SAMPLING – general

• Is a statistical method of obtaining

representative data or observations from a group
(lot, batch, population or universe).

Two sampling techniques (probability & non

probability sampling):

Probability sampling also called random sampling

• Probability samples are selected in such a way to

allow representativeness of the population.
• They provide the most valid or credible results
because they reflect the characteristics of the
population from which they are selected.
• In probability sampling, each individual in the
population of interest has an equal likelihood
(chance) of selection/involvement.
• The appropriateness of this sampling method
is that it increases the likelihood of obtaining
samples that are representative of the
population.
• On the appropriate of this approach, the
researcher uses specific statistical
processes to ensure this equal probability of
every item in the population is employed.
• This method enables more reliable results of
statistical analysis with measurable margins
of errors and degree of confidence.
The most common types of random sampling
methods: (1) simple random sampling (2)
systematic sampling (3) stratified sampling and
(4) cluster sampling

Simple random sampling:

-used for small population groups that are
usually, a homogenous characterized
- the method ensures that each possible sample
has an equal probability of being selected

Example, select twenty five people (from a

homogenous group) out of two hundred fifty,
- (25:250, = 1 in 10) – means from every 10
people, select 1; to get a total of 25
 Systematic sampling
• It is a technique that allows than a simple randomly
sampling method large population
• It also affords homogeneity of the group
• Is the method frequently chosen due to its quality and
reliability.
• The items are selected from the population at a uniform
interval that may be defined in terms of: time, order/list or
space.

For example:
- An observation may be made every half an hour,
- From a long queue of people, every fourth person may be
selected, or
- In a bunch of documents, every tenth document may be
selected, or
- In a linear meters that measure a distance, the sample
occurs at every hundred meters.
To create systemic random sample, there are
seven steps:
(a) define/organize the population, preferably, put
it in homogenous;
(b) choose your sample size (which size required,
e.g. n = 20);
(c) List/arrange the population in a defined order;
(d) assign the numbers;
(e) calculate the sampling fraction = sample size
vs a population. This gives you the selection
interval;
(f) select the first unit; and
(g) Select/identify your sample.
Stratified sample:
• The entire population is divided into relatively
homogeneous groups or categories.

• Example, - all students in a certain school may be

divided into groups of boys and girls. Or classes of
contractors (I – VII), or soil on the earth is formed
in layers, etc. to remove heterogeneity

• Once this is done random sample from each of

such groups is drawn independently

• This approach is suitable when there are

identifiable sub-groups exist within the population
that differ significantly in respect of characteristic
under study.
Stratified sampling
 Cluster sampling
• Cluster sampling refers to a type of sampling
method in which the researcher divides the
population into separate groups, called
clusters.
• Then, a simple random sample of clusters is
selected from that population group.
• For example, a city may be divided in a cluster
of small localities; medium and large localities;
and a sample of these localities may be drawn
using random sampling methods.
• All the households within each of the locality
may be studied for the research.
 Non probability samples (Non-
representative samples)
• Non-probability sampling is a technique where
the samples are gathered in a process that
does not give all the individuals the equal
chances of being selected.

• As they are not truly representative, non-

probability samples are less desirable than
probability samples.
• Non random sample may be deployed when a
researcher is unable to obtain a random or
stratified sample, or it may be too expensive.

• Or, nature of the research may be of a specified

purpose, requiring this type.

• In that case, a researcher may not care about

generalizing information.

• Conclusions may be drawn as, ‘these groups

have so and so ....features’
• In using this approach, the importance is for a
researcher to minimize as many sources of
bias as possible.

• Therefore, this method depends on the

purpose of research, need a researcher to
establish and go to those who:
• qualify,
• are willing to participate,
• are available and
• are accessible.
Many names are given to represent non-
probability sampling, e.g.:
• ‘Quota sample’ - The researcher sets the
proportions or strata within the sample and
work on it.
This is generally done to ensure the inclusion of
a particular segment of the population. e.g. 3% of
certain group, 20% of other quota, 8% of the other
quota etc. are used

• Purposive sample: is considered to serve a

very specific need or purpose. A researcher
may have a specific group in mind, such as
high level organizational executives who are
targeted as a sample group.
• Snowball sampling/a subset of a purposive
is where research participants recruit other
participants for a study.

• It is used where potential participants are

hard to find, and therefore, a third party is
involved.

• The method involves a ‘primary data source’

nominating ‘other potential data sources’ that
will be able to participate in the research
studies.
• The method is purely based on referrals and it
is also called the chain-referral sampling
method.

• Snowball sampling method is extensively

used where a population is unknown and rare
and it is tough to choose subjects to
assemble them as samples for research.

• This sampling technique can go on and on,

just like a snowball increasing in size (in this
case the sample size) till the time a
researcher has enough data to analyze, to
draw conclusive results that can help an
organization make informed decisions – the
figure, next slide.
In the Figure, Research
source data start from a
single person and
continue to grow – giving
rich research data not
known previously

It’s called snowball

sampling because (in
theory) once you have the
ball rolling, it picks up
more “snow” along the
way and becomes larger
and larger.

The approach is
particularly useful in hard-
to-track populations, such
as drug users, etc.
Convenience sample:

• Definition: it is a method adopted to collect research data

from a conveniently available pool of respondents.

• It has incredibly prompt, uncomplicated, and economical.

• In many cases, members are readily approachable to be a

part of the sample.

• It uses the correct definition of everyone in the population

having equal chance of being selected.

• It allows researchers to use various sampling techniques

in situations where there are large populations, and
testing the entire community is practically impossible.
Samples of physical items in the construction
context

Attention need to be considered while looking on

samples relating to specified construction
activities such as:
• Cube tests
• Slump tests
• Soil materials
• Block walling etc.
 SAMPLING FOR CONCRETE CUBES
• Usually a minimum of 3 cubes are taken from
each sample, so ensure the sample is taken
enough from the pour before it finishes.

• Check the specification you are working with, as

sometimes, the quantity of cubes may vary. (
e.g. it is appropriate to pick 150% of the
expected amount required

• The frequency of sampling should be identified

based on the specifications. This could be per
batch /load or even per volume poured.
• Cube moulds are usually 150mm x
150mm x 15 mm (or 100mm x100mm x
10mm) and can be made from steel or
polyurethane.

• The cube moulds must be manufactured

to the specifications / standards of the
relevant body, in the UK it is the British
Standards Institute to this specification BS
EN 12390-1:2000.
• Before the concrete is scooped into the moulds,
the moulds must be lightly coated in a mould
release agent.

• This ensures that the concrete does not stick to

the mould and makes it easier to remove the cube.

• When using a 150mm mould, the concrete sample

is scooped into the mould in 3 equal layers
(50mm) and compacted between each layer.

• When using a 150mm mould, each layer

compacted is tampered using a certified
compacting rod /bar, 35 tamps per layer is
required.
• Once the 3 layers have been tampered,
tap/knock the side of the mould with a
hammer.

• Tampering and tapping removes trapped air in

the concrete and allows compaction of the
sample.

• Once complete, the concrete is levelled off

using a concrete float or trowel to give a
smooth surface flush with the top of the
mould.
Labelling
• It’s very important to uniquely identify each of
the cubes (and moulds) and to record where
they have come from, and the dates involved.
• Management must ensure labelling or
tracking the cubes to ensure properly
recording
Storage
• The cubes should be covered with a damp
cloth and a plastic sheet and stored in a dry
environment at a temperature range of 20 ± 5
degrees.
• The concrete cubes are removed from the
moulds between 16 to 72 hours, usually it is
preferable to be done after 24 hours.

• Make sure the cube ID is transferred to the cube

from the mould before placing into a curing
tank.

• The curing tank needs to operate at a

temperature between 20 ± 2 degrees and
provides a moist environment that allows the
cubes to hydrate properly.

• Ensure the cubes are fully submersed at all

times and record the tank water temperature at
least daily.
 Cube Testing
• The cubes are generally tested at 3, 7 & 28
days unless specific tests are required.

• The cubes are tested using a calibrated

compression machine.

• Usually, 1 cube will be tested at 7 days and

2 cubes at 28 days, however this may vary
depending on the requirements.
• The cubes are tested on the face
perpendicular to the casting face.
• The compression machine exerts a constant
progressing force on the cubes till they fail.
• The reading at failure is the maximum
compressive strength of the concrete.
• Once the cubes have reached failure, the
shape of the cube has been altered due to the
compression. The failure shape can indicate
whether it’s a satisfactory / unsatisfactory
failure.
 SAMPLING FRESH CONCRETE – SLUMP
TESTS
• It is very important that the concrete put into
the moulds is representative sample of the
concrete to be used.
• Never take a sample from the first or last
section of the pour, it won’t be a true
representation of the batch.
• The concrete is usually sampled after the 1 or
st

2 minutes, when concrete has been poured to

nd

ensure a good sample is taken.

• Take a few samples throughout the pour for the
best representation of the batch and take 150%
of what you think you’ll need (BS EN 12350-1).

• The sample must be a good cohesive mix.

 SLUMP TEST
If the slump test fails to meet the range limit
as dictated by the British Standard (shown in
the table below) then, the load should be
rejected.

Apparatus
• Slump cone,
• Scale for measurement,
• Temping rod (steel)
 The mold for the slump test is a frustum of a cone, 300 mm (12
in) of height. The base is 200 mm (8in) in diameter and it has a
smaller opening at the top of 100 mm (4 in).

Procedure of Concrete Slump test:

• The base is placed on a smooth surface and the container is filled with
concrete in three layers, whose workability is to be tested .
• Each layer is temped 25 times with a standard 16 mm (5/8 in) diameter steel
rod, rounded at the end.
• When the mold is completely filled with concrete, the top surface is struck off
(leveled with mould top opening)
• The mould must be firmly held against its base during the entire operation so
that it could not move due to the pouring of concrete and this can be done by
means of handles or foot - rests brazed to the mould.
• Immediately after filling is completed and the concrete is leveled, the cone is
slowly and carefully lifted vertically, an unsupported concrete will now slump.
• The decrease in the height of the center of the slumped concrete is called
slump.
• The slump is measured by placing the cone just besides the slump concrete
and the temping rod is placed over the cone so that it should also come over
the area of slumped concrete.
• The decrease in height of concrete to that of mould is noted with scale.
(usually measured to the nearest 5 mm).
 Types Of Slump
The slumped concrete takes various shapes, and according
to the profile of slumped concrete, the slump is termed as;
• Collapse Slump
• Shear Slump
• True Slump
Collapse Slump
• In a collapse slump the concrete collapses completely. A
collapse slump means that the mix is too wet or that it is
a high workability mix, for which slump test is not
appropriate.
Shear Slump
• In a shear slump the top portion of the concrete shears off
and slips sideways. OR
• If a shear or collapse slump is achieved, a fresh sample
should be taken and the test is repeated.
True Slump
• Here concrete simply keeps more or less to shape. This is
the only slump which is used in various tests.
 Slump
Degree of workability Use for which concrete is suitable
mm in

Very low 0-25 0-1 Very dry mixes; used in road making.

Low workability mixes; used for

Low 25-50 1-2 foundations with light
reinforcement..

Medium workability mixes;

manually compacted flat slabs
using crushed aggregates. Normal
Medium 50-100 2-4
reinforced concrete manually
compacted and heavily reinforced
sections with vibrations.

High workability concrete; for

sections with congested
High 100-175 4-7
reinforcement. Not normally
suitable for vibration
 SOIL SAMPLE
Picking samples for laboratory testing
• Much skills, experience and time is needed
by the soil investigation team in obtaining
the sample from the site under consideration.

Understanding the following;

• method of sampling,
• exact location with respect to plan and
elevation,
• date of sampling etc must be correctly
recorded in the laboratory.
• Importance is that, the soil should be
representative of the mass of the strata
from which they have been taken. This
involves the size of sample, method of
sampling and location of sampling,
amount of sample taken, etc.

• Undisturbed samples can be obtained by

employing some types of sample such as
sample tube or by taking sample in a trial
peat and immediately covering it with a
protective (impervious layer of wax).
Samples are taken and tested within short time of
arrival at the laboratory due to various reasons:

Storage problem
• Sometimes, it is a challenge to maintain natural
moisture content and other properties
• Demand use – sample taken may become little than
required, therefore, addition sample need to be taken
instantly
• Facilities required for conducting tests must be
considered, and
• Space requirements.

Picking samples at the relevant intervals on the area or

the length concerned to ensure representativeness, if
a need is for buildings or road stretches.
SAMPLING OF BRICKS OR BLOCK WALLING
• Ensure brick/block in manufactures using
suitable materials, well proportioned, good
workmanship is guided, product is cured
timey as required
• When the brick or block walling is already
constructed, samples of such bricks or
blocks may be picked randomly at different
locations and levels of the wall, thereby, taken
into the compressive tests in the laboratory.

• Compressive laboratory test results would

indicate their appropriateness.
 Sampling benefits in the engineering economics
Studying sampling in engineering economics helps:
• Reasonable amount of sample materials would
be taken in the laboratories and worked on
• Rather that conducting tests into the whole of
the population, that is expensive, a
representative sample serves the time & cost
efforts
• If you do not take samples, for example, in
concrete works, how can you guarantee a
strength requirement?, therefore, there could be
no confidence with the materials/items used.

[Generally, samples serve costs, because we either

deal with few which represent the whole or few
which qualify to meet the intended needs]