Sampling

Glance
 Introduction
 Sampling is a complex and technical
process. In course of daily activities we
get information make decisions and
develop predictions through sampling.
Sampling is the critical part of research
design.
 Population
 A population is the entire aggregation of
cases in which a researcher is interested
 Whatever the basic unit, the population
always comprises the entire aggregate of
elements in which the researcher is
interested
 Target population
 The accessible population
 Census

 An investigation of all the individual element that make

up the population, a total enumeration rather than a
sample
 When a population is sampled in its entirety.
 No statistical testing is necessary. All needed are
descriptive statistics.
 Census conclusions are the conclusions about the
population.
 Collect and analyze data from every possible case or
group member
 Sampling

 A range of methods that enable

researcher to reduce the amount of data
by only data from a subgroup rather than
all possible cases or elements
 Sampling frame
 List of all the sampling units from which
sample is drawn
 e.g. children < 5 years of age, households,
health care units…

Sampling scheme
 Method of selecting sampling units from
sampling frame
 Sampling unit (element)
 Subject under observation on which
information is collected
 Example: children <5 years attending
balwadi, Ayanambakkam, hospital discharges
after DHF,
Sampling fraction
 Ratio between sample size and population
size
 Example: 100 out of 2000 (5%)
Sampling: central concepts
Universe Population Final
Original
(theoretical target (empirical target sample
sample
population) population) (data

sometimes called
Sampling Sampling elements
sampling frame
units
The aim of sampling is to produce a miniature copy of the population. Each member
of
the population has an equal likelihood of being selected into the sample. Hence we
can
make inferences about the larger population based on the sample
Sampling method and Census method
Sampling method Census method
census is a
Sample refers to complete
selecting a typical or enumeration of all
representative fraction or the items in the
part of the elements in ‘universe’ of study.
the ‘universe’. Census method
Sampling method refers to selecting
requires only a all the elements in
representative part or the ‘universe’ as in
portion of the ‘universe’ the case of
for the research purposes national census
 The need to sample
 Budget constraints prevent you from
surveying the entire population
 Time constraints prevent you from
surveying the entire population
 Impracticable to survey the entire
population
 You have collected all the data but need
the results quickly
 Purpose of sampling
 To draw conclusions about populations
from samples
 To determine a population’s
characteristics by directly observing only a
portion (or sample) of the population.
 Basic sampling classifications

Probability samples Non-probability samples

Ones in which members of Instances in which the
the population have a chances (probability) of
known chance (probability) selecting members from the
of being selected population are unknown
 Process of probability sampling
 Identify a suitable sampling frame based
on your research questions or objectives
 Decide on a suitable sample size
 Select the most appropriate
sampling technique and select the
sample
 check the sample is representative of
the population
 Probability sampling methods
simple random sampling
 Simple random sampling: the probability
of being selected is “known and equal” for
all members of the population
 Blind Draw Method (e.g. names “placed in a
hat” and then drawn randomly)
 Random Numbers Method (all items in the
sampling frame given numbers, numbers then
drawn using table or computer program)
 Advantages:
 Known and equal chance of selection
 Easy method when there is an electronic
database
Disadvantages (Overcome with
electronic database)
 Complete accounting of population
needed
 Cumbersome to provide unique
designations to every population
member
 Very inefficient when applied to skewed
population distribution (over- and
under-sampling problems) – this is not
“overcome with the use of an electronic
database)
 Systematic sampling
 Systematic sampling: way to select a probability-based
sample from a directory or list. This method is at times
more efficient than simple random sampling. This is a
type of cluster sampling method.
Sampling interval (SI) = population list size (N) divided
by a pre-determined sample size (n)
 How to draw: 1) calculate SI, 2) select a number
between 1 and SI randomly, 3) go to this number as the
starting point and the item on the list here is the first in
the sample, 4) add SI to the position number of this
item and the new position will be the second sampled
item, 5) continue this process until desired sample size
is reached.
 Systematic sampling
 Advantages:
 Known and equal chance of any of the SI
“clusters” being selected
 Efficiency..do not need to designate
assign a number to every population
member, just those early on on the list
unless there is a very large sampling
frame.
 Less expensive…faster than SRS
 Disadvantages:
 Small loss in sampling precision
 Potential “periodicity” problems
 Cluster sampling
 Cluster sampling method by which the
population is divided into groups
(clusters), any of which can be considered
a representative sample. These clusters
are mini-populations and therefore are
heterogeneous. Once clusters are
established a random draw is done to
select one or more clusters to represent
the population. Area and systematic
sampling are two common methods.
 Cont . . .
 Advantages
 Economic efficiency … faster and less
expensive than SRS
 Does not require a list of all members of the
universe

 Disadvantage:
 Cluster specification error…the more
homogeneous the cluster chosen, the more
imprecise the sample results
Cluster sampling – Area method
 Divide the geographical area into sectors (sub
areas) and give them names/numbers,
determine how many sectors are to be sampled
(typically a judgment call), randomly select
these sub areas. Do either a census or a
systematic draw within each area.
 To determine the total geo area estimate add
the counts in the sub areas together and
multiply this number by the ratio of the total
number of sub areas divided by number of sub
areas.
 Cluster Sample
Companies (Clusters)
Divide Population
into Clusters
 If Managers are
Elements then
Companies are
Clusters
Randomly Select
Sample
Clusters
Survey All or a
Random Sample of
Elements in Cluster
 A two-step area cluster sample (sampling
several clusters) is preferable to a one-step
(selecting only one cluster) sample unless
the clusters are homogeneous
Stratified Sampling Method
 This method is used when the
population distribution of items is
skewed. It allows us to draw a
more representative sample. Hence
if there are more of certain type of
item in the population the sample
has more of this type and if there
are fewer of another type, there are
fewer in the sample.
 Cont . . .
 Stratified sampling: the population is
separated into homogeneous
groups/segments/strata and a sample is
taken from each. The results are then
combined to get the picture of the total
population.
 Sample stratum size determination
 Proportional method (stratum share of total
sample is stratum share of total population)
 Disproportionate method (variances among
strata affect sample size for each stratum)
. . .Cont
 Cont . . .

 Advantage:
 More accurate overall sample of skewed
population.
 Disadvantage:
 More complex sampling plan requiring
different sample sizes for each stratum
 Why is Stratified Sampling more accurate
when there are skewed populations?
 The less the variance in a group, the smaller the
sample size it takes to produce a precise
answer.
 Why? If 99% of the population (low variance)
agreed on the choice of brand A, it would be
easy to make a precise estimate that the
population preferred brand A even with a small
sample size.
 But, if 33% chose brand A, and 23% chose B,
and so on (high variance) it would be difficult to
make a precise estimate of the population’s
preferred brand…it would take a larger sample
size….
 Cont . . .

 Stratified sampling allows the

researcher to allocate a larger
sample size to strata with more
variance and smaller sample size to
strata with less variance. Thus, for
the same sample size, more precision
is achieved.
 This is normally accomplished by
disproportionate sampling.
 Stratified Sample
All Students
Divide Population
into Subgroups
 Mutually Exclusive
 Exhaustive Part-time Full-time
 At Least 1 Common
Characteristic of
Interest
Select Simple
Sample
Random Samples
from Subgroups
 Multistage sampling technique
 Primary area selection (towns)
 Sample location selection
 Chunk selection
 Segment selection
 Housing unit selection
 Non probability sampling methods

 Convenience samples: samples drawn at the convenience

of the interviewer. People tend to make the selection at
familiar locations and to choose respondents who are like
themselves.
 Error occurs 1) in the form of members of the
population who are infrequent or nonusers of that
location and 2) who are not typical in the population
 Judgment samples: samples that
require a judgment or an “educated
guess” on the part of the interviewer
as to who should represent the
population. Also, “judges” (informed
individuals) may be asked to suggest
who should be in the sample.
Subjectivity enters in here, and
certain members of the population
will have a smaller or no chance of
selection compared to others
 Referral samples (snowball samples)
samples which require respondents to
provide the names of additional
respondents
 Members of the population who are less
known, disliked, or whose opinions conflict
with the respondent have a low probability of
being selected.
 Quota samples samples that set a specific
number of certain types of individuals to
be interviewed
 Often used to ensure that convenience samples
will have desired proportion of different
respondent classes
 Online sampling techniques
 Random online intercept sampling: relies
on a random selection of Web site visitors
 Invitation online sampling: is when
potential respondents are alerted that they
may fill out a questionnaire that is hosted
at a specific Web site
 Online panel sampling: refers to consumer
or other respondent panels that are set up
by marketing research companies for the
explicit purpose of conducting online
surveys with representative samples
Guidelines for selecting a sample size (Gay and
Airasian 2003)
 For small populations with fewer than 100

people or other units, there is little point in

sampling. Survey the entire population.
 If the population size is around 500,50% of the

population should be sampled.

 If the population size is around 1,500,20%

should be sampled.
 Beyond a certain at about 5,000 units or more,

the population size is almost irrelevant, and a

sample size of 400 should be adequate.
 Developing a sample plan
 Sample plan: definite sequence of
steps that the researcher goes
through in order to draw and
ultimately arrive at the final sample
 Step 1 Define the relevant
population.
 Specify the descriptors, geographic locations,
and time for the sampling units.

 Step 2 Obtain a population list, if

possible; may only be some type of
sample frame
 List brokers, government units, customer lists,
competitors’ lists, association lists, directories,
etc.
 Step 2
 Incidence rate (occurrence of
certain types in the population,
the lower the incidence the
larger the required list needed
to draw sample from)
 Step 3: Design the sample method
(size and method).
 Determine specific sampling method
to be used. All necessary steps must
be specified (sample frame, sample
size, recontacts, and replacements)
 Step 4 Draw the sample.
 Select the sample unit and gain the
information
 Step 4
 Drop-downsubstitution
 Over sampling

 Resampling

 Step 5: Assess the sample.

 Sample validation – compare

sample profile with

population profile; check
non-responders
 Step 6: Resample if necessary.
 Errors in Sampling
Sampling frame error: error that
occurs when certain sample elements
are excluded or when the total
population is not accurately
represented in the sampling frame
Random sampling error: the
difference between the sample result
and the result of a census conducted
using identical procedures; a
statistical fluctuation that occurs
because of chance variation in the
elements selected for the sample
 Random sampling error is a function of
sample size. As sample size increase,
sampling error decreases.
 Non sampling error (systematic error):
error resulting from some imperfect aspect
of the research design that causes
response error or from a mistake in the
execution of research; error that comes
from such sources as sample bias,
mistakes in recording responses, and non
responses from persons who were not
contacted or who refused to participate.
 Type 1 error
 The probability of finding a difference with
our sample compared to population, and
there really isn’t one….

 Known as the α (or “type 1 error”)

 Usually set at 5% (or 0.05)

 Representativeness (validity)
 A sample should accurately reflect
distribution of
 relevant variable in population

 Person e.g. age, sex

 Place e.g. urban vs. rural
 Time e.g. seasonality

 Representativeness essential to generalise

 Ensure representativeness before starting,

Problems with process of sampling
 Subject withdrawal from the study

 Lost follow up

 Exclusion / inclusion criteria limit

applicability of the sample

 Incomplete data

Problem with samples

 Bias in subject recruitment

 Selectivity

 Response rate
 Conclusion

It can be said that using a sample in

research saves mainly money and time if a
suitable sampling strategy is used,
appropriate sample size selected and
necessary precautions taken to reduce on
sampling and measurement errors, then a
sample should yield valid and reliable
information.
THANK YOU