1. 3.

Significant figures, rules Precision and accuracy;

applied to significant mean, median, deviation,
figures in reporting standard deviation,
measurements and relative standard
results deviation.

TOPIC
Types and sources of error:
determinate (systematic)
and indeterminate
(random) Q-test and confidence limit

2. 4.
 A WAY TO EXPRESS

Scientific
NUMBERS IN A
FORM THAT MAKES
NUMBERS THAT ARE

Notation
TOO SMALL OR TOO
LARGE MORE
CONVENIENT TO
WRITE.
HOW ITS WORK
Rule of Scientific Notation= N x 10n

Where;
N= a number between 1 and 9 Examples
n =a positive or negative integer

1. Case n > 0
The number of atoms in 12 g of carbon is
602,200,000,000,000,000,000,000
Scientific Notation= N x 10n= 6.022 x 1023

2. Case n < 0
The mass of a single carbon atom in grams is
0.0000000000000000000000199
Scientific Notation= N x 10n=1.99 x 10-23
 Scientific Notation= N x 10n

Addition

Addition or Question:4.31 x 104 + 3.9 x 103

Subtraction Step 1: 4.31 x 104, 0.39 x 104

Step 2: 4.31 x 104 + 0.39 x 104
Steps: Step 3: 4.70 x 104

1) Write each quantity with the Subtraction

same exponent (n) Question:4.31 x 104 - 3.9 x 103

2) Combine N1 and N2 (add or Step 1: 4.31 x 104, 0.39 x 104

subtract) Step 2: 4.31 x 104 - 0.39 x 104
Step 3: 3.92 x 104
3) The exponent, n, remains the
same
Multiplication & Division
Multiplication Division
(a x 10m) x (b x 10n) = (a x b) x 10m+n (a x 10m) ÷ (b x 10n) = (a ÷ b) x 10m-n

1. Multiply a and b 1. Divide a and b

2. Add exponents m and n 2. Subtract exponents m and n

(4.0 x 10-5) x (7.0 x 103) = ? 8.5 x 104 ÷ 5.0 x 109 = ?

= (4.0 x 7.0) x (10-5+3) = (8.5 ÷ 5.0) x 104 - 9
= 28 x 10-2 = 1.7 x 10-5
= 2.8 x 10-1
01-
Significant
Figures
THE MEANINGFUL DIGITS IN
A MEASURED OR
C A LC U L AT E D Q U A N T I T Y
 Rule Examples No. of
significant
figures
ALL non-zero numbers are always 3.456 4
significant 2.81456 6
627 3
ALL zeroes in between non-zero
numbers
5.024
820.264
4
6 RULES
are significant 10011 5
Zeros to the left of the first non-zero 0.0073 2 Significant Figures
digit are not significant 0.0000001 1
0.00118 3
If a number is greater than 1, then all 69.0 3
zeros to the right of the decimal point 0.580 3
are significant 0.40072 5
Rule of Rounding Off
➢If the last digit to be removed is greater than ➢If the last digit is 5 and 2nd last digit is an
5, add one to the 2nd last digit. even num., thus the 2nd last digit does not
change.
Example:
Example:
22.486 >> 22.49
73.285 >> 73.28

➢If the last digit to be removed is smaller than

5, then the 2nd last digit does not change. ➢If the last digit is 5 and the 2nd last digit is an
Example: odd num., thus add one to the last digit.
Example:
31.392 >> 31.39
63.275 >> 63.28
Addition or Subtraction
1. The num. of digits to the right of the Addition:
decimal point in this operation should
89.332
least d.p is 1
remain. +1.1 The answer-round off to
90.4
2. The answer to this operation has a value 90.432
with the least decimal point (key number).

3.70 Subtract:
Least d.p is 2
-2.9133 The answer-round off to
0.7867 0.79
 Multiplicity

Question= 4.51 x 3.6666 (least s.f is 3)

Multiplication = 16.536366 = 16.5 (the answer must same - 3 s.f.)

or Division
Divided
The num. of significant figures in
Question= 6.8 ÷ 112.04 (lease s.f is 2)
this operation should be the same
as the num. with the least = 0.0606926 = 0.061 (the answer must same – 2 s.f)
significant figure in the data (key
number)
LOGARITHMS &
ANTILOGARITHMS
In changing from logarithms to antilogarithms and vice versa,
the number being operated and the logarithm mantissa have
the same num. of s. f.

All zeros in mantissa are significant.

Logarithm Logarithm antilogarithm

log 957 = 2.981 log 9.57 x 10-4 10 0.072 = 1.18
= -3.019
characteristic 2 -3 1
mantissa 0.981 0.019 0.18
02-EXPERIMENTAL
ERROR
N AT U R E O F S C I E N T I F I C M E A S U R E M E N T T H AT U N C E R TA I N T Y
I S A S S O C I AT E D W I T H E V E R Y Q U A N T I TAT I V E R E S U LT
Types of Errors
Determinate or systematic
errors

Indeterminate or random
errors

Gross error
 A constant error that originates from a
fixed cause, such as flaw in the design
of an equipment or experiment.
Type of error:
Determinate systematic error will affect the
or systematic accuracy of results
errors
mean of the data set will be differ
from the accepted value
Characteristics

Cause of error is Consistency, that is Will give effect to Can be corrected

known the values are accuracy of the and avoided
almost the same method
 caused by clumsiness,
carelessness or not using
Personal errors the right techniques by the
operators

Will occur if the reagents

Reagent errors used are not pure

TYPES
Caused by the nature of the
Methods errors methods used

The faulty equipments,

Instrumental errors uncalibrated weight and
glasswares
Personal Errors
Involve measurements that require personal Example
judgement.
❑estimation of a pointer between two scale
to preserve the integrity of the collected divisions.
data, who makes measurements must
❑color of solution at the end point in titration
constantly guard against personal bias
❑level of liquids with respect to a graduation
in a burette or pipette.

❑prejudice or bias in estimating

measurements: a way considered to be
unfair.
REAGENT
ERROR
Reagent containing impurities
Using impure or less pure solution
for an analytical experiment
Expired reagent
Reagents prepared incorrectly
Dilution and pipetting error
Methods errors
Source of error

❑Slow or incomplete reaction

Non-ideal analytical
methods are often
These errors are difficult
to detect.
❑Instability of reacting species
sources of systematic
errors.
❑Occurrence of side reaction

❑Interference

The most serious of the

3 types of systematic
errors.
Instrumental Errors
•Example:
✓ All measuring devices contribute to systematic
errors.
Cause by non ideal instrument
behavior, by faulty calibrations, or ✓ Glassware such as pipets, burets, and
volumetric flasks may hold volume slightly
use under inappropriate conditions different from those indicated by their
graduations. This is due to significant
difference in temperature from the calibration
temperature.
All these error can be detected and
✓ Systematic error from electronic instrument:
be corrected not calibrated instrument frequently or
calibrated incorrectly.

✓ Decreased in power supply voltage

Detection and Control
Standard reference materials
(SRM) Independent analysis Analysis of blank sample

• Analysis of standard sample • Use different method that give • Blank contains the reagents
• There are certified samples same results and solvents used in analysis
containing a known If SRM are not available, an but no analyte.
concentration or quantities of independent and largely • Reveals errors due to
particular analytes. different analysis can be used interfering contaminants from
• Can be purchased from a in parallel with the method the reagents and vessels used
number of governmental or evaluated. in analysis.
industrial; sources such as U. • A statistical test must be used
S. National Institute of to determine whether the
Standards and Technology difference is due to random
(NIST). errors in the 2 methods.
 occur at all measurements do not have
specific values and are unpredictable
Type of error:
Indeterminate Cause data to be scattered more or less
symmetrically around a mean value. This
or random error will affect the measurement precision.
errors
This error is caused by the many
uncontrollable variables in physical or
chemical measurements.
Characteristics
➢Cause of error is unknown
➢Spreads randomly around the middle value
➢Most contributors to random error cannot be positively identified.
➢It is usually impossible to measure because the error are so small that they cannot be
detected individually
➢Have effects on precision of measurement
➢Cannot be corrected-e.g: the change of humidity and temperature in the balance room that
cannot be controlled.
Type of error: Gross Error
Examples:

part of precipitate is lost before weighing,

analytical results will be low.
They usually occur
Differ from only occasionally, touching a weighing bottle with your fingers
indeterminate and may cause a result after empty mass will cause a high mass
determinate errors. to be either high or reading for a solid weighed.
low.
Lead to outliers, results in replicate
measurements that differs significantly from
the rest of the results.
 Accuracy – how close a set of measurement to true

03-Precision and value

precision – how close a set of measurement to each
accuracy other
Precision
Deviation from mean

Deviation from median

Standard deviation
Mean or average
the summation of all data and
divided by the number of
measurements (average
value)
• Where:
• X= average or mean
• Xi=individual value (sum value of all
data)
• n= number of replicate
measurements (number of data)
Mean calculation
Question: Below is the amount of chloride (%Cl) in sea water
% of chloride: 24.39, 24.20, 24.28

Equation:

Sum Xi=24.39 + 24.20 + 24.28=72.87

n=3
X= 72.87/3 = 24.29
Median
Is the middle value for a set of data

In a set of even data, the median is the In a set of odd data, the median is in the
average of the central pair, whereby the middle
values are arrange in an ascending or
e.g:
descending order
Find the median for the following data:
e.g:
5, 7, 9, 11, 12
Find the median for the following data:

5, 7, 9, 9, 11, 12
Median
Question: Below is the amount of chloride (%Cl) in sea water

% of chloride: 24.39, 24.20, 24.28

Arrange the given value:

24.20, 24.28, 24.39

Median = 24.28
Replicate % Cl- Deviation from Deviation from
Mean Median
(Xi) (Xi-mean value) (Xi-median value)
I 24.39 0.10 0.11
II 24.20 0.09 0.08
III 24.28 0.01 0.00
d = 0.02/3 d= 0.19/3
=0.07 =0.06

Question: Below is the amount of chloride (%Cl) in sea water

Deviation from % of chloride: 24.39, 24.20, 24.28,

mean= 24.29,
mean & median median= 24.28
Relative Precision (%)
Relative towards mean or median is expressed as percentages or parts per million (ppm).

Example (100%)
Relative deviation from mean = (dev. from mean/mean)X 100%
=(0.07/24.29) x 100%
= 0.29%

Example (ppm)
Relative deviation from median (ppm)= (dev. from median/median) X 106
= (0.06/24.28) X 106
= 2.47 X103 ppm

* Small amount = more precise

Standard Deviation
For n (num. of measurements) < 30 For n > 30
EXAMPLE 2.1
Question:
Calculate the mean and the standard deviation of the following set of analytical results:
15.67, 15.69, 16.03 g.
Solution
Xi Xi-X (Xi-X)2
15.67 0.13 0.0169
15.69 0.11 0.0121
16.03 0.23 0.0529
Σ 47.39 Σ 0.0819
n= 3
mean= 15.80
s= √0.0819/2
s= 0.202
Example 2.2
A student performed four titrations of a standard NaOH to determine its conc. The vol. Of NaOH used
are as given in the following data.

titration 1 2 3 4
Vol. NaOH 22.6 22.1 23.7 22.4
(mL)
Determine the mean, median and range of data set.
Calculate the deviation for mean
Calculate relative deviation from mean in %
Calculate the SD of the data
Example 2.2
Determine the mean, median and range of data set. Xi Deviation from (Xi-X)2
Mean
Mean, X=22.7 [Xi-mean value]

Median= 22.1, 22.4, 22.6, 23.7=22.5 22.6 0.1 0.01

22.1 0.6 0.36

Range= 23.7-22.1=1.6
23.7 1.0 1.00
Calculate the deviation for mean= 0.5
22.4 0.3 0.09
Calculate relative deviation from mean in %
X= D= 2/4= 0.5 1.46
Relative deviation from mean = (dev. from mean/mean)X 100% Xi/n=90.8/4
=22.7
=(0.5/22.7) x 100%
=2.2%
Calculate the SD of the data= (1.46/3)=
Ways of
Describing
Accuracy Absolute Error, Relative Error
E
Errors will show the closeness of
measurements to the accepted or
true value.

Relative
Accuracy
Absolute Error, E
Definition: The difference between the true value and the measured value

E = xi – xt

Where
xi = measured value
xt = true or accepted value

Example: If 2.62 g sample of material is analyzed to be 2.52 g, so the absolute error is − 0.10g.
xi = 2.52 g
xt = 2.62 g
Relative Error
The absolute or mean error expressed as a percentage of the true value.

Er = xi – xt x 100%
xt

Also can be expressed in parts per thousand. The above analysis has a relative error of

− 0.10 g x 100% = -3.8% or -38 ppt

2.62 g

Example: If 2.62 g sample of material is analyzed to be 2.52 g

xi = 2.52 g
xt = 2.62 g
Relative Accuracy
Definition: The measured value or mean expressed as a percentage of the true value.

Er = xi x 100%
xt

The above analysis has a relative accuracy of

2.52 g x 100% = 96.2 %

2.62 g

Example: If 2.62 g sample of material is analyzed to be 2.52 g

xi = 2.52 g
xt = 2.62 g
04- Confident Limit:
The objective of a measurement or analysis is to determine the value that is possibly true value.
How sure are you?
Even if the true value is not known, the interval can be expressed within the mean value. This
interval is known as confidence interval (C.I)
Confidence interval: the range on both sides of the mean in which the probability of the
true value exists.
Confidence limit: the limit of the confidence interval
Confidence level: the likelihood that the true value falls within the confidence limit. Often
expressed as percentage (%).
Confident limit equation
Confidence limit =
Where:
t= a statistical factor that depends on the num. of degrees of freedom and the confidence level
desired.
* the num. of degrees of freedom is one less than the num. of measurements.
s= standard deviation

N= num. of determinations (data)

X=mean
Value of t
Example 2.3
The true mean and standard deviation obtained from the 5 times measurements of
ascorbic acid in vitamin C tablet is 68.35% and 0.12%. Calculate the confidence limit at
95% confidence interval.

Confidence limit= 68.35% ± 2.78 (0.12%)/√5

= 68.35% ± 0.14%
= 68.20% and 68.50%

Confidence interval= 68.20% < μ < 68.50%

μ = true value of ascorbic acid content in vitamin C tablet

Example 2.3
The true mean and standard deviation obtained from the 5 times measurements of
ascorbic acid in vitamin C tablet is 68.35% and 0.12%. Calculate the confidence limit at
95% confidence interval.

Confidence limit= 68.35% ± 2.78 (0.12%)/√5

= 68.35% ± 0.14%
= 68.20% and 68.50%

Confidence interval= 68.20% < μ < 68.50%

μ = true value of ascorbic acid content in vitamin C tablet

 Used for identification of outlier
Num. of measurements (n) is 3 to 10
used sparingly and never more than once in a data set
The data must be arranged in increasing order values
Formula for Q-test
Qc = gap / range
04- where

Rejection of ‘gap’ = different between the outlier and the closest

number to it.

Data: Q-test Range =highest value-lowest value

The calculated Qc is compared with a rejection

quotient;
If Qc≥Q0.90, reject the result being test.
Value of Q
Example 2.4
Question:
Volume of 0.1 M NaOH used for the titration with 0.1 M HCl (repeated for 4 times, n=4)
27.33, 27.56, 27.37, 27.30
Step 1: Arrange in increasing order: 27.30, 27.33, 27.37, 27.56 ml
Step 2- Find Outlier = 27.56 ml
Step 3-Calculate Qc:
Qc= gap = 27.56-27.37 = 0.731
range 27.56-27.30

With 4 observations, Qc (0.731)<Q0.90 (0.765), therefore the value of 27.56 at 90% confidence was keep in
the data
Example 2.5
Question:

Given the data: 26.20, 26.24, 26.21, 26.32, 26.22

Identify a possible outlier in this set of data.

Used the Q-test to determine whether it can be retained or rejected at 95% Confidence level.
Example 2.5
Question:

Given the data: 26.20, 26.24, 26.21, 26.32, 26.22

◦ Step 1: arrange in increasing order: 26.20, 26.21, 26.22, 26.24, 26.32

Identify a possible outlier in this set of data.

◦ Step 2: outlier= 26.32

Used the Q-test to determine whether it can be retained or rejected at 95% Confidence level.
◦ Q= gap/range= (26.32-26.24)/ (26.32-26.20)= 0.08/0.12=0.667 < Q at 95%, retained
EXAMPLE
Consider the data set
39.25, 39.83, 39.61, 39.68
1. Find mean
2. Find SD
3. 95% confidence limit for the mean
4. If the accepted value for the set of data is 39.50, calculate the value for the absolute error
5. relative error in %
6. Determine whether 39.25 is an outlier or otherwise at 95% confidence level.
Thank You
DR. NUR RAHIMAH SAID