18/03/2022

Experimental Uncertainties
Errors in Chemical Analysis
• No measurement is perfectly
accurate or exact. Many
instrumental, physical and human
limitations cause measurements to
deviate from the "true" values of the
quantities being measured. These
deviations are called "experimental
uncertainties," but more commonly
the shorter word "error" is used.

1 4

Steps in a Typical Quantitative

Analysis
Terms and Definitions

• Replicates - two or more determinations on

the same sample
– the more replicates and trials you have, the
better
Example: One student measures Fe (III)
concentrations six times.
• The results are listed below:
– 19.4, 19.5, 19. 6, 19.8, 20.1, 20.3 ppm (parts
per million)
• 6 replicates = 6 measurements

2 5

Terms and Definitions

• Data of unknown quality are useless!
• All laboratory measurements contain • Mean: average or arithmetic mean
N
experimental error.  x i
x  i 1

N
• Median: the middle value of replicate data;
• It is necessary to determine the make sure that the array of numbers are in
magnitude of the accuracy and ascending order
reliability in your measurements. – If an odd number of replicates, the middle
value of replicate data
• Then you can make a judgment about
– If an even number of replicates, the middle two
their usefulness. values are averaged to obtain the median

3 6

1
 18/03/2022

Calculation: Mean and Median Calculation: Mean and Median

• Example I: measurements of Fe (III) concentrations:
• 19.4, 19.5, 19. 6, 19.8, 20.1, 20.3 ppm (parts per • 5 replicates: An odd number of
million)
• What are the mean and median of these measurements
replicates !!!
• Mean = 19.4 + 19.5 + 19.6 + 19.8 + 20.1 + 20.3 • Rearrange the number from the low to
6 the high:
= 19.78 ppm = 19.8 ppm • 19.4, 19.5, 19. 6, 19.8, 20.1
• 6 replicates
• An even number of replicates !!!
• Median = 19.6 ppm
• Median = 19.6 + 19.8
2
= 19.7 ppm

7 10

Calculation: Mean and Median Terms and Definitions

• Example II: measurements of Fe (III)
• Precision - describes the reproducibility of
concentrations: 19.4, 19.5, 19. 6, 19.8, measurements.
20.1 ppm (parts per million) • How close are results which have been obtained in
• What are the mean and median of these exactly the same way?
measurements • The reproducibility is derived from the deviation
• Mean = 19.4 + 19.5 + 19.6 + 19.8 + 20.1 from the Mean:
• Deviation from the mean
5 – Standard deviation
= 19.68 ppm = 19.7 ppm – Variance
• An odd number of replicates !!! – Coefficient of variation
• Median = 19.6 ppm – range

8 11

Calculation: Mean and Median Terms and Definitions

• Example III: measurements of Fe (III)
concentrations:19. 6, 19.4, 20.1, 19.5, 19.8 • Absolute Deviation (D) of an
ppm (parts per million) element of a data set is the absolute
• What are the mean and median of these difference between that element and a
measurements given point. Typically the point from
• Mean = 19.4 + 19.5 + 19.6 + 19.8 + 20.1 which the deviation is measured is the
5 value of either the median or the
= 19.68 ppm = 19.7 ppm mean of the data set.
Number mean

9 12

2
 18/03/2022

Example:
Terms and Definitions Find the standard deviation of 4, 9, 11, 12, 17, 5, 8, 12,
14
• Average absolute deviation or Average First work out the mean: 10.222
deviation or mean absolute deviation is Now, subtract the mean individually from each of the
the average of the absolute deviations and numbers in the question and square the result. This is
is a summary statistic of statistical equivalent to the (x - xbar)² step. x refers to the
values in the question.
dispersion or variability.
x 4 9 11 12 17 5 8 12 14
• Sum up all the absolute deviation (x – x )² 38.7 1.49 0.60 3.16 45.9 27.3 4.94 3.16 14.3
• No need to divide it to the number of Now add up these results (this is the 'sigma' in the
results of measurements formula): 139.55
Divide by n-1. n is the number of values, so in this case
is 8: 17.44
And finally, square root this: 4.18 (4 is the final answer)

13 16

Terms and Definitions Terms and Definitions

• Standard Deviation (sd) is a measure • Variance is a measure of statistical

of the dispersion of a collection of dispersion
numbers • Variance is the average squared
deviations about the mean. Thus,
variance is the square of the standard
deviation.
• sd2

14 17

Standard Deviation Terms and Definitions

• measures the spread of the data about the mean • Coefficient of Variation (CV) is a normalized
value. It is useful in comparing sets of data measure of dispersion of a probability
which may have the same mean but a different distribution. It is defined as the ratio of the
range. standard deviation to the mean
• Example, the mean of the following two is the • It should be expressed in percentage, so you have
same: 15, 15, 15, 14, 16 and 2, 7, 14, 22, 30 to multiply it by 100
• However, the second is clearly more spread out.
If a set has a low standard deviation, the values
are not spread out too much.
• The smaller standard deviation, the closer the
results • Range – highest value minus the lowest
value

15 18

3
 18/03/2022

Problem Terms and Definitions

The following results were obtained for the • Outlier - Occasional error that
analysis of aspirin in aspirin tablets. obviously differs significantly from
Determination % Aspirin w/w the rest of the results.
• 1 91.30
• 2 91.25
• 3 91.20
• 4 91.31
• 5 91.26

19 22

Problem Calculation: Mean and Median

• Example I: measurements of Fe (III) concentrations:
Calculate • 19.4, 19.5, 19. 6, 19.8, 20.1, 20.3 ppm (parts per
• a) The mean of the data set million)
• b) The median of the data set • What are the mean and median of these measurements
• Mean = 19.4 + 19.5 + 19.6 + 19.8 + 20.1 + 20.3
• c) The absolute deviation from the mean of
6
the 3rd determination
= 19.78 ppm = 19.8 ppm
• d) The absolute average deviation from the • 6 replicates
mean of the data set
• An even number of replicates !!!
• e) standard deviation • Median = 19.6 + 19.8
• f) variance 2
• g) coefficient of variation = 19.7 ppm

20 23

Terms and Definitions Mean and True Value

• Accuracy - the closeness of the measurement to the true or
accepted value.
• This "closeness" called as the error:
absolute or relative error of a result to its true value.

Accuracy & Precision

__
Mean : X Xt = true value

21 24

4
 18/03/2022

Accuracy and Precision Terms and Definitions

• Relative Error (Er) - the absolute error
corrected for the size of the measurement
or expressed as the fraction, %, or parts-
per-thousand (ppt) of the true value. Er
has a sign, but no units.
Er = xi − xt
xt
• parts per hundred (pph) = Er x100
• parts per thousand (ppt) = Er x1000
Relative Average Error (RAE)

25 28

Calculations: Absolute and

Terms and Definitions
Relative Error
• Example: measurements of Fe concentrations:
• Precision: Described by the standard
• 19.4, 19.5, 19. 6, 19.8, 20.1, 20.3 ppm
deviation, the variance and the
• Assumed we already knew the true value of Fe (III)
coefficient of variation (all are function of concentration at 20.0 ppm. What are absolute and
the deviation from the mean). relative errors of each measurement?
• Accuracy: Described by the absolute and • E = 19.4 - 20.0 = -0.6 ppm Er =(-0.6/20)x100% = - 3%
• E = 19.5 - 20.0 = -0.5 ppm Er =(-0.5/20) x100%= -2.5% ~ -3%
relative error.
• E = 19.6 - 20.0 = -0.4 ppm Er =(-0.4/20) x100% = -2%
• E = 19.8 - 20.0 = -0.2 ppm Er =(-0.2/20) x100% = -1%
• E = 20.1 - 20.0 = 0.1 ppm Er=(0.1/20)x100% = 0.5%
• E = 20.3 - 20.0 = 0.3 ppm Er = (0.3/20)x100% =1.5% ~ 2%

26 29

Terms and Definitions Problem

• Absolute Error (E) - the difference between A standard, National Bureau of Standards
the experimental value and the true value. (NBS), sample known to contain 21.37%
Has a sign and experimental units: arsenic was analyzed with the following
E = xi − xt results.
= Experimental value – true (acceptable) Determination % Arsenic
value • 1 21.40
Absolute Average Error (AAE) • 2 21.35
= ∑ (xi − xt) • 3 21.32
• 4 21.39
• 5 21.30

27 30

5
 18/03/2022

Systematic or Determinate
Problem
Errors
Calculate Potential Instrument Errors
• a) The mean of the data set • Variation in temperature
• b) The median of the data set
• Contamination of the equipment
• c) The absolute error of the 2nd
determination • Power fluctuations
• d) The absolute average error of the data set • Component failure
• e) The relative error of the 4th determination All of these can be corrected by
in ppt calibration or proper instrumentation
• f) The relative average error of the data set maintenance.
in ppt

31 34

Systematic or Determinate
Types of Errors
Errors
• Systematic or determinate errors Method Errors
affect accuracy! • Slow or incomplete reactions
• Random or indeterminate errors • Unstable species
affect precision! • Nonspecific reagents
• Gross errors or blunders lead to • Side reactions
outlier’s and require statistical
techniques to be rejected. These can be corrected with proper
method development.

32 35

Systematic or Determinate Systematic or Determinate

Errors Errors
• 1. Instrument errors - failure to calibrate, Personal Errors
degradation of parts in the instrument, power • Misreading of data
fluctuations, etc.
• Improper calibration
• 2. Method errors - errors due to no ideal
physical or chemical behavior - • Poor technique/sample preparation
completeness and speed of reaction, • Personal bias
interfering side reactions, sampling • Improper calculation of results
problems
• 3. Personal errors - occur where These are blunders that can be
measurements require judgment, result from minimized or eliminated with proper
prejudice, color acuity problems training and experience

33 36

6
 18/03/2022

The Effect of Systematic Error - normally "biased” and

often very "reproducible".
Normal or Gaussian Curve
1. Constant errors - Es is of the same magnitude, • named after Carl Friedrich Gauss
regardless of the size of the measurement.
• This error can be minimized when larger samples
• characteristic symmetric "bell shape
are used. In other words, the relative error decreases curve" that quickly falls off towards
with increasing amount of analyte. plus/minus infinity
• Er = (Es/Xt )x100%
• eg. Solubility loss in gravimetric analysis
• eg. Reading a buret
2. Proportional errors - Es increases or decreases
with increasing or decreasing sample size,
respectively. In other words, the relative error
remains constant.
• Typically a contaminant or interference in the
sample

37 40

Detection of Systematic
Method Errors Normal or Gaussian Curve

• 1. Analysis of standard samples

• 2. Independent Analysis: Analysis
using a "Reference Method" or
"Reference Lab"
• 3. Blank determinations
• 4. Variation in sample size: detects
constant error only

38 41

Random Errors Gross Errors

• caused by uncontrollable variables which • gross errors cause an experimental value to
normally cannot be defined be discarded.
• The accumulated effect causes replicate • Lead to outlier’s and require statistical
measurements to fluctuate randomly around techniques to be rejected.
the mean. • Examples of gross error are an obviously
• Random errors give rise to a normal or "overrun end point“ (titration), instrument
Gaussian curve. breakdown, loss of a crucial sample, and
• Results can be evaluated using statistics discovery that a "pure" reagent was
• Usually statistical analysis assumes a actually contaminated.
normal distribution • We do NOT use data obtained when gross
error has occurred during collection.

39 42