Академический Документы
Профессиональный Документы
Культура Документы
Atomic Spectroscopy
•Atomic Emission Spectroscopy
•Atomic Absorption Spectroscopy
Statistics and Chemometrics for Analytical Chemists
Understanding ‘basic’ Chemometrics terminology
Precision versus Accuracy
Random errors versus Systematic errors
Quality planning
Quality assurance versus Quality control
Quality improvement
Quality management
Quality Control charts e.g., Shewhart charts, X-bar charts,
Range charts and S-charts (std devn charts).
Chemical standards and reference materials
Reference material and certified reference material
Q-test
Textbook: Statistics and Chemometrics for Analytical Chemistry, 4th edn by Miller & Miller
Test 2 will be written on Monday 29 October 2018.
Time: 8-10 am or 3-5 pm
Venue: First year Chemistry lab
Some more important definitions in Statistics
Mean, x-bar: is ∑x divided by n, where n is number of data.
Mode is the score that occurs with the highest frequency. If
no value occurs more than once, then there is no mode.
When there is a tie between two values for the greatest
count, the data is said to be bimodal.
Median occupies the middle position after all the values
have been arranged in an ascending or descending order.
Sample Population
n for sample size (a lot smaller) N for population size (much bigger)
Summary on measures of central tendency
Type of average Advantages Disadvantages
A B
C D
B has very precise results but the mean is NOT accurate. C has
an accurate mean but the data points are imprecise. D is ideal.
Random Errors Systematic Errors
Random
Random errors Errors
in experimental measurements Systematic Errors
Systematic errors in experimental observations
are caused by unknown and unpredictable usually come from the measuring instruments.
Random errors in experimental measurements Systematic errors in experimental observations
changes inare
thecaused
experiment.
by unknown These changes
and unpredictable They
usuallymay
comeoccur because:
from the measuring instruments.
may occur in the measuring
changes instruments
in the experiment. or in
These changes there
They may is something wrong with the instrument or
occur because:
may occurconditions.
the environmental in the measuring instruments or in there
its is something
data handlingwrong with the
system, orinstrument or
the environmental conditions. its data handling
because system, or is wrongly used by the
the instrument
because the instrument is wrongly used by the
experimenter.
experimenter.
Random errors usually have a Gaussian
Random errors usually have a Gaussian
normal distribution
normal distribution
Examples ofExamples
causesofof random
causes errors
of random are:
errors are: Two types
Two types of systematic
of systematic error
error can can
occur withoccur with
instruments having
instruments a linear
having response:
a linear response:
electronic noise in the circuit of an electrical Offset or zero setting error - the instrument
electronic noise in the circuit of an electrical Offset or zero setting error - the instrument
instrument does not read zero when the measurant is zero.
instrument irregular changes in the heat loss rate from a does not read zero when the measurant is zero.
Multiplier or scale factor error in which the
irregular changes in thedue
solar collector heat loss rate
to changes from
in the wind.a Multiplier or scale factor
instrument consistently error in
reads changes in which
the the
solar collector due to changes in the wind. measurant ≥ than
instrument the actual changes.
consistently reads changes in the
measurant ≥ than the actual changes.
Precision is limited by random errors. The accuracy of measurements is usually
Random errors are usually determined by reduced by systematic errors.
Precision is limited by random errors. The accuracy of measurements is usually
repeating the measurements. Systematic errors are difficult to detect even for
Random errors are usually determined by reduced by systematic errors.
experienced research workers.
repeating the measurements. Systematic errors are difficult to detect even for
experienced research workers.
Random and systematic errors in routine analyses
In everyday analyses, one has to look-out for random and
systematic errors. For results to be accurate and precise,
random and systematic errors have to be kept at a minimum.
𝐬𝐭𝐝 𝐝𝐞𝐯𝐧
𝐂𝐨𝐞𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐭 𝐨𝐟 𝐕𝐚𝐫𝐢𝐚𝐭𝐢𝐨𝐧 𝐂𝐕 = • 𝟏𝟎𝟎%
𝐱𝐛𝐚𝐫
Random errors – these cause replicate results to differ from
one another, so that the individual results fall on both sides of
the mean value. Random errors affect the precision or
reproducibility of an experiment. Random errors are statistical
fluctuations (in either direction) in the measured data due to the
precision limitations of the measurement device. Random
errors usually result from the experimenter's inability to take the
same measurement in exactly the same way to get exact the
same number. Systematic errors, by contrast, are
reproducible inaccuracies that are consistently in the same
direction. Systematic errors are often due to a problem which
persists throughout the entire experiment.
Note that systematic and random errors refer to problems
associated with making measurements. Mistakes made in the
calculations or in reading the instrument are not considered in
error analysis. It is assumed that the experimenters are
careful and competent!
How to minimize random errors
Take more data. Random errors can be evaluated through statistical
analysis and can be reduced by averaging over a large number of
observations.
Errors that can be reliably estimated by repeating measurements are called
random.
How to minimize systematic errors
Systematic errors are difficult to detect and cannot be analyzed statistically,
because all of the data is off in the same direction (either to high or too low).
Spotting and correcting for systematic error takes a lot of care.
Systematic errors are errors associated with a flaw in the equipment or in
the design of the experiment. Systematic errors cannot be estimated by
repeating the experiment with the same equipment. Consider again the
example of measuring an oscillation period with a stopwatch. Suppose that
the stopwatch is running slow. This will lead to underestimation of all our
time results. Systematic errors, unlike random errors, shift the results always
in one direction.
Systematic errors are much harder to estimate than random errors. After all,
how could we have known beforehand that our stopwatch was unreliable?
Quality and Quality Management
In other words,
Quality control is about adherence to requirements. Quality assurance is
generic and does not concern the specific requirements of the product
being developed.
Quality assurance activities are determined before production work begins
and these activities are performed while the product is being developed. In
contrast, Quality control activities are performed after the product is
developed.
Quality Assurance Quality Control
What: Prevention of quality problems through What: The activities or techniques used to
planned and systematic activities including achieve and maintain the product quality,
documentation. process and service.
How: Establish a good quality management How: Finding and eliminating causes of quality
system and the assessment of its adequacy problems through tools and equipment so
and conformance audit of the operation that customer’s requirements are continually
system and the review of the system itself. met.
Quality Assurance is a complete system to Quality Control just measures and determines
assure the quality of products or services. It is the quality level of products or services. It is a
not only a process, but a complete system process itself.
including also control. It is a way of
management.
Process Product
Pro-active Reactive
Staff Function Line Function
Prevent the defects Find the defects
Quality Audit Walk through
Defining Process Testing
Selection of tools Inspection
Trainings Checkpoint Review
We provide more definitions of QA and QC because people
have different preferences. These definitions are ALL correct
because they come from qualitygurus.com so 1 is good enough.
QualityGurus.com gives many QA and QC definitions
•mean
•standard deviation (sample or population ... you decide!)
•position of warning limits (WLs)
•position of action limits (ALs)
𝟐 𝐬𝐭𝐚𝐧𝐝𝐚𝐫𝐝 𝐝𝐞𝐯𝐢𝐚𝐭𝐢𝐨𝐧
𝐖𝐚𝐫𝐧𝐢𝐧𝐠 𝐥𝐢𝐧𝐞𝐬 𝐚𝐫𝐞 𝐚𝐭: 𝐦𝐞𝐚𝐧 ±
𝐧
𝟑 𝐬𝐭𝐚𝐧𝐝𝐚𝐫𝐝 𝐝𝐞𝐯𝐢𝐚𝐭𝐢𝐨𝐧
𝐀𝐜𝐭𝐢𝐨𝐧 𝐥𝐢𝐧𝐞𝐬 𝐚𝐫𝐞 𝐚𝐭: 𝐦𝐞𝐚𝐧 ±
𝐧
The standard deviation of a sample is given a symbol s and is calculated using the following
equation:
x − mean 2
s=
n−1
Where x represents each value in the sample, x-bar is the mean value of the sample, ∑ is
the summation (or total), and n-1 is the number of values in the sample minus 1.
The standard deviation of an entire population is known as σ (sigma) and is calculated using
the following equation:
x− μ 2
σ=
N
Where x represents each value in the population, μ is the mean value of the population, ∑ is
the summation (or total), and N is the number of values in the population.
Problem on a Shewhart chart Number of
Sample
defectives
A high-level computer technology has
developed bit-sized microprocessors for use in 1 5
2 6
operating industrial ‘robots’. To monitor the 3 4
fraction of defective microprocessors produced 4 7
5 1
by a manufacturing process, 50 6 3
microprocessors are sampled each hour. The 7 6
results for 20 hours of sampling are provided in 8 5
9 4
the Table on the right. In this problem, the 10 5
number of replicates is three. 11 8
12 3
13 2
(a) Show clear calculations of warning and 14 1
action lines. 15 0
16 1
(b) Plot the Shewhart chart for these results. 17 1
(c) Make concise comments on the Shewhart 18 2
chart you have just plotted. 19 3
20 3
Does the control chart indicate any out-of-control
signals?
Use these general guide-lines with care. Furthermore, these
guide-lines apply more to Shewhart charts than to other
charts.
Introduction to Electrochemistry.
Applications of Standard Electrode Potentials.
Ecell = Ecathode - Eanode
From thermodynamics, ΔG = - RTlnK and ΔG = - nFEcell
The Nernst equation
Balancing redox equations in acidic, alkaline and unspecified
conditions.
Redox Titrations
Make sure you know how to do the necessary calculations in
redox titrations, including the plotting of the graph.
Applications of Redox Titrations.
I2 (aq) + I- ⇌ I3-
Please note that ppm = μg/g = mg/kg = mg/L so when you use
ppm be clear on your units, that is, are you saying mg/kg or
mg/L?
Key steps in the memo ... Iodometric titrations