Sampling Methods

Important statistical terms

Population:
a set which includes all
measurements of interest
to the researcher
(The collection of all
responses, measurements,
or counts that are of interest)

Sample:
A subset of the population
Why sampling?

Get information about large populations

 Less costs
 Less field time
 More accuracy i.e. Can Do A Better
Job of Data Collection
 When it’s impossible to study the
whole population
 Target Population:
The population to be studied/ to which the
investigator wants to generalize his results
Sampling Unit:
smallest unit from which sample can be
selected
Sampling frame
List of all the sampling units from which sample
is drawn
Sampling scheme
Method of selecting sampling units from
sampling frame
Defining the Target Population
 It is critical to the success of the
research project to clearly define
the target population.

 Rely on logic and judgment.

 The population should be defined in

connection with the objectives of
the study.
Technical Terminology
 An element is an object on which a
measurement is taken.
 A population is a collection of elements
about which we wish to make an
inference.
 Sampling units are nonoverlapping
collections of elements from the
population that cover the entire
population.
Technical Terms
 A sampling frame is a list of sampling
units.
 A sample is a collection of sampling units
drawn from a sampling frame.
 Parameter: numerical characteristic of a
population
 Statistic: numerical characteristic of a
sample
Errors of nonobservation
 The deviation between an estimate
from an ideal sample and the true
population value is the sampling
error.
 Almost always, the sampling frame
does not match up perfectly with the
target population, leading to errors of
coverage.
Errors of nonobservation
 Nonresponse is probably the most serious
of these errors.
 Arises in three ways:
 Inability
of the person responding to
come up with the answer
 Refusal to answer
 Inability
to contact the sampled
elements
Errors of observation
 These errors can be classified as
due to the interviewer, respondent,
instrument, or method of data
collection.
Interviewers
 Interviewers have a direct and dramatic
effect on the way a person responds to a
question.
 Most people tend to side with the view
apparently favored by the interviewer,
especially if they are neutral.
 Friendly interviewers are more successful.
 In general, interviewers of the same gender,
racial, and ethnic groups as those being
interviewed are slightly more successful.
Respondents
 Respondents differ greatly in motivation
to answer correctly and in ability to do so.
 Obtaining an honest response to sensitive
questions is difficult.
 Basic errors
 Recall bias: simply does not remember
 Prestige bias: exaggerates to ‘look’ better
 Intentional deception: lying
 Incorrect measurement: does not understand
the units or definition
Census Sample
 A census study occurs if the entire
population is very small or it is
reasonable to include the entire
population (for other reasons).

 It is called a census sample because

data is gathered on every member
of the population.
Why sample?
 The population of interest is usually
too large to attempt to survey all of
its members.
 A carefully chosen sample can be
used to represent the population.
 The sample reflects the characteristics
of the population from which it is
drawn.
Probability versus Nonprobability
 Probability Samples: each member of
the population has a known non-zero
probability of being selected
 Methods include random sampling, systematic
sampling, and stratified sampling.

 Nonprobability Samples: members are

selected from the population in some
nonrandom manner
 Methods include convenience sampling,
judgment sampling, quota sampling, and
snowball sampling
Methods used in probability samples

 Simple random sampling

 Systematic sampling
 Stratified sampling
 Multi-stage sampling
 Cluster sampling
Random Sampling
Random sampling is the purest form of
probability sampling.
 Each member of the population has an equal and
known chance of being selected.
 When there are very large populations, it is often
‘difficult’ to identify every member of the
population, so the pool of available subjects
becomes biased.
 You can use software, such as minitab to generate
random numbers or to draw directly from the
columns
Simple random sampling
Table of random numbers

684257954125632140
582032154785962024
362333254789120325
985263017424503686
 Systematic Sampling
 Systematic sampling is often used instead
of random sampling. It is also called an Nth
name selection technique.
 After the required sample size has been
calculated, every Nth record is selected from a
list of population members.
 As long as the list does not contain any
hidden order, this sampling method is as good
as the random sampling method.
 Its only advantage over the random sampling
technique is simplicity (and possibly cost
effectiveness).
Systematic sampling
 Stratified Sampling
 Stratified sampling is commonly used
probability method that is superior to random
sampling because it reduces sampling error.
 A stratum is a subset of the population that share
at least one common characteristic; such as
males and females.
 Identify relevant stratums and their actual
representation in the population.
 Random sampling is then used to select a sufficient
number of subjects from each stratum.
 Stratified sampling is often used when one or more
of the stratums in the population have a low
incidence relative to the other stratums.
 Cluster Sampling
 Cluster Sample: a probability sample in which
each sampling unit is a collection of elements.
 Effective under the following conditions:
 A good sampling frame is not available or costly,
while a frame listing clusters is easily obtained
 The cost of obtaining observations increases as the
distance separating the elements increases

 Examples of clusters:
 City blocks – political or geographical
 Housing units – college students
 Hospitals – illnesses
 Automobile – set of four tires
 Cluster sampling
Section 1 Section 2

Section 3

Section 5

Section 4
Convenience Sampling
 Convenience sampling is used in
exploratory research where the
researcher is interested in getting an
inexpensive approximation.
 The sample is selected because they are
convenient.
 It is a nonprobability method.
 Often used during preliminary research efforts
to get an estimate without incurring the cost or
time required to select a random sample
Judgment Sampling
 Judgment sampling is a common
nonprobability method.
 The sample is selected based upon
judgment.
 an extension of convenience sampling

 When using this method, the researcher

must be confident that the chosen
sample is truly representative of the
entire population.
Quota Sampling
 Quota sampling is the nonprobability
equivalent of stratified sampling.

 First identify the stratums and their

proportions as they are represented in
the population

 Then convenience or judgment sampling

is used to select the required number of
subjects from each stratum.
 Snowball Sampling
 Snowball sampling is a special nonprobability
method used when the desired sample
characteristic is rare.
 It may be extremely difficult or cost prohibitive
to locate respondents in these situations.
 This technique relies on referrals from initial
subjects to generate additional subjects.
 It lowers search costs; however, it introduces
bias because the technique itself reduces the
likelihood that the sample will represent a good
cross section from the population.
Sample Size?
 The more heterogeneous a population is,
the larger the sample needs to be.
 Depends on topic – frequently it occurs?
 For probability sampling, the larger the
sample size, the better.
 With nonprobability samples, not
generalizable regardless – still consider
stability of results
Response Rates
 About 20 – 30% usually return a
questionnaire
 Follow up techniques could bring it up to
about 50%
 Still, response rates under 60 – 70%
challenge the integrity of the random
sample
 How the survey is distributed can affect
the quality of sampling