agreement between individual
analyses, the more precise the results.
 For example, in determining the concentration of K+ in serum, the results shown in
Figure 5(a) are more precise than those in Figure 5(b).
 It is important to realize that precision does not imply accuracy. That the data in Figure
5(a) are more precise does not mean that the first set of results is more accurate. In fact,
both sets of results may be very inaccurate.
 As with accuracy, precision depends on those factors affecting the relationship between
the signal and the analyte (equations 1 and 2).

Two determinations of the concentration of K+ in

serum, showing the effect of precision. The data in (a) are less scattered and, therefore, more
precise than the data in (b).
(Figure 5)
Sensitivity
 The ability to demonstrate that two samples have different amounts of analyte is an
essential part of many analyses.
 The detection limit is the smallest amount of analyte that can be determined with
confidence. The detection limit, therefore, is a statistical parameter.
 Sensitivity is the change in signal per unit change in the amount of analyte and is
equivalent to the proportionality constant, k, in equations 1 and 2.
 If ∆SA is the smallest increment in signal that can be measured, then the smallest
difference in the amount of analyte that can be detected is

9
Statistical Treatment of Analytical Data I Semester Unit-I

(total analysis method)

(concentration method)

Selectivity
An analytical method is selective if its signal is a function of only the amount of
analyte present in the sample. In the presence of an interferent, equations 1 and 2 can be
expanded to include a term corresponding to the interferent’s contribution to the signal, SI,

Ssamp = SA + SI = kAnA + kInI (total analysis method) -------3

Ssamp = SA + SI = kACA + kICI (concentration method) ------4

where Ssamp is the total signal due to constituents in the sample; kA and kI are the sensitivities
for the analyte and the interferent, respectively; and nI and CI are the moles (or grams) and
concentration of the interferent in the sample.
The selectivity of the method for the interferent relative to the analyte is defined by a
selectivity coefficient, KA,I
------5

which may be positive or negative depending on whether the interferent’s effect on the signal
is opposite that of the analyte. A selectivity coefficient greater than +1 or less than –1
indicates that the method is more selective for the interferent than for the analyte. Solving
equation 5 for kI
kI = KA,I × kA -------6
substituting into equations 3.3 and 3.4, and simplifying gives

Ssamp = kA(nA + KA,I ´ nI) (total analysis method) ---------7

Ssamp = kA(CA + KA,I ´ CI) (concentration method) -------- 8

The selectivity coefficient is easy to calculate if kA and kI can be independently

determined. It is also possible to calculate KA,I by measuring Ssamp in the presence and
absence of known amounts of analyte and interferent.
Knowing the selectivity coefficient provides a useful way to evaluate an interferent’s
potential effect on an analysis. An interferent will not pose a problem as long as the term

10
Statistical Treatment of Analytical Data I Semester Unit-I

KA,I × nI in equation (7) is significantly smaller than nA, or KA,I ×CI in equation(8) is
significantly smaller than CA.
Robustness and Ruggedness
 Methods are subject to a variety of chemical and physical interferences that contribute
uncertainty to the analysis.
 When a method is relatively free from chemical interferences, it can be applied to the
determination of analytes in a wide variety of sample matrices. Such methods are
considered robust.
 Random variations in experimental conditions also introduce uncertainty.
 If a method’s sensitivity is highly dependent on experimental conditions, such as
temperature, acidity, or reaction time, then slight changes in those conditions may lead to
significantly different results.
 A rugged method is relatively insensitive to changes in experimental conditions.
Scale of Operation
 Another way to narrow the choice of methods is to consider the scale on which the
analysis must be conducted.
 The scale of operations in Figure 6 shows the analyte’s concentration in weight percent
on the y-axis and the sample’s size on the x-axis.
 For convenience, we divide analytes into
a) major (>1% w/w)
b) minor (0.01% w/w – 1% w/w)
c) trace (10–7 % w/w – 0.01% w/w) and
d) ultra trace (< 10–7% w/w) components
 we divide samples into
a) macro (>0.1 g)
b) meso (10 mg – 100 mg)
c) micro (0.1 mg – 10 mg) and
d) ultra micro (<0.1 mg) sample sizes.
 Note that both the x-axis and the y-axis use a logarithmic scale.
 The analyte’s concentration and the amount of sample used provide a characteristic
description for an analysis.
 For example, samples in a macro–major analysis weigh more than 0.1 g and contain more
than 1% analyte.

11
Statistical Treatment of Analytical Data I Semester Unit-I

Diagonal lines connecting the two axes show combinations of sample size and
concentration of analyte containing the same absolute amount of analyte. As shown in Figure
6, for example, a 1-g sample containing 1% analyte has the same amount of analyte (0.010 g)
as a 100-mg sample containing 10% analyte or a 10-mg sample containing 100% analyte.

Scale of operation for analytical methods.

(Figure 6)
Equipment
Finally, analytical methods can be compared in terms of their need for equipment, the time
required to complete an analysis, and the cost per sample.
 Methods relying on instrumentation are equipment-intensive and may require significant
operator training.
 For example, the graphite furnace atomic absorption spectroscopic method for
determining lead levels in water requires a significant capital investment in the instrument
and an experienced operator to obtain reliable results.
Time
 The time needed to complete an analysis for a single sample is often fairly similar from
method to method.
 This is somewhat misleading, however, because much of this time is spent preparing the
solutions and equipment needed for the analysis.

12
Statistical Treatment of Analytical Data I Semester Unit-I

 Once the solutions and equipment are in place, the number of samples that can be
analyzed per hour differs substantially from method to method.
 This is a significant factor in selecting a method for laboratories that handle a high
volume of samples.
Cost
 The cost of an analysis is determined by many factors, including the cost of necessary
equipment and reagents, the cost of hiring analysts, and the number of samples that can
be processed per hour.
 In general, methods relying on instruments cost more per sample than other methods.
Making the Final Choice
 Attempts to minimize cost and analysis time may decrease accuracy.
 Selecting a specific method requires a careful balance among these design criteria.
Usually, the most important design criterion is accuracy, and the best method is that
capable of producing the most accurate results.
 When the need for results is urgent, as is often the case in clinical labs, analysis time may
become the critical factor.
 The best method is often dictated by the sample’s properties. Analyzing a sample with a
complex matrix may require a method with excellent selectivity to avoid interferences.
 Samples in which the analyte is present at a trace or ultratrace concentration usually must
be analyzed by a concentration method.
 If the quantity of sample is limited, then the method must not require large amounts of
sample.
Determining the concentration of lead in drinking water requires a method that can
detect lead at the parts per billion concentrations. Selectivity is also important because other
metal ions are present at significantly higher concentrations. Graphite furnace atomic
absorption spectroscopy is a commonly used method for determining lead levels in drinking
water because it meets these specifications. The same method is also used in determining lead
levels in blood, where its ability to detect low concentrations of lead using a few microliters
of sample are important considerations.

13
Statistical Treatment of Analytical Data I Semester Unit-I

ERRORS & TREATMENT OF ANALYTICAL DATA

Two main classes of errors can affect the accuracy or precision of a measured quantity,
1. Determinate error/ systematic/ constant error.
2. Indeterminate error/ random error.
1. Determinate error/ systematic/ constant error:
 Are those that, as the name implies, are determinate and that presumably can be either
avoided or corrected.
 Determinate errors are often reproducible, and in many cases they can be predicted by
person who thoroughly understands all the aspects of the measurement.
Ex: sources of determinate errors are incorrectly calibrated instruments, such as a burette,
balances or PH meter and impurities in the reagent, a side reaction in a titration and heating a
sample at too high a temperature
Determinate errors have been classified as:
1. Methodical error.
2. Operative error/personal error.
3. Instrumental error.
Methodical error:
These are the most serious errors of an analysis. Most of the above errors can be
minimized or corrected for, but errors that are inherent in a method cannot be changed unless
the conditions of the determination are altered. some sources of methodical errors includes,
co precipitation of impurities, slightly solubility of a precipitate, side reactions, incomplete
reactions, impurities in reagents.
Operative error/personal error:
These includes personal error and can be reduced by experience and care of the
analyst in the physical manipulations involved, operations in which these errors may occur
include, transfer of solutions, effervescences & bumping of samples, incomplete drying of
samples, improper washing of ppt., errors in reading burette, allowing hygroscopic materials
to absorb moisture, errors in calculations.
Instrumental error:
These includes faulty equipment, un calibrated weights, un calibrated glass wares,
failure of measuring devices to perform in accordance with required standards.
Determinate errors can also be classified as

14
Statistical Treatment of Analytical Data I Semester Unit-I

 Constant errors: a constant error is independent of a magnitude of the measured

quantity and becomes less significant as the magnitude.
Ex: if a constant end point error of 0.10ml is made in a series of titrations, these
represents a relative error of 1% for a sample requiring 10ml of titrant, but only 0.2%
if 50 ml of titrant is used
 Proportional errors; the absolute value of this type of error varies with sample size
in such a way that the relative error remains constant. A substance that interferes in an
analytical method may lead to such an error if present in the sample.

Ex: in the iodometric determination of an oxidant like chlorate, another oxidizing agent
such as bromate would cause high results if its presence were unsuspected and not
corrected for. Taking larger samples would increase the absolute error, but the relative
error would remain constant provided the sample was homogeneous.
2. Indeterminate error/ random error:
Indeterminate errors, as the name implies, cannot be attributed to any known cause,
but they inevitably attend measurements made by human beings. They are random in nature
and lead to both high and low results with equal probability. They cannot be eliminated or
corrected and are the ultimate limitation on the measurement. They can be treated by
statistics, & repeated measurement of the same variable can have the effect of reducing their
importance.
Minimization of errors:
Detection of Systematic Instrument and Personal Errors
 Some systematic instrument errors can be found and corrected by calibration.
 Periodic calibration of equipment is always desirable because the response of most
instruments changes with time as a result of wear, corrosion, or mistreatment.
 Many systematic instrument errors involve interferences in which a species present in
the sample affects the response of the analyte.
 Most personal errors can be minimized by care and self-discipline.
 It is a good habit to check instrument readings, notebook entries, and calculations
systematically.
 Errors due to limitations of the experimenter can usually be avoided by carefully
choosing the analytical method.

15
Statistical Treatment of Analytical Data I Semester Unit-I

Detection of Systematic Method Errors:

 Analysis of Standard Samples
The best way of estimating the bias of an analytical method is by the analysis of standard
reference materials, materials that contain one or more analytes at known concentration
levels. Standard reference materials are obtained in several ways
 Standard reference material can be purchased from a number of governmental
and industrial sources. For example, the National Institute of Standards and
Technology (NIST) (formerly the National Bureau of Standards) offers more
than 1300 standard reference materials, including rocks and minerals, gas
mixtures, glasses, hydrocarbon mixtures, polymers, urban dusts, rain waters,
and river sedirnents."
 Independent Analysis
If standard samples are not available, a second independent and reliable analytical method
can be used in parallel with the method being evaluated.
 Blank Determinations
A blank contains the reagents and solvents used in a determination, but no analyte. Often,
many of the sample constituents are added to simulate the analyte environment, often called
the sample matrix.
 Variation in Sample Size
The size of a measurement increases, the effect of a constant error decreases. Thus,
constant errors can often be detected by varying the sample size.
Characterizing Measurements and Results
Let’s begin by choosing a simple quantitative problem requiring a single
measurement.
Measures of Central Tendency
Two common ways to report this estimate of central tendency are the mean and the
median.
Mean
 The mean, x̅, is the numerical average obtained by dividing the sum of the individual
measurements by the number of measurements

where Xi is the ith measurement, and n is the number of independent measurements.

16
Statistical Treatment of Analytical Data I Semester Unit-I

 The mean is the most common estimator of central tendency. It is not

considered a robust estimator, however, because extreme measurements, those much
larger or smaller than the remainder of the data, strongly influence the mean’s value.
Median:
The median, Xmed, is the middle value when data are ordered from the smallest to the
largest value. When the data include an odd number of measurements, the median is the
middle value. For an even number of measurements, the median is the average of the n/2 and
the (n/2) + 1 measurements, where n is the number of measurements.
Measures of Spread
Although spread is often defined relative to a specific measure of central tendency, its
magnitude is independent of the central value. Changing all measurements in the same
direction, by adding or subtracting a constant value, changes the mean or median, but will not
change the magnitude of the spread.
Three common measures of spread are range, standard deviation, and variance.
Range:
The range, w, is the difference between the largest and smallest values in the data set.
Range = w = Xlargest – Xsmallest
The range provides information about the total variability in the data set, but does not provide
any information about the distribution of individual measurements.
Standard Deviation:
The absolute standard deviation, s, describes the spread of individual measurements about
the mean and is given as

---------------1

where Xi is one of n individual measurements, and is the mean.

Frequently, the relative standard deviation, sr, is reported.

sr =

The percent relative standard deviation is obtained by multiplying sr by 100%

Variance:
Another common measure of spread is the square of the standard deviation, or the
variance. The standard deviation, rather than the variance, is usually reported because the
units for standard deviation are the same as that for the mean value.

17