TOPIC 1B :

SAMPLING TECHNIQUES
 Objectives

1. Concept of sampling

2. Types of sampling techniques

3. Some useful tips in sampling

4. Method of data collection

 Definition : Concept of sampling

➢ A process of selecting units from a population

➢ A process of selecting a sample to determine certain

characteristics of a population
 Concept of sampling
Why sample needed?

➢ Economy

➢ Timeliness

➢ The large size of many populations

➢ Inaccessibility of some of the population

➢ Destructiveness of the observation – accuracy

➢ In most cases, census is unnecessary!

 Sampling Error

Sampling error is any type of bias

that is attributable to mistakes
in either drawing a sample or
determining the sample size
 Sampling Frame

➢ Definition: A complete and updated list of

population elements (people, companies, houses,
cities, etc.) from which units to be sampled can be
selected.

➢ Difficult to get an accurate list.

➢ Sample frame error occurs when certain elements

of the population are accidentally omitted or not
included on the list.
 Sampling Methods

Probability Nonprobability
Sampling sampling

- Element will not be selected

- Element will be selected randomly randomly
- Unbias - Bias (Most of the time the result is
- Sampling frame is needed inaccurate)
- Sampling error can be calculated - Sampling frame is not needed
- Sampling error cannot be calculated
 Type of Sampling Methods
Probability Sampling Non-probability
( Random selection) (does not involve random selection)
1. Simple random sampling 1. Convenience sampling
2. Systematic random 2. Quota sampling
sampling 3. Snowball sampling
3. Stratified random
sampling
4. Cluster sampling
5. Multi-stage sampling
 Simple Random Sampling

Simple random sampling is a method of

probability sampling in which
every unit has an equal nonzero
chance of being selected.

When there are very large populations, it is often difficult or

impossible to identify every member of the population.
 Simple Random Sampling
Population
Sample

A T Y W
B P G E B T
S C K L G K
G N Q
element
population
➢ Probability selected = ni/N
➢ When population is rather uniform (e.g. school/college students, low-
cost houses)
➢ Simplest, fastest, cheapest Population not uniform

➢ Could be unreliable, why? Wrong procedure

 Random Selection

1. Pick any element randomly, Eg thru lottery concept

2. Use random table

 Systematic Random Sampling
Systematic random sampling is a
method of
probability sampling
in which the defined
target population is ordered
and the sample is selected
according to position using a skip interval

Systematic in the “picking-up” of element.

E.g. every 5th. visitor, every 10th. House, every 15th. minute
 Steps in Drawing a Systematic
Random Sampling
1. Compute the skip interval, k = N/n.
2. Determine a random start point between 1 and k
using lottery method
3. Beginning at the start point, select the units by
choosing each unit that corresponds to the skip
interval.
 Systematic Sampling (contd.)
Example:
1. There are 100 students in your class.
2. You want a sample of 20.
3. You have a class listing in alphabetical
order.
4. Divide 100 by 20, you will get 5.
5. Randomly select any number between 1
and five. That is your starting number.
6. From there select every 5th name until
you reach the last one.
 Stratified Random Sampling

Stratified random sampling is a

method of probability sampling
in which the population is divided
into different subgroups and samples
are selected from each.

1. Commonly used because it reduces sampling error.

2. A stratum is a subset of the population that share at least
one common characteristic.
3. Examples of stratums might be males and females, or managers and non-managers.
 Steps: Drawing a Stratified Random Sample

1. Divide the target population into homogeneous subgroups or

strata

2. Draw random samples fro each stratum (show the calculation)

3. Combine the samples from each stratum into a single sample of

the target population
 Stratified Random Sampling
Population Sample
Stratum 1 = odd no.
1 4 8 12 3 7
3 6 13 2 10 16
10 20 15
7 14 11 16 Stratum 2 = even no.

➢Break population into “meaningful” strata and take random sample from each stratum
➢When:
* key sub-groups need to be represented → more
precision
* Homogenous within elements
* Heterogenous between groups
* sub-group inferences are needed (Sample has to be selected from each group)
 Cluster Sampling
➢Research involves spatial issues
➢Sampling involves analysis of geographic units
➢Steps:
* Divide population into “clusters” (localities). Elements between
clusters are homogeneous but the subgroups are heterogenous
* Choose clusters randomly
* Take all cases from each cluster
Cluster Sampling
 Multi-Stage Sampling

Among choices:
* Two-stage cluster (cluster first, then,
stratify within cluster).

Tmn Bukit Tmn

Kingfisher Tmn Sulaman
Sepanggar

Cluster

Strata

M C I M C I M C I
 Multi-Stage Sampling

* Three-stage stratified (Locality first, then, ethnic, then, family status).

Inner Suburb Locality

Outskirt

C M C I M Ethnic
M C I I

Family
MD UD UD MD MD UD
status
 Non-Probability Sampling :
Convenience Sampling
➢Naïve sampling
➢Does not intend to represent the population
➢Selection based on one’s “convenience”, by “accident”, or “haphazard”
way
➢Common in popular surveys, public “view” or “opinion” (e.g. by-the-
road-side “interviews”)
➢Serious bias – only one group included
➢Must be avoided
 Quota Sampling

➢ Select cases non-randomly according to some fixed quota.

➢Proportional quota
* Represent major characteristics of the population by
proportion. E. g. 40% women and 60% men
* Have to decide the specific characteristics for the quota
(e.g. gender, age, education race, religion, etc.)
 Snowball Sampling

➢Identify a case that meets criteria for inclusion in the study.

➢Find another case, that also meets the criteria, based on the first one.

➢Next, search for others based on the previous ones, and so on.

➢Hardly leads to representative sample, but useful when population is

inaccessible or hard to find. E.g.
* the homeless
* wound-up companies
 Data Collection Methods

1. Personal/Face-to-face interview

2. Telephone interview
3. Mail/Postal interview
4. Direct observation
5. Online interview
 Methods of Data Collection : Advantages & Disadvantages
Method of data collection Advantages Disadvantages

Personal Interview More accurate High cost/expensive

Highest response rates Time inefficient
Visual aids can be used Training to avoid bias
Follow-up questions can be asked
Allows clarification
Telephone Interview Less expensive compared to personal interview Difficult to develop rapport
Higher response rates compared to mail interview No visual aids
Larger radius than personal Time is limited

Postal/Mail Interview Larger target Lowest response rates

Economical Not all subjects accessible
Survey fraud ( answer may not be honest)

Online Interview Ease of data collection (can be conducted fast) Questions might be left unanswered
Easily accessible Poor response
Low price Not reliable
Easy to analyze
Direct Observation Provides direct information Time consuming
More accurate May require training
Can be used in natural or experimental settings Potential for bias (overlook meaningful
aspects)
 EXERCISE 1
A group of researcher wanted to investigate the perception of the married couple towards
on the factors that contributed to problems in marriage in a particular state. The
questionnaires were distributed to 500 randomly selected couples from five districts in the
state. Among the variables selected in survey were age, gender, occupation, number of
years of marriage, number of children, the total household income and the perception on
the factors contributed to the marriage problems on a Likert scale where 1 indicates
strongly agree and 5 indicates strongly disagree. The distribution on the number of married
couples is shown in the following table.
District Number of Married Couples
A 320
B 370
C 280
D 300
E 230

I. State the population and sample of the above study

II. State TWO variables in the above study. Hence, identify the corresponding scales of
measurements
III. Calculate the number of married couples to be selected from each district if
stratified sampling is used. (or how to conduct stratified sampling)
 SOLUTION

1. Population : 1500 married couples, Sample : 500 married couples

2. Variable : Age (ratio), gender (nominal), occupation (nominal), no of years of marriage
(ratio), no of children (ratio), total household income (ratio), perception on the factors
contributed to marriage problems (ordinal)
3. District A : 500/1500 * 320 = 107
District B : 500/1500 * 370 = 123
District C : 500/1500 * 280 = 93
District D : 500/1500 * 300 = 100
District E : 500/1500 * 230 = 77
Total = 500
 EXERCISE 2

A survey on the workers’ satisfaction levels was carried out at Company

X. The company has 24 branches with the same setting. A sample of 6
branches was selected at random. All workers who work at these 6
branches were then selected for the study.

I. State the population of the study

II. State the sampling frame for the survey
III. State the variable, type of variable & level of measurement for this
study
IV. Name the sampling technique used in this study
V. What is the most suitable data collection method to be used in this
study.