Research

Methodology
Sampling

Week 8
 Learning Outcomes

Sampling methods:
Quantitative approach
1.The sampling process
2.Probability sampling
3.Non-probability sampling
4.Sample size

Saunders, M., Lewis. P & Thornhill, A. (2012) Research

methods for business students. 6th ed. Harlow: Pearson.
Module Code and Module Title Title of Slides Dr Jugindar Singh
 What is Sampling?
A sample is “a smaller (but
STUDY POPULATION
hopefully representative) Sampling Frame
collection of units from a
population used to determine
truths about that population”
(Field, 2005)
SAMPLE

The process of selecting a

number of individuals for a
study in such a way that the
individuals represent the larger
group from which they were
TARGET POPULATION
selected
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Barriers towards Entrepreneurship Inclination: A
quantitative study among APU students

• Population : All APU students

10,000

• Sampling Frame: A list of

elements from which the
sample may be drawn
1,000 SAMPLE

• Sample : Subset of a larger

population
100

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Illustration

Barriers towards Entrepreneurship Inclination: A quantitative study among APU

students

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Sampling Process
1. Defining target population.

2. Choosing sampling frame.

3. Determining sampling size

4. Selecting a sampling method

5. Applying the chosen

sampling method in practice.

6 SAMPLING BREAKDOWN
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Types of sampling in quantitative researches

Non-
Probability
samples probability
samples
Non-probability sampling techniques
Samples are selected based on the subjective judgment of the
researcher, rather than random selection

Probability sampling techniques

Probability sampling techniques is that units are selected from the
population at random using probabilistic methods.
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Classification of Sampling
Techniques
Sampling Techniques

Nonprobability Probability
Sampling Sampling Techniques
Techniques

Convenience Judgmental Quota Snowball

Sampling Sampling Sampling Sampling

Simple Systematic Stratified Cluster Multi stage

Random Sampling Sampling Sampling Sampling
Sampling
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Sampling Techniques

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Probability Sampling
• No biasness
• Best method to achieve a representative sample
• Five techniques
1. Random
2. Stratified random
3. Cluster
4. Systematic
5. Multistage

Simple Stratified
Systematic
Random Sampling
Sampling
Sampling

Cluster Multistage
Sampling area sampling
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Simple Random
Sampling
– Selected by using
Each member of the population has an equal

chance or random Define the

and independent chance of being chosen

numbers population

– Each individual List all

subject (human or members of
otherwise) has an population
equal chance of Assign
being selected numbers to
– Examples: each member
of population
• Drawing names
from a hat Use criterion
• Random Numbers to select
sample
Advantages: The sample should represent the target population
Module Code and Module Title
Eliminate sampling bias
Title of Slides Dr Jugindar Singh
 STRATIFIED RANDOM
SAMPLING
• Population is divided into two or more groups called
strata Eg: Age, Gender, geographic location
• Subsamples are randomly selected from each strata

The proportional representation of each class is determined

(e.g., 40% females & 60% males)

Advantage: The sample should be highly representative of the target

population and therefore we can generalize from the results obtained.
Each
Module subgroup
Code and Module Title within the population receives proper representation.
Title of Slides Dr Jugindar Singh
 CLUSTER SAMPLING
• The process of randomly selecting intact groups,
not individuals, within the defined population
sharing similar characteristics
• The population is divided into subgroups (clusters)
like families.
• A simple random sample is taken from each cluster
Examples
– Neighborhoods
– School districts
– Schools
– Classrooms
Module Code and Module Title Title of Slides Dr Jugindar Singh
 SYSTEMATIC RANDOM SAMPLING
A sampling procedure in which a starting point is selected by
a random process and then every nth number on the list is
selected.

• Order all units in the sampling frame

• Then every nth number on the list is selected
• N= Sampling Interval

Advantage: This method is that is should provide a representative

Eliminate biasness

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Advantages of Systematic sampling
1. Chosen data is evenly distributed. For example, in a population
of 10,000 people, a statistician might select every 100th person
for sampling

2. This method is that is should provide a representative.

Researchers generally assume the results are representative of
most normal populations

3. Eliminate biasness

4. Low risk factor. The primary potential disadvantages of the

system carry a distinctly low probability of contaminating the
data.

5. Easy to construct, execute, compare and understand

6. It also gives researchers a degree of control

Module Code and Module Title Title of Slides Dr Jugindar Singh

 SYSTEMATIC SAMPLING
1. Jane 18. Steve 35. Fred
2. Bill 19. Sam 36. Mike
3. Harriet 20. Marvin 37. Doug 1. Divide the population
4. Leni 21. Ed. T. 38. Ed M.
by the size of the
5. Micah 22. Jerry 39. Tom
6. Sara 23. Chitra 40. Mike G.
desired sample: e.g.,
7. Terri 24. Clenna 41. Nathan 50/10 = 5
8. Joan 25. Misty 42. Peggy 2. Select a starting point
9. Jim 26. Cindy 43. Heather
at random: e.g., 43 =
10. Terrill 27. Sy 44. Debbie
11. Susie 28. Phyllis 45. Cheryl
Heather
12. Nona 29. Jerry 46. Wes 3. Select every 5th name
13. Doug 30. Harry 47. Genna from the starting point
14. John S. 31. Dana 48. Ellie
15. Bruce A. 32. Bruce M. 49. Alex
16. Larry 33. Daphne 50. John D.
17. Bob 34. Phil

Module Code and Module Title Title of Slides

 Non Probability sampling
methods...
1. Convenience sampling
2. Purposive sampling/Judgemental Sampling
3. Quota sampling
3. Snowball sampling

Module Code and Module Title Title of Slides Dr Jugindar Singh

 CONVENIENCE SAMPLING
• Convenience sampling involves choosing respondents
at the convenience of the researcher.
• called “accidental” or “haphazard” sampling
Advantages
• Very low cost
• Extensively used/understood
Disadvantages
• Variability and bias cannot be measured or controlled
• Projecting data beyond sample not justified
• Restriction of Generalization.
The sampling procedure of obtaining those people or units
that are most conveniently available.
Module Code and Module Title Title of Slides Dr Jugindar Singh
 QUOTA SAMPLING
Sampling procedure that ensures that various subgroups
of a population will be represented on pertinent
characteristics to the exact extent that the investigator
desires.
Example: age, gender and socioeconomic status or social
class
To select a quota sample you:
1 Divide the population into specific groups. Eg: male and Female
2 Calculate a quota for each group based on relevant/available data.
3 Give each interviewer an ‘assignment’, which states the number of
cases in each quota from which they must collect data.
4 Combine the data collected by interviewers to provide the full
sample.
Module Code and Module Title Title of Slides Dr Jugindar Singh
 JUDGEMENTAL SAMPLING
Sampling technique in which an experienced individual
selects the sample based on personal judgment about some
appropriate characteristics of the sample member

Example: Select only CEOs in SMEs

• The process whereby the researcher selects a sample
based on experience or knowledge of the group to be
sampled
• Advantages
• There is a assurance of Quality response
• Meet the specific objective.
Disadvantages
• Bias selection of sample may occur
• Time consuming process.
Module Code and Module Title Title of Slides Dr Jugindar Singh
 SNOWBALL SAMPLING
The research starts with a key person and introduce the next
one to become a chain
–
Make contact with one or two cases in the population.
–
Ask these cases to identify further cases.
–
Ask these new cases to identify further new cases (and so on).
–
Stop when either no new cases are given or the sample is as large
as is manageable.
Advantages
• Low cost
• Useful in specific circumstances & for locating rare populations
Disadvantages
• Not independent
• Projecting data beyond sample not justified
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Sample Size: Krejcie & Morgan
This is the sub-population to be studied
in order to make an inference to a (1970)
reference population
(A broader population to which the findings
from a study are to be generalized)

Formula (Krejcie & Morgan, 1970) may be used to

determine the sample size.

S = X 2NP (1-P)/d2 (N-1) + X2P(1-P)

Where:
S = Required Sample size
X = Z value (e.g. 1.96 for 95% confidence level)
N = Population Size
P = Population proportion (expressed as
decimal) (assumed to be 0.5 (50%)
d 2 = Degree of accuracy (5%), expressed as a
2 proportion
Dr Jugindar Singh (.05); It is margin of error
 What governs the sample size
Sample size governed by:
• The confidence Level –the level of certainty that the characteristics of the
data will represent the characteristics of the total population (normally 95%)
• The margin of error that you can
tolerate the accuracy you require for any estimates made from your sample
.05%)
• The types of analyses to undertake
Eg: you wish to subdivide your data, as many statistical techniques have a
minimum threshold of data cases for each cell
• The size of the total population
from which your sample is being drawn.

95 per cent level of certainty means that if your Where the population in the category
sample was selected 100 times, at least 95 of these is less than 30, collect data from all
samples would be certain to rep. the characteristics cases in that category.
of the population

Dr Jugindar Singh
 Salant, P.,
& Dillman,
D. A.
(1994).
How to
conduct
your own
survey.
New York:
John Wiley
and Sons.
 Confidence Level
For example, if a random sample of 400 APU
students (out of 12,000 students) showed that
80% of them found that the food in university
café is expensive, researchers can be 95%
confident that between 75% and 85% of ALL
APU students would have the same comments,
even though they (12,000-400 = 11,600) are
NOT been interviewed.
 What is meant by split
A 50/50 split means that the population is completely
divided in their responses. The researcher would expect
50% of the population to answer one way (e.g., support,
agree) and 50% to answer the other way (e.g., oppose,
disagree). A 50/50 split is the most conservative value
possible.

An 80/20 split means that answers are less variable; many

people respond one way or have a certain characteristic,
whereas a few do not. If researchers have little or no
knowledge about the diversity of characteristics and
opinions among the population, the conservative 50/50
split approach is recommended (Salant & Dillman, 1994).
 Sample Size: Cohen Statistical Power Analysis (1998)
Five Factors considered
1. Significance level or criterion
2. effect size
3. desired power
4. estimated variance
5. sample size
The values of significance level,
effect size, power and estimated
variance have to be pre-determined. this statistical analysis.
• Significance for most studies : = .05.
• Effect sizes to test the significance of
product moment correlation
coefficient, r, are, .10, .30, and .50, for
small, medium and large
• Effect size index, f2 for small, medium
and large effect sizes are f2 = .02, .15,
and .35 respectively
• Statistical Power: Cohen (1992)
suggested fixing the power at .80 (β =
Dr Jugindar Singh
.20),
Correlational study
To find out the degree of the relationship
between a DV and 10 IVs
With a predetermined effect size
of r = .30 (medium), a significant alpha = .05
and a statistical power of .80
Desired sample size to test these
relationships is 85 (Cohen, 1992). This means
that 85 respondents are sufficient to perform
Regression Analysis
Calculation can performed on the maximum
of 10 independent variables (u = 10).
With the specified power of .80, a medium
effect size of f2 = .15, a significant alpha of
.05
Sample size is 116
 Tabachnick and Fidell (2013)
Tabachnick and Fidell (2013) proposed a formula
“50 + 8m” where “m” is the number of
variables. In this study, a total of 168 staff
completed and returned the questionnaires.
 FORMULAE FOR SAMPLE SIZE CALCULATION
There are many formulae for
calculating sample size
Determined largely by 3 factors
1. Population size: The total
number of people who qualify
for the study
2. The desired level of
confidence.
Usually 95%≡1.96 in business
research
3. The acceptable margin of error.
Desired in the study. ~1%-5%
or 0.01 and 0.05

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Sample Size for Regression Analysis
Multiple Regression Sample Size SEM Sample Size
For multiple regression
analyses the desired
level is between 15 to 20
observations for each
predictor variable

For example the predictor

variables are 5, require a
sample size of 100
Hair, J.F., W.C. Black, B.J. Babin,
R.E. Anderson and R.L. Tatham,
2010. Multivariate Data Analysis,
Prentice Hall, New Jersey
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Tutorial
1. Working in groups, read the attached
sample of research (or any other past
research) and explain the
methodology used.

• The study population

• The type of sampling
• Calculation of sample size

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Cohen: GPower

Power: 95%
Alpha at: 0.05
Medium effect of 0.15

Number of Predictors: 2

Module Code and Module Title Title of Slides Dr Jugindar Singh

 G-Power Example

The priori power analysis assumed a medium effect size

of .15 (f2 = .15) and a = .05 to achieve a power of .80 and
.95
Module Code and Module Title Title of Slides Dr Jugindar Singh
 Sample Questions

1. Explain 4 types of non probability

sampling
2. Explain 4 types of probability sampling

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Chapter 3

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Example

RELATIONSHIP AMONG Knowledge

Management, ORGANIZATIONAL
CULTURE AND Organization
PERFORMANCE

Module Code and Module Title Title of Slides Dr Jugindar Singh

3.9 Sampling (Type, population, frame, size)
1. State the Study Population, sampling frame and units of analysis. What was
the sample size and how was it calculated
•Describe the characteristics of your study population—who are they and where are
they situated and how big is this population size?
•Describe your sample size, how you derive the sample size (state the formula)
2. State the sampling method (probability or non-probability) and sampling
procedure.
• What was the sampling method (eg Convenience sampling)
• Reasons to support the sampling method
Example:
For all research questions where it would be impracticable for you to collect data from the entire
population, you need to select a sample (Saunders et al., 2012). For this study, this will be
important because of the use of questionnaires for data collection technique. Sampling also saves
time, an important consideration when you have tight deadlines and the organization of data
collection is more manageable as fewer people are involved. The sampling techniques available
to you can be divided into two types ie probability or representative sampling and non-
probability sampling. Non-probability sampling sampling was used as this was the most practical
(Saunders et al., 2012). The sample size was based on the formula by Morgan and Krejcie
(1970). Convenience sampling technique was used to select potential respondents in this survey.
Due to time constraints, the sample size for this study was 80. For this study, Convenience
sampling (or haphazard sampling) was used and this involves selecting haphazardly those cases
that are easiest to obtain for your sample (Saunders et al., 2012).
Module Code and Module Title Title of Slides Dr Jugindar Singh
Note:
For qualitative studies, the detail steps may be
different, but the overall expectation of a clear and
complete description of study population,
instrumentation, data collection and analysis is the
same as the quantitative studies.

Module Code and Module Title Title of Slides Dr Jugindar Singh

Methodology for Survey Research (adapted from
Creswell, 2009)
1. Introduction
• Chapter 3 begins with an introduction, where you briefly remind the reader of
the purpose of the study and the research questions/objectives/hypotheses.
This introduction can be used to describe the overall paradigm and rationale
for it (qualitative, quantitative, or mixed methods), as well as to describe the
overarching research purpose (exploratory, descriptive, or explanatory).

2. Design
• Introduce readers to the purpose of survey research and the rationale for its
selection for your proposed study.
• This section should indicate if the survey will be cross-sectional or
longitudinal, and should also indicate the specific form of data collection
(self-administered questionnaire; in-person or telephone interview;
structured observation; web-based self-administered questionnaire; etc.) A
rationale for the form specified should be included.

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Population and Sample
• Indicate the population (size, characteristics, etc.) and
accessible population and sampling frame if applicable.

• Describe the sampling design (single or multi-stage) and

selection process (random or non-random), and if stratification
will be used.

• Indicate the number of people in the sample and how that

number was calculated (see applicable research texts for
sampling methods and sample size formulas and tables).

Module Code and Module Title Title of Slides Dr Jugindar Singh

 Instrumentation
• Describe in detail the actual survey instrument (was it developed for this
study, modified, or an intact instrument?).
• Describe the major sections of the instrument, including the instructions,
items, and response scales.
• Consider creating a table of specifications to align research questions,
constructs, dimensions, and items. Consider including sample items
• Indicate if the instrument will be analyzed at the item level, subscales, etc.
(or include this in the Data Analysis section)
• Indicate how validity and reliability will be (or have been)
established/demonstrated.
• Describe plans for pilot or field testing a self-developed or modified
instrument. Indicate how results from this testing will translate into
instrument revisions.
• Describe steps for administering the survey and ensuring the desired
response rate (this could be in a separate section Data Collection
Procedures). See survey research texts for survey administration processes
designed to ensure a high response rate.

Module Code and Module Title Title of Slides Dr Jugindar Singh