9/4/2024

Sampling Techniques: Need

Presumed that you know: SAMPLE

A relatively dominating word in the definition of

Sample: ???
Sampling Techniques
Population ---- Sample

Sampling: What is it ---- Any Hint

Sampling Techniques: Definition Sampling Techniques: As A Process

The process of selecting a small number of elements Define the Population
from a larger defined target group of elements
such that
Determine the Sampling Frame
the information gathered from the small group
will allow judgments to be made about
Select Sampling Technique(s)
the larger groups

Determine the Sample Size

The process to select units from the population to be used in the
sample Execute the Sampling Process
 9/4/2024

Sampling Techniques: Classification Sampling Techniques: Non-Probability

Sampling Techniques
Instances in which the chances (probability) of selecting
members from the population are unknown
Non-Probability Probability
Sampling Techniques Sampling Techniques

 It does not involve random selection

Convenience Judgmental Quota Snowball
Sampling Sampling Sampling Sampling  May or may not represent the population well

 Used when researcher lacks a sampling frame for the population

Simple Random Systematic Stratified Cluster Other Sampling

Sampling Sampling Sampling Sampling Techniques

Convenience Sampling: Non-Probability Convenience Sampling: Non-Probability

Attempts to obtain a sample of convenient elements.
Error Occurs:
Often, respondents are selected because they happen to be in the right place
at the right time. 1. In the form of members of the population who are infrequent or nonusers
of that location and

 Relies upon convenience and access 2. Who are not typical in the population
 Use of students, and members of social organizations
 Mall intercept interviews without qualifying the respondents Also Known As:
 Department stores using charge account lists
Accidental Sampling / Haphazard Sampling
 “People on the street” interviews
 9/4/2024

Judgment Sampling: Non-Probability Quota Sampling: Non-Probability

A form of convenience sampling in which the population elements are May be viewed as two-stage restricted judgmental sampling.
selected based on the judgment of the researcher.
The first stage consists of developing control categories, or quotas, of
population elements.
1. Relies upon belief that participants fit characteristics
In the second stage, sample elements are selected based on convenience or
judgment.
2. must be confident that the chosen sample is truly
representative of the entire population.

Subjectivity enters in here, and certain members of the population will have a Often used to ensure that convenience samples will have desired proportion of
smaller or no chance of selection compared to others. different respondent classes

Quota Sampling: Non-Probability Quota Sampling: Non-Probability

1. Equivalent of stratified sampling. Steps To Do:
2. Like stratified sampling, the researcher first identifies the stratums and
1. The population is first segmented into mutually exclusive sub-groups.
their proportions as they are represented in the population.

3. Then convenience or judgment sampling is used to select the required 2. People are selected non-randomly according to some fixed quota.
number of subjects from each stratum.
3. Judgment is used to select the subjects or units from each segment
4. This differs from stratified sampling, where the stratums are filled by based on a specified proportion.
random sampling.
4. Convenience sampling within population groups.

Emphasizes representation of specific characteristics

 9/4/2024

Quota Sampling: Non-Probability Quota Sampling: Non-Probability

Example:
The quota sampling can be two types
Population Sample
Proportional and Non-Proportional Composition Composition

Proportional quota sampling: It is representing the major characteristics of Control

the population by sampling a proportional amount of each. Characteristic

Non-proportional quota sampling: The minimum number of sampled units Sex % % Number
in each category is specified. Male 48 48 480
Female 52 52 520
100 100 1000

Snowball Sampling: Non-Probability Snowball Sampling: Non-Probability

Remarks:
It begins by identifying someone who meets the criteria for inclusion in
the study. 1. When the desired sample characteristic is rare. It may be extremely
difficult or cost prohibitive to locate respondents in these situations.
They are asked to recommend others who they may know who also meet the
criteria. 2. As it relies on referrals from initial subjects to generate additional subjects,
this technique can dramatically lower search costs.

1. Relies upon respondent referrals of others with like 3. It comes at the expense of introducing bias because the technique itself
characteristics. reduces the likelihood that the sample will represent a good cross section
from the population.
2. Also known by the name of Referral Sampling
 9/4/2024

Others Types of Sampling: Non-Probability Others Types of Sampling: Non-Probability

Type of Purposive Sampling
Purposive Sampling
 Modal-Instance, Expert, Quota Sampling
1. Sampling with a purpose in mind
Sampling for specific types of people
2. Handpicking supposedly typical or interesting cases
 Heterogeneity Sampling
Sampling for diversity
3. Reaches a targeted sample quickly
 Snowball Sampling
4. Also known as judgmental sampling
Capitalize on informal social networks to identify
specific respondents who are hard to locate otherwise

Others Types of Sampling: Non-Probability Others Types of Sampling: Non-Probability

Heterogeneity Sampling
Modal Instance Sampling
When all opinions or views are to be included and are not concerned
1. Sampling the most frequent case, or the "typical" case about representing these views proportionately
2. Is only sensible for informal sampling contexts

Other Kinds Of Sampling

Expert Sampling
Event Sampling: Using routine or special events as the basis for sampling
It involves the assembling of a sample of persons with known or
demonstrable experience and expertise in some area Time Sampling: Recognizing that different parts of the day, week or year may be
significant
 9/4/2024

Others Types of Sampling: Non-Probability Simple Random Sampling: Probability

Other Kinds Of Sampling A sample selected such that each possible sample combination has equal
probability (nonzero) of being chosen.
Voluntary Sampling: The sample is self selected
This implies that every element is selected independently of every other
 Sample consists of people who chose themselves by responding to a general element.
appeal.

 They often over represent people with strong opinions, most often negative
opinions. Also known as Unrestricted Random Sampling.

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics

Simple Random Sampling: Probability Simple Random Sampling: Probability

Types of Simple Random Sampling
Methods of Selection:
Simple Random Sampling Without Replacement (SRSWOR)
 Lottery Method / Blind Draw Method
 In this method the population elements can enter the sample only For Example: Names “placed in a box” and then
drawn randomly
once
 Random Number Method: Table, Calculators or
 The units once selected is not returned to the population before the
Computers
next draw
For Example: All items in the sampling frame
Simple Random Sampling With Replacement (SRSWR) given numbers, numbers then drawn using
table or computer program
 The population units may enter the sample more than once

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics
 9/4/2024

Simple Random Sampling: Probability Simple Random Sampling: Probability

Advantages:
Example:
• 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 & under-sampling problems)
Not “overcome with the use of an electronic database”

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics

Systematic Sampling: Probability Systematic Sampling: Probability

Steps in Drawing a Systematic Sampling:
A method of probability sampling in which the defined target population is
Population size (N)
ordered and the sample is selected according
 Sampling interval (SI) = ------------------------------------
to position using a skip interval.
Pre-determined sample size (n)
 Calculate SI
 More efficient than simple random sampling  Select a number between 1 and SI randomly
 Use this number as the starting point to select the first sample in the list.
 A type of cluster sampling method
 Add SI to the position number of this item and the new position will be the
second sampled item
 Also called Quasi-random sampling
 Continue this process until desired sample size is reached.

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics
 9/4/2024

Systematic Sampling: Probability Systematic Sampling: Probability

Advantages:
Disadvantages:
1. Known and equal chance of any of the SI “clusters” being selected
1. Small loss in sampling precision
2. Efficient as do not need to designate every population member, just those
early on the list (unless there is a very large sampling frame). 2. Potential “periodicity” problems
3. Less expensive as faster than Simple Random Sampling
3. Decrease the representativeness of the sample, if
4. Increases the representativeness of the sample, when the ordering of the the ordering of the elements produces a cyclical
elements is related to the characteristic of interest pattern

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics

Systematic Sampling: Probability Systematic Sampling: Probability

Example: Pictorial Depiction:

1. Let there are 100,000 elements in the population

2. And a sample of 1,000 is desired
3. In this case the sampling interval, SI, is 100
4. A random number between 1 and 100 is selected
5. Let us say, this number is 23
6. The sample consists of elements 23, 123, 223,
323, 423, 523, and so on.

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics
 9/4/2024

Stratified Sampling: Probability Stratified Sampling: Probability

A method of probability sampling in which the population is separated into Remarks:
homogeneous groups/segments/strata
and samples are selected from each. The results  This method is used when the population distribution of items is skewed.
are then combined to get the picture
 It allows us to draw a more representative sample.
of the total population.
 Hence if there are more of certain type of item in the population the
Sample stratum size determination:
sample has more of this type
Proportional Method: Stratum share of total sample is stratum share of total
 And if there are fewer of another type, there are fewer in the sample.
population

Disproportionate Method: Variances among strata affect sample size for

each stratum

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics

Stratified Sampling: Probability Stratified Sampling: Probability

Remarks:
The elements within a stratum should be as homogeneous as possible, but Remarks:
the elements in different strata should be as heterogeneous as possible.
 The stratification variables should also be closely related to the
The strata should be mutually exclusive and collectively exhaustive in that characteristic of interest.
every population element should be assigned to one and only one stratum
and no population elements should be omitted.
 The variables should decrease the cost of the stratification process by
A major objective of stratified sampling is to increase precision without being easy to measure and apply.
increasing cost.
 The less the variance in a group, the smaller the sample size it takes to
produce a precise answer.

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics
 9/4/2024

Stratified Sampling: Probability Stratified Sampling: Probability

Remarks:
Advantage
In proportionate stratified sampling, the size of the sample drawn from
each stratum is proportionate to the relative size of that stratum in the • More accurate overall sample of skewed population.
total population.

Disadvantage
In disproportionate stratified sampling, the size of the sample from each
stratum is proportionate to the relative size of that stratum and to the • More complex sampling plan requiring different
standard deviation of the distribution of the characteristic of interest sample sizes for each stratum
among all the elements in that stratum.

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics

Stratified Sampling: Probability Cluster Sampling: Probability

Pictorial Depiction: 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.

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics
 9/4/2024

Cluster Sampling: Probability Cluster Sampling: Probability

Remarks:
Remarks:
 Used, when units of population are natural groups or clusters like wards, villages etc
Elements within a cluster should be as heterogeneous as possible, but clusters themselves
 The group is taken as a sampling unit should be as homogeneous as possible. Ideally, each cluster should be a small-scale
representation of the population.
 The target population is first divided into mutually exclusive and collectively exhaustive
subpopulations, or clusters.
In probability proportionate to size sampling, the clusters are sampled with probability
 Then a random sample of clusters is selected, based on a probability sampling technique proportional to size. In the second stage, the probability of selecting a sampling unit in a
such as SRS. selected cluster varies inversely with the size of the cluster.

 For each selected cluster, either all the elements are included in the sample (one-stage)
or a sample of elements is drawn probabilistically (two-stage).

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics

Cluster Sampling: Probability Cluster Sampling: Probability

Advantages Pictorial Depiction:

 Economic efficient as 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

Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics Dr. Pramod K. Gupta, Assistant Professor, Department of Biostatistics
 9/4/2024

Cluster Sampling: Probability Other Sampling: Probability

Types of Cluster Sampling: Multi- Stage Sampling

• Used in large scale investigations

Cluster Sampling • First stage- preparation of large sized sampling units
• Randomly selecting a certain number
• Second stage- Another list prepared from them
One-Stage Two-Stage Multistage
• Sub-samples drawn by random sampling
Sampling Sampling Sampling
Multi –Phase Sampling

• Used to obtain supplementary information

Simple Cluster Probability
Sampling • Certain items of information collected from all units of sample
Proportionate
to Size Sampling • Other items collected from only some of sampling units

Other Sampling: Probability Other Sampling: Probability

Area Sampling Remarks

 Divide the geo area into sectors (subareas) and give them names/numbers  To determine the total geo area

 Determine how many sectors are to be sampled (typically a judgment call),  Estimate add the counts in the subareas together
randomly select these subareas.
 And multiply this number by the ratio of the total number of subareas
 Do either a census or a systematic draw within each area. divided by number of subareas.
 9/4/2024

See You Next Time …