Chapter 3 Flashcards
▸ Sample type
▸ Size of sample
▸ Sample preparation needed
SELECTING A METHOD
▸ Sample type/ homogeneity/ size
▸ Sampling statistics/ errors
OBTAINING A REPRESENTATIVE SAMPLE
a ‘portion of material selected
from a larger quantity of material
SAMPLE
▪ Homogeneous
▪ Heterogeneous
▪ Static (contained) system
▪ Dynamic system
Representative Sample
▪ Simple Random Sampling
▪ Stratifies Random
Sampling
▪ Systematic Sampling
Random Sample
a sample that is typical of the
parent material for the
characteristic under
inspection
Representative Sample
▸ the parent material is changing with respect to time
▸ removal of a portion at any instant represents only a ‘snapshot’ of
that moment in time and in that particular location.
* The fact that it can never be reproduced presents difficulties in
applying statistical control and consequently cannot be the subject
of conventional statistical sampling plans
Dynamic system
▸ the composition of the parent material is permanent with
respect to position in space and stable over the time of
sampling and testing
▸ e.g. oil in a drum; tins of fruit in a warehouse
Static (contained) system
a sample deliberately chosen by
using a sampling plan that
screens- out materials with certain
characteristics and/or selects only
material with other relevant
characteristics
Selective Sample
(directed/ focused sampling)
a sample is selected by a
random process to eliminate
problems of bias in selection
and/or to provide a basis for statistical interpretation of measurement data.
Random Sample
3 Types of Random Sampling
Simple random
sampling, Stratified random
sampling, Systematic sampling
Any sample has an
equal chance of
selection
. Simple random
sampling
the lot is subdivided/
stratified and a simple
random sample selected
from each stratum.
Stratified random
sampling
The first sample is
selected at random and
then the subsequent
samples are taken
according to a previously
arranged interval, e.g.
every 5th, 10th or
whatever is appropriate
Systematic sampling
▸ consists of two or more portions of
materials (collected at the same
time) selected so as to represent
the material being investigated
▸ taken in proportion to the amount
of the material that they represent
Composite Sample
can take a number of forms (a single pile of
grain, soil present in an area of contaminated land or a
shipload of coal.
- not in the form of separate permanently
identifiable units.
Bulk materials
comprised of identifiable units, which may
be assigned number.
Packaged goods
bulk materials that are partially packaged into
smaller units, such as in bags or drums.
Segments
(for both bulk materials and packaged goods)
- a quantity of material transferred on one occasion
and covered by a single set of shipping documents.
- may be made up of one or more lots
Consignment
a quantity of material which is assumed to represent a
single population for sampling purposes
- may consist of one or more batches
Lot
a quantity of material which is known or assumed to
have been produced under uniform conditions
Batch
portions of the material obtained from the lot/batch
by using a sampling device
- often combined to produce a primary or gross sample
Increments
portion of material delivered to the
laboratory for analysis
- often obtained from the primary or composite
sample by a series of division and reduction processes (e.g. coning
and quartering, riffling
Laboratory sample
intermediate sample that the analyst
obtains from the laboratory sample
Test/ analytical sample-
the quantity of material that is
actually submitted for analysis, taken from the test sample
- has to go through a number of treatment steps
before the final aliquot is obtained
Test portion/ analytical portion
❖ describes when, where and how samples are to be taken.
❖ IUPAC: A predetermined procedure for the selection, withdrawal,
preservation, transportation and preparation of the portions to be
removed from a population as samples
SAMPLING PLAN
▸ should specify the number and size of primary samples to be
obtained from the lot/batch
▸ should include the number, location and size of the portions to be
taken, and instructions for the extent of compositing and reduction
of these portions to produce the laboratory sample
▸ should address whether the process of sampling should be a ‘oneoff’ or whether it should be repeated and if so, how often
▹ sampling scheme/ sampling schedule- sampling plan with
regular requirement for analysis
▹ sampling programme - describes a combination of procedures
where several related sampling schemes are combined
TYPES OF SAMPLING
Probability Sampling
Bulk Sampling
Acceptance Sampling
Non-Probability Sampling
allows a statistical evaluation to
be applied to the data.
▸ used when a representative
sample is required
Probability Sampling
(random sampling)
involves taking increments from the bulk
material in such a way that any portion of
The bulk has an equal probability of being sampled.
▸ used when little information is available about the material
that is being sampled, or
▸ sampling from a batch or consignment of manufactured
product for quality control purposes
Simple random sampling
requires the consignment to be subdivided
into groups (strata) according to predefined
criteria. A simple random sampling plan is then applied to
each stratum. The number of samples taken from each
stratum is proportional to its size (e.g. weight or volume).
▸ aims to produce a more representative sample than would
be obtained by simple random sampling.
Stratified sampling
most commonly used sampling technique.
▸ involves taking increments from the bulk material at
predetermined intervals, as defined by the sampling
plan.
Systematic sampling
▸ used when a representative
sample cannot be collected or is
not appropriate
▸ the correct sampling approach
to use to produce a selective
sample
Non- Probability Sampling
use of knowledge
about the material to
be sampled, and the
reason for sampling, to
select specific samples
. Judgement sampling
requires the consignment to be subdivided into groups (as for stratified sampling) then judgement sampling is used to select samples from each group
Quota sampling
involves selecting
samples on the basis
of availability and/or
accessibility
Convenience
sampling
involves the taking of a sample
from material which does not
consist of discrete, identifiable
or constant units
▸ bulk material may be gaseous,
liquid or solid.
Bulk Sampling
involves the application of a
predetermined sampling plan to
decide whether a batch of goods
meets the defined criteria for
acceptance
Acceptance Sampling
Examples of measurement objectives
✓ Control of limits, such as product specifications, environmental quality standards
✓ (EQS), maximum contaminant levels (MCL) or maximum residue limits (MRL)
✓ Investigation of trends (variation in time) or distributions (variation in space)
The quality requirements are in many cases set as the analytical quality
✓ HORRATR –value (requirement on between laboratory precision) of less than 1.5 in
interlaboratory method study
✓ For environmental control in Denmark: maximum standard deviation at low
concentrations, maximum relative standard deviation at higher concentrations,
maximum bias and maximum error, values set individually and for quality classes
Conventionally, arbitrary quality requirements have been applied
✓ Limit of detection (LOD) below 10% of limit, repeatability standard deviation better
than 5%, bias less than 20% and all measurements within linear range
How can knowledge on the type of sampling error help in designing the sampling?
The major source of uncertainty in contaminant measurements
for a row of soil piles from the previous example was shown to be
varying contaminant concentration with depth, i.e. a systematic
effect. Sampling or sub-sampling of an increased number of
replicates would not have reduced uncertainty because the
effect of doing so is mostly limited for systematic errors.
Instead, a more suitable sampling method was designed, as
described