Business Research methods
Chapter 4a Sampling
1
Conten
t
1. What is a sample?
2. Stages in the selection of a sample
3. Sampling method
2
1 .What is a sample ?
• A population (universe) is any complete that
shares some common set of characteristics.
• A census is an investigation of all the
individual elements that make up the
• A sample is a subset, or some part, of a
population. The purpose of sampling is to
estimate an unknown characteristic of a
population.
3
Why Sample?
• Pragmatic reasons
– the cost, labour, and time implications.
• Accurate and Reliable Results:
– A sample may on occasion be more accurate than
a census because non- sampling errors
may increase during
• Destruction of Test Units : sampling becomes
necessary when the measurement requires
the destruction of the items being tested.
4
2. Stages in the
selection of a sample
5
1. Defining the Target Population
“ To whom do we want to talk to? ”
2. Selecting a sampling Frame
•
A list of elements from which
the sample may be drawn
•
A sampling frame error when
the entire population is not
accurately represented
6
3.3 Sampling method
1. Errors associated with sampling
Two basic causes of differences between
statistics and parameters are
– Random sampling errors
– Systematic (non sampling) error
7
8
3.3. 2 Probability versus Non probability Sampling
• The main alternative sampling plans
– probability techniques, and
– non probability techniques.
• probability sampling
– every member of the population has a known, nonzero
probability of selection.
• Non probability sampling
– the sample are selected on the basis of
personal judgment or convenience.
– The probability of any particular member of the
population being chosen is unknown.
• No appropriate statistical techniques exist for measuring
random sampling error from a non probability sample.
Therefore, projecting the data beyond the sample is,
technically speaking, statistically inappropriate.
9
Non probability sampling
Some of the non-probabilty sampling methods
include convenience , purposive , quota,
and snowball sampling.
• convenience sampling : sampling those which are
conveniently available.
• Purposive (Judgment) sampling : selecting
the sample based on expert’s judgment
• Quota sampling: sampling to ensure that
the various subgroups in a population are
represented
• Snowball sampling : selecting
respondents based on information provided
by the initial respondent.
10
Probability Sampling
• The various probability sampling methods including
simple random, systematic, stratified, cluster, and
multistage sampling.
• Simple Random Sampling
– each element in the population have an equal
chance of being selected.
– for large population sample selection is based on
tables of random numbers or computer-generated random
numbers.
• Systematic Sampling
– A sampling procedure in which a starting point is selected
by a random process; and then every nth number on the
11
list is selected
• stratified sampling:
– Subdividing a population into groups
which are internally homogeneous but have
comparative differences between groups.
– A subsample is drawn using simple random
sampling within each stratum
– either a proportional or disproportional
stratified sampling method is used
12
Cluster sampling
– primary sampling unit is not the individual element in
the population but a large cluster of elements
– clusters are selected randomly , and selection of
elements within each cluster is also
random
– Ideally a cluster should be as heterogeneous as the
population itself
– The area sample is the most popular type of
cluster sample.
multistage area sampling
– It is a sampling that involves using a combination of
two or more probability sampling techniques
– Typically, geographic areas are randomly
selected in progressively smaller (lower-population)
units. 13
What Is the Appropriate Sample Design?
• The criteria for selecting an
appropriate sample design ( prob. or
non prob.) may include
– degree of accuracy ,
– resources , and
– time.
14
Sample size determination
• Three factors are required to specify sample size:
– the heterogeneity (i.e., variance or the
standard deviation) of the population;
– the magnitude of acceptable error (i.e. ± some
amount); and
– the confidence level (i.e., 90 percent, 95
percent, 99 percent etc...).
2
• Sample size
zs
n
E
where Z is the standardized value that corresponds to the
confidence level
• S is the sample standard deviation or estimate of
the population standard deviation
• E is the acceptable magnitude of error, plus or minus
error factor (range is one-half of the total confidence
Annex: Sample size determination ( confidence level = 0.05)
16
17
18
