Accuracy and precision,

Errors and significant

figures

Errors: Sources of errors, types of errors, methods of minimizing errors,accuracy,

precision and significant figures
 Good precision does not
assure good accuracy

The accuracy of an analytical method expresses the closeness of agreement between

the true value and the experimental value found

Precision may be defined as the concordance of a series of

measurements of the same quantity
 1.3 Errors in
Analysis
Accuracy , reproducibility and reliability are basis of
Analytical chemistry
The job of analytical chemist is that to device
method which gives result that are accurate,
precise (reproducible) and reliable.

Every measurement made, no matter how

systematically and
carefully, is subjected to some degree of uncertainty
or error
Magnitude of error = Measured value – True value is not determinable

All we can do it is to decrease the margin of uncertainty so that

the results become sufficiently accurate to an acceptable
level
Errors
Errors can affect the accuracy or precision of measured quantity

The difference between the experimental value and a true

value is termed as error

Error

Absolute error Relative

error

Absolute Error: The difference between experimental mean and a

true value is termed as Absolute error. It may be positive or
negative

Relative Error = (Measured mean value – True value) / True value

Types of Error
1) Determinate (Systematic) errors :
As the name implies, magnitude of these errors can be determined and
can be avoided. The determinate error can be classified
as

I) Operational and personal errors

II) Instrumental and reagent errors

III) Errors of method

I) Operational and personal errors
These errors are mostly physical in nature and
occur when sound analytical technique is not
followed Operational errors

1) Incomplete drying of analytical samples before

Weighing, allowing hygroscopic materials to
absorb moisture before or during weighing
2) Mechanical loss of
materials during sampling ( from
effervescence or bumping, underwashing
or overwashing of precipitates; ignition of
precipitates
at incorrect temperatures;
3) insufficient cooling of crucibles before
weighing;; and use of reagents
containing harmful impurities
Personal errors

In ability of an individual to make certain observation accurately

Personal errors may arise from the constitutional inability of an

individual to make certain observations accurately. Thus some persons
are unable to judge colour changes sharply in visual titrations, which
may result in a slight overstepping of the end point.
II) Instrumental and reagent
errors

These errors arises due to

1) Faulty constructions of balances
2) The use of uncalibrated or improperly
calibrated weights, graduated glassware and
other instruments.
3) The attack of reagents upon glassware,
porcelain, resulting in the introduction of foreign
materials and
4) Use of reagents containing impurities
III) Errors of
method
These are most serious errors and often difficult to detect

1) Wrongly standardized pH meter

2) Faulty detector response in chromatographic and
spectroscopic method.
3) In gravimetric analysis errors may arise owing to
appreciable solubility of precipitates, co-precipitation, and
post-precipitation, decomposition, or volatilisation of
weighing forms on ignition, and precipitation of
substances other than the intended ones.
4)Errors arises in titrimetry If there is
difference between
the observed end
point and the stoichiometric
equivalence point of reaction
5) Matrix effects- When there is difference between
composition of the analyte sample solution and
composition of standard solution used to establish
 How to reduce systematic errors ?????

1) Calibration of apparatus and application of corrections

All instruments should be calibrated, and the appropriate corrections applied
to original measurements.
Calibration of apparatus like burettes , pipettes, weighing balance,
spectrophotometers is an important step prior to analysis

2) Running blank determination

Here separate determination is carried out by omitting sample, under
exactly same experimental conditions as employed in the actual analysis of
the sample.
The objective is to find out effect of the impurities introduced
through reagents and vessels, or
to determine excess of standard solution necessary to establish end point in
case of back titrations
3) Running control determination
Here determination is carried out on a standard substance under conditions
as close as possible to the experimental conditions
 upon a quantity of a standard substance which contains the same
weight of the constituent as is contained in the unknown sample. The
weight of the constituent in the unknown can then be calculated
from the relation:

standard samples which have been analysed by a number of skilled analysts are
commercially available(Certified Reference Materials (CRM), Certified Reference
Materials (CRM). These include certain primary standards (sodium oxalate,
potassium hydrogenphthalate, arsenic(III) oxide, and benzoic acid) and ores,
ceramic materials, irons, steels, steel-making alloys, and non-ferrous alloys.
4) Use of independent method of
analysis
Here accuracy of result is established by carrying out the analysis in an
entirely different manner.
Ex- Water hardness, the calcium and magnesium conc. Is determined by
Atomic absorption spectroscopy and may be compared with the results
obtained by EDTA titration.
Iron may first be determined gravimetrically by precipitation as iron(III)
hydroxide after removing the interfering elements, followed by ignition of the
precipitate to iron(III) oxide. It may then be determined titrimetrically by
reduction to the iron(II) state, and titration with a standard solution of an
oxidising agent, such as potassium dichromate or cerium(IV) sulphate.

The determination of the strength of a hydrochloric acid solution both by

titration with a standard solution of a strong base and by precipitation and
weighing as silver chloride.
5) Running parallel determinations
• The value obtained for constituents should not
vary among themselves by more than three parts per
thousand.
• If larger variations are shown determination must be repeated
in duplicate and at most triplicate.
• Good agreement between duplicate and triplicate do not
justify that
result
variation of
is determinate
correct, it merely
errorsshows
are same or
that
nearly
accidentalsame
error in
or
parallel determinations.
6) Standard addition
A known amount of constituent being determined is added to the
sample, which is then analyzed for the total amount of
constituent present.

The difference between the analytical results for samples with and
without the added constituents gives the recovery of the amount of
added constituent.
7) Internal standards
This procedure is of particular value in chromatographic determinations.
It involves adding a fixed amount of a reference material (the internal
standard) to series of known concentrations of material to be measured.
The ratios of the physical value (absorption or peak size) of the internal
standard and the series of known concentrations are plotted against the
concentration values. This should give a straight line. Any unknown
concentration can then be determined by adding the same quantity of
internal standard and finding where the ratio obtained falls on the
concentration scale

8) Amplification method
If very small amount of material is to be determined, this may be beyond
the limit of instruments available.
In this case small amount of material can be reacted in such a way that every
molecule produces two or more molecules of some other measurable
material. The resultant amplification may then bring the quantity to be
determined within the scope of the apparatus or method available.
8) Isotopic dilution.

A known amount of the element being determined, containing a radioactive

isotope, is mixed with the sample and the element is isolated in a pure form
(usually as a compound), which is weighed or otherwise determined. The
radioactivity of the isolated material is measured and compared with that of
the added element: the weight of the element in the sample can then be
calculated.
2) Indeterminate (random) / accidental errors :
As the name implies, these errors can‟t be
determined and can‟t be avoided.
These errors are revealed by small differences in
successive measurements made by same analyst
under with the greatest care under as nearly identical
conditions as possible

Analyst do not have any control on

these errors

These accidental errors will follow a

random
distribution;
applied to and are at
arrive not constant,conclusion
therefore,
mathematical
some
regarding most probable results of laws
a of
probability
of measurements. can be
series
• These random errors cause variation in results, some of which
may be too high and some too low.

• The average of the replicate determinations is accurate, but

each individual determination may vary slightly from the true
value.

• Indeterminate errors arise from sources that cannot be

corrected,
avoided, or even identified, in some cases.

• All analytical procedures are subject to indeterminate error.

However, because indeterminate error is random, the errors
will follow a random distribution.
 Examples of random errors

• Due to changes in the environmental conditions like

temperature and humidity which can vary the results of analysis

• Appearance of noise in chromatographic analysis

• Drift from an electronic circuit

• Vibrations in balance caused while weighing

If a sufficiently large number of observations is taken it can be shown
that these errors lie on a curve of the form shown in

An inspection of this error curve shows: (a) small errors occur more
frequently than large ones; and (b) positive and negative errors of the
same numerical magnitude are equally likely to occur.
 Random errors often have a Gaussian normal distribution (see Fig. 2). In such cases
statistical methods may be used to analyze the data. The mean m of a number of
measurements of the same quantity is the best estimate of that quantity, and the standard
deviation s of the measurements shows the accuracy of the estimate. The standard error
of the estimate m is s/sqrt(n), where n is the number of measurements.

The Gaussian normal distribution. m = mean of

measurements. s = standard deviation of
measurements.
68% of the measurements lie in the interval m -
s < x < m + s;
95% lie within m - 2s < x < m + 2s;
and 99.7% lie within m - 3s < x
 The amount of information that the
measuring instrument can
provide.
The precision of a measuring tool
 depends on how readable the tool
Ruler example
is.
◦ 14.5 cm is all you can read off of a cm ruler
◦ To say that the line was 14.5678 would give the
impression that a very precise tool had been used
not a ruler.
 When measuring with a mechanical instrument
(ruler, triple beam balance etc), record all the digits
that are marked on the instrument’s scale and
estimate (and only one) more digit.

 When measuring with an electronic instrument,

record all the digits on the readout. Consider the
last digit to be approximate.

 Round calculated answers only once, at the end of

the calculation, so that the number of significant
digits reflects the precision of the original
measurements.
 Every experimental
measurement has a
degree of uncertainty.
The volume, V, at
right is certain in the
10’s place,
 10mL<V<20mL
The 1’s digit is also
certain,
17mL<V<18mL
 A best guess is needed
for the tenths place.

34
 1 2 3
4 cm
 We can see the markings between 1.6-1.7cm
 We can’t see the markings between the .6-.7
 We must guess between .6 & .7
 We record 1.67 cm as our measurement
 The last digit an 7 was our guess...stop there

35
What is the length of the wooden
stick?
1) 4.5 cm
2) 4.54 cm
3) 4.547 cm
Significant
figures
Definition- The number of digits necessary to express result ofmeasurement
consistent with measured precision
Since there is uncertainty(imprecision) in any measurement of at least 1
in
the last significant figure.
Observed quantities should be recorded with one uncertain figure retained.

Sample weighed on rough balance Sample weighed on analytical

balance

15.6 g 15.5988 g
Three significant figures Six significant figures

15.6 0.1 g 15.5988 0.0001 g

Significant figures are independent of decimal point placement.

The weights 515g,5.15g, 0.515g all contain three significant
figures
Significant figures
Weighing and measuring can be carried out with only a certain
maximum degree of accuracy

Result always be approximate value due to many sources

of error such as Temp., limitations of the instruments
employed personal factors and so on…………………………

Pharmacist must achieve the greatest accuracy possible

with the equipment

You can’t weigh 1mg of solid on 10mg sensitive balance

You can’t measure 76.32 ml of liquid in 100 ml measuring cylinder,
which can be read only to 1ml

When quantities are written, the number should contain only

those digits that are significant within the precision of
the instrument

Significant figures are digits that have practical meaning

  Do you see why Measured Numbers have
error…you have to make that Guess!
 All but one of the significant figures are known
with certainty. The last significant figure is
only the best possible estimate.
 To indicate the precision of a measurement, the
value recorded should use all the digits known
with certainty.
Below are two measurements of the mass of
the same object. The same quantity is
being described at two different levels of
precision or certainty.
 38.57 has four significant figures but two
decimal places
 283 has three significant figures but no decimal
places
 0.0012 has two significant figures but four
decimal places
 A balance always weighs to a fixed number of
decimal places. Always record all of them
◦ As the weight increases, the number of significant
figures in the measurement will increase, but the
number of decimal places is constant
◦ 0.0123 g has 3 S.F.; 10.0123 g has 6 S.F.
Zero
0
The digit zero is significant figure except when it is the first figure in
the
number.

Thus
1.2680ingthe quantities
1.0062 g The zero is significant
&
0.0025 kg The zeros are not significant figures
They serves only to locate the decimal point
2.5 g Can be omitted by proper choice of units

The first two number 1.2680 & 1.0062 contain five significant figures,
but 0.0025 contains only two significant figures
 Nonzero digits(1,2,3,4,5,6,7,8,9) are always
significant
38.57 (four) 283 (three)
 Zeroes are sometimes significant and sometimes
not
◦ Zeroes at the beginning: never significant
0.052 (two)
◦ Zeroes between: always 6.08 (three)
◦ Zeroes at the end after decimal: always
39.0 (three)
◦ Zeroes at the end with no decimal point may or
may not: 23 400 km (three, four, five)
 A significant figure is any non-zero digit or any embedded
or trailing zero. Leading zeros are not significant.
 38.57 has four significant figures but two
decimal places
 283 has three significant figures but no decimal
places
 0.0012 has two significant figures but four
decimal places
 A balance always weighs to a fixed number of
decimal places. Always record all of them
◦ As the weight increases, the number of significant
figures in the measurement will increase, but the
number of decimal places is constant
◦ 0.0123 g has 3 S.F.; 10.0123 g has 6 S.F.
How many
significant digits
are in each of the
following Answers:
examples? a) 3
a) 438 b) 4
b) 26.42 c) 2
c) 1.7 d) 3
d) .653
How many
significant digits
are in each of the
following Answers:
examples?
a) 3

a) 506 b) 5

b) 10,052 c) 6

c) 900.431
How many
significant digits
are in each of the Answers:
following
examples? a) 3
b) 1
a) 4830 c) 1
b) 60
c) 4,000
How many
significant digits
are in each of the Answers:
following
examples?
a) 4
b) 2
a) 4830.
c) 4
b) 60.
c) 4,000.
How many
significant digits
are in each of the
following Answers:
examples?
a) 1

a) 0.06 b) 2

b) 0.0047 c) 1

c) 0.005
How many
significant digits
are in each of the
following Answers:
examples?
a) 3

a) .870 b) 2
c) 4
b) 8.0
d) 5
c) 16.40
e) 3
d) 35.000
e) 1.60
 When we are using measurements in
different calculations in Chemistry and
Physics and even Biology, we need to
account for the level of precision. To do so,
scientists use significant figures.
 The last sig fig in a measurement is always
the doubtful digit. But that is not always
clear. For example

 12340cm or 12340.cm or 12340.0cm

 Scientists use scientific notation as a method
to show the proper amount of sig. figs.
 300 km can be written as 3 x 10 2 km.
(10 2 is 100 – so 3 x 100 =300)
 300 km (with 2 sig. figs) can be written as
3.0 x 10 2 km. Notice we clearly have 2 sig.
figs.
 When multiplying by 10 + x we can move the
decimal x times to make the number bigger.

Ex. 4 x 10 5 N = 400000 N

Ex. 3.00 x 10 4 kg = 30000 kg (3 sig. figs

– shown in sci. notation)
 When multiplying by 10 - x , we can move the
decimal x times to make the number smaller.

 6.01 x 10 - 2 m = 0.0601 m (3 sig. figs)

 3.10 x 10 - 5 N = 0.0000310 N (3 sig. figs.)

 1,000,000 = 1 x 10 6
 0.000 001 = 1 x 10 - 6
 234,000 = 2.34 x 10 5
 0.00234 = 2.34 x 10 - 3
 2.3400 x 10 4 (five S.F.)
 2.340 x 10 4 (four S.F.)
 2.34 x 10 4 (three S.F.)

 23 400. also indicates five S.F.

 23 400.0 has six S.F.
Measurement # of Sig Figs
1) 1400.0
2) 300
3) 0.0050
4) 6001.30
5) 11232.0
6) 5.00
RULE 1. In carrying out a multiplication or
division, the answer cannot have more
significant figures than either of the
original numbers.
32.27  1.54 = 49.6958 49.7

3.68  .07925 = 46.4353312 46.4

1.750  .0342000 = 0.05985 .05985

3.265010 6  4.858 = 1.586137  10 7 1 . 5 8 6 10 7

6.02210 2 3  1.66110 - 2 4 = 1.000000 1 . 0 0 0

 Rule for addition/subtraction: The last digit
retained in the sum or difference is
determined by the position of the first
doubtful digit
37.24 + 10.3 = 47.5
1002 + 0.23675 = 1002
225.618 + 0.23 = 225.85
 Position is key
RULE 2. In carrying out an addition or
subtraction, the answer cannot have more
digits after the decimal point than either
of the original numbers.
.56 + .153 = .713 .71 Look for
the last
82000 + 5.32 = 82005.32 82000 important
10.0 - 9.8742 = 0.12580 0.1 digit

10 – 9.8742 = 0.12580 0.
Make the following into a 3 Sig Fig number

Your Final number

1.5587 1.56 must be of the same
.0037421 .00374 value as the number
you started with,
1367 1370 129,000 and not 129
128,522 129,000
1.6683 106 1.67 106
 Often when doing arithmetic on a
pocket calculator, the answer is
displayed with more significant figures
than are really justified.
 How do you decide how many digits to
keep?
 Simple rules exist to tell you how.
 5 or above goes up
◦ 37.45 → 37.5 (3 S.F.)
◦ 123.7089 → 123.71(5 S.F.); 124 (3 S.F.)
< 5 goes down
◦ 37.45 → 37 (2 S.F.)
◦ 123.7089 → 123.7 (4 S.F.)
 For example you want a 4 Sig Fig number

4965.03 4965 0 is dropped, it is < 5

780,582 780,600 8 is dropped, it is > 5 ;

Note you must include
the 0’s
1999.5 2000.
5 is dropped it is = 5;
note you need a 4 Sig Fig
1) Retain as many significant figures in a result or in any data as
will give only one uncertain figure.
 Thus a volume which is known to be between 20.5mL and
20.7mL should be written as 20.6 mL, but not as 20.60 mL,
since the latter would indicate that the value lies between
20.59mL and 20.61mL.

 Also, if a weight, to the nearest 0.1 mg, is 5.2600 g, it should

not be written as 5.260 g or 5.26 g, since in the latter case
an accuracy of a centigram is indicated and in the former a
milligram.
2) Multiplication and Division

Find out key factor

36.63 x 0.5481 x 0.05300

100 % = 88.5470578 = 88.55
1.1689

It should have least degree of certainty

The final answer has to rounded upto number of significant figures as the key
factor has.
3) Addition and Subtraction

When adding or subtracting, retain in sum or remainder no more decimal

places than the least number entering into calculations. i.e retain only as
many decimal places as that appear in the number that has fewest decimals.

4) Rounding off
If digit following last significant figure is If the last digit is a 5, the number
is
greater than 5, the number is rounded to
rounded off to the next higher digit
the next higher digit. If it is less than
five, the number is rounded to the 8.65 =
present value of last significant figure 8.7
8.75 =
8.8
9.47 =
8.55 =
9.5
8.6

9.43 =
9.4
 Which glassware would give you the most precise
volume measurement?
A rule of thumb: read the volume to 1/10 of the smallest division.
This means that the error in reading (called the reading error) is 1/10
or 0.1 of the smallest division on the glassware

Beaker Graduated cylinder Burette

Smallest div. 10 ml Smallest div. 1ml Smallest div. 0.1 ml

Reading error = 10/10= 1ml
The volume in this beaker is
47 1
So, How many significant
figures does our volume of
47 1 mL have?
Answer - 2! The "4" we know
for sure plus the "7" we had to
estimate
Reading error = 1/10= 0.1ml
The volume in this cylinder is
36.5  0.1 mL.
So, How many significant figures
does our volume of
36.5  0.1 mL have?
Answer- 3! The "3" and the "6"
we know for sure and the "5”we
had to estimate
Reading error = 0.1/10= 0.01ml
The volume in this burette is
20.38  0.01 mL.
So, How many significant figures
does our volume of
20.38  0.01 mL have?
Answer- 4! The “2" and the “0"
and „3” we know for sure and
the “8”we had to estimate

. .