Sampling
Techniques
DR. FAISAL KALOTA
1
Statistics
Before we discuss the details related to population, samples and
sampling techniques, lets go over some basic statistical concepts.
Statistics: “Is the science of conducting studies to collect, organize,
summarize, analyze, and draw conclusions from data” (Bluman,
2012, p. 3).
Descriptive Statistics: “Consists of the collection, organization,
summarization, and presentation of data” (Bluman, 2012, p. 4).
Infrential Statistics: “Consists of generalizing from samples to
populations, performing estimation and hypothesis tests,
determining relationship among variables, and making predictions”
(Bluman, 2012, p. 4)
58%23%
10%
9%
Data
1st Qtr 2nd Qtr 3rd Qtr 4th Qtr
2
Population
A Population Consists of all subjects (Humans or otherwise) that are being studied. (Bluman,
2012, p. 4)
A population may not always consists of humans.
Example:
Let’s say your research objective is to study British Restaurants in UAE. Then the population in
would be all the British Restaurants in UAE.
Let’s say your research objective is to all the Skyscrapers with more than 80 floors in the UAE.
Then the population would be ____________________________?
3
Sample
A Sample is a group of subjects from the population (Bluman, 2012, p. 4).
Generally, it is not easy study the whole population for various reasons such as lack of
resources, costly, size of population, etc.
Therefore, you select a Sample from the Population. The goal is to select a sample that
represents the population, so that the results be generalized to the Population.
4
Population vs. SamplePOPULATION
Sample
So how do you
select a Sample?
5
Sample Selection
Probability Sampling: “The chance, or probability, of each case being selected from the
population is known and is usually equal in all cases. This means it is possible to answer
research questions and to achieve objectives that require you to estimate statistically the
characteristics of the population from the sample” (Saunders et al., 2012, p. 261)
Non Probability Sample: “The probability of each case being selected from the total population
is not known and it is impossible to answer research questions or to address objectives that
require you to make statistical inference about the characteristics of the population” (Saunders
et al., 2012, p. 262).
6
Probability Sampling Stages
1. Identify the suitable sampling frame based on your research questions(s) and objectives.
2. Decide on suitable sample size.
3. Select the most appropriate sampling technique and select the sample.
4. Check that the sample is representative of the population. (Saunders et al, 2011, p. 262)
Sampling Frame for any probability sample is the complete list of all cases in the population
from which your sample is drawn (Saunders et al, 2012, p. 262)
7
Probability Sample Size
There is a difference of opinion among the scholars regarding the sample size. Additionally, it
also depends on the size of the population and the degree of confidence. Below are some
general guide lines.
According to Fraenkel & Wallen (2006, p 104):
For Descriptive Studies minimum sample size should be 100
For Correlational Studies minimum sample size should be 50
For Experimental & Causal-Comparative studies a minimum of 30 individuals per group.
Table 7.1 in Saunders et al (2012, p. 266) provides different sizes of population at 95% confidence
level (see next slide)
8
Sampling Techniques
Source:(Saundersetal,2012p.261,Figure7.2)
9
Probability Sample Size
10
Saunders et al. (2012, p. 266)
11
SelectingaProbabilitySample
Saunders et al. (2012, p. 271)
12
Saunders et al. (2012, p. 272)
ImpactofvariousfactorsonChoiceof
ProbabilitySamplingTechnique
Non-Probability Sample Size
There is a difference of opinion among the scholars regarding the non-probability sample size.
Additionally, it also depends on the size objective of the study. Below are some general
guidelines
13
Saundersetal.(2012,p.283)
14
SelectingaNon-Probability
Sample
Saunders et al. (2012, p. 282)
15
ImpactofvariousfactorsonChoiceof
Non-ProbabilitySamplingTechnique
Saunders et al. (2012, p. 284)
Sampling Techniques
Source:(Saundersetal,2012p.261,Figure7.2)
16
Simple Random Sampling
Simple Random Samples are selected by using a chance method or random numbers. (Bluman,
2012, p. 10)
You can assign a number to everyone in the population and select the numbers randomly using
different methods such as computer generated numbers, random number tables, etc.
A H O V
B I P W
C J Q X
D K R Y
E L S Z
F M T
G N U
H
B P W
J
D Y
S Z
F M
Population Sample
17
Systematic Sampling
Numbering each subject of the population, and then selecting every Kth subject (Bluman, 2012,
p. 11).
For example lets say there are 120 subjects and 10 subjects are needed. You number all of them
from 1-120. Then divide 120/10 = 12. So K will be 12. So you select the first participant at
random, and from that point onward you start selecting every 12th subject.
Care must be taken because lets say the subjects are wife and husband, and also numbered as
such. So you might end up only with all the wives or all the husbands.
18
Stratified Sampling
“Dividing the population into group (called Strata) according to some characteristic that is
important to the study, then sampling from each group” (Bluman, 2012, p. 12).
“… subgroups, or strata, are selected form the sample in the same proportion as they exist in the
population” (Fraenkel & Wallen, 2006, p. 96).
25% 50% 25%
A H O V
B I P W
C J Q X
D K R Y
E L S Z
F M T
G N U
25% 50% 25%
A H O V
I P
T
G N U
19
Cluster Sampling
“Population is divided into groups called Clusters by some means such a geographic location,
school, etc. Then the researcher randomly selects some of the clusters and uses all the
members of the clusters as the subjects of the sample” (Bluman, 2012, p. 12).
A, B, C, D E, F, G, H
L, M, N, O I, J, K
T, U, V
P, Q, R, S W, X, Y, Z
A, B, C, D
W, X, Y, Z
Population Cluster Random
I, J, K
20
Quota Sampling
“Quota Sampling is entirely non-random and is often used for structured interviews as part of a
survey strategy” (Saunders et al, 2012, p. 284).
It can be considered as a non-probabilistic version of Stratified Sampling.
To select a quota sample (Saunders et al, 2012, p. 285):
1. Divide the population into specific groups
2. Calculate a quota for each group based on relevant and available data.
3. Give each interviewer an ‘assignment’, which states the number of cases in each quota for
which they must collect data.
4. Combine the data collected by the interviewer to provide the full sample
21
Purposive
“Based on previous knowledge of a population and specific purpose of the research,
investigators use personal judgment to select a sample. Researchers assume they can use their
knowledge of the population to judge whether or not a particular sample will be representative”
(Fraenkel & Wallen, 2006, p. 100)
Example:
A teacher may select 2 High-GPA, 2 Medium-GPA, 2 Low-GPA students to find about the class’
opinion on a particular teaching strategy.
22
Volunteer (Self Selection & Snowball)
Self-Selection Sampling: “Non-probability sampling procedure in which the case, usually an
individual is allowed to identify their desire to be part of the sample” (Saunders et al, 2012, p.
681)
Snowball Sampling: “Non-probability sampling procedure in which subsequent respondents are
obtained from information provided by initial respondents ” (Saunders et al, 2012, p. 682)
23
Haphazard (Convenience)
In a Convenience Sample the researcher uses the subjects that are convenient (Bluman, 2012, p.
12). Example: Interview subjects at the entrance of a park.
This type of sample may not be a representative of the population. For example, when you are
interviewing people at the entrance of the park, the subjects may change based on the time,
date, day, weather, etc.
24
References
Bluman, A. (2012). Elementary statistics: A step by step approach (8th ed.) [International
edition]. New York, NY: Mc-Graw Hill.
Fraenkel, J. R., & Wallen, N. E. (2006). How to design and evaluate research in education (6th ed.).
New York, NY: Mc-Graw Hill
Saunders, M., Lewis, P., & Thornhill, A. (2012). Research methods for business students (6th ed.).
Harlow, England: Prentice Hall.
25

7. lo1 lo4 sampling

  • 1.
  • 2.
    Statistics Before we discussthe details related to population, samples and sampling techniques, lets go over some basic statistical concepts. Statistics: “Is the science of conducting studies to collect, organize, summarize, analyze, and draw conclusions from data” (Bluman, 2012, p. 3). Descriptive Statistics: “Consists of the collection, organization, summarization, and presentation of data” (Bluman, 2012, p. 4). Infrential Statistics: “Consists of generalizing from samples to populations, performing estimation and hypothesis tests, determining relationship among variables, and making predictions” (Bluman, 2012, p. 4) 58%23% 10% 9% Data 1st Qtr 2nd Qtr 3rd Qtr 4th Qtr 2
  • 3.
    Population A Population Consistsof all subjects (Humans or otherwise) that are being studied. (Bluman, 2012, p. 4) A population may not always consists of humans. Example: Let’s say your research objective is to study British Restaurants in UAE. Then the population in would be all the British Restaurants in UAE. Let’s say your research objective is to all the Skyscrapers with more than 80 floors in the UAE. Then the population would be ____________________________? 3
  • 4.
    Sample A Sample isa group of subjects from the population (Bluman, 2012, p. 4). Generally, it is not easy study the whole population for various reasons such as lack of resources, costly, size of population, etc. Therefore, you select a Sample from the Population. The goal is to select a sample that represents the population, so that the results be generalized to the Population. 4
  • 5.
    Population vs. SamplePOPULATION Sample Sohow do you select a Sample? 5
  • 6.
    Sample Selection Probability Sampling:“The chance, or probability, of each case being selected from the population is known and is usually equal in all cases. This means it is possible to answer research questions and to achieve objectives that require you to estimate statistically the characteristics of the population from the sample” (Saunders et al., 2012, p. 261) Non Probability Sample: “The probability of each case being selected from the total population is not known and it is impossible to answer research questions or to address objectives that require you to make statistical inference about the characteristics of the population” (Saunders et al., 2012, p. 262). 6
  • 7.
    Probability Sampling Stages 1.Identify the suitable sampling frame based on your research questions(s) and objectives. 2. Decide on suitable sample size. 3. Select the most appropriate sampling technique and select the sample. 4. Check that the sample is representative of the population. (Saunders et al, 2011, p. 262) Sampling Frame for any probability sample is the complete list of all cases in the population from which your sample is drawn (Saunders et al, 2012, p. 262) 7
  • 8.
    Probability Sample Size Thereis a difference of opinion among the scholars regarding the sample size. Additionally, it also depends on the size of the population and the degree of confidence. Below are some general guide lines. According to Fraenkel & Wallen (2006, p 104): For Descriptive Studies minimum sample size should be 100 For Correlational Studies minimum sample size should be 50 For Experimental & Causal-Comparative studies a minimum of 30 individuals per group. Table 7.1 in Saunders et al (2012, p. 266) provides different sizes of population at 95% confidence level (see next slide) 8
  • 9.
  • 10.
  • 11.
  • 12.
    12 Saunders et al.(2012, p. 272) ImpactofvariousfactorsonChoiceof ProbabilitySamplingTechnique
  • 13.
    Non-Probability Sample Size Thereis a difference of opinion among the scholars regarding the non-probability sample size. Additionally, it also depends on the size objective of the study. Below are some general guidelines 13 Saundersetal.(2012,p.283)
  • 14.
  • 15.
  • 16.
  • 17.
    Simple Random Sampling SimpleRandom Samples are selected by using a chance method or random numbers. (Bluman, 2012, p. 10) You can assign a number to everyone in the population and select the numbers randomly using different methods such as computer generated numbers, random number tables, etc. A H O V B I P W C J Q X D K R Y E L S Z F M T G N U H B P W J D Y S Z F M Population Sample 17
  • 18.
    Systematic Sampling Numbering eachsubject of the population, and then selecting every Kth subject (Bluman, 2012, p. 11). For example lets say there are 120 subjects and 10 subjects are needed. You number all of them from 1-120. Then divide 120/10 = 12. So K will be 12. So you select the first participant at random, and from that point onward you start selecting every 12th subject. Care must be taken because lets say the subjects are wife and husband, and also numbered as such. So you might end up only with all the wives or all the husbands. 18
  • 19.
    Stratified Sampling “Dividing thepopulation into group (called Strata) according to some characteristic that is important to the study, then sampling from each group” (Bluman, 2012, p. 12). “… subgroups, or strata, are selected form the sample in the same proportion as they exist in the population” (Fraenkel & Wallen, 2006, p. 96). 25% 50% 25% A H O V B I P W C J Q X D K R Y E L S Z F M T G N U 25% 50% 25% A H O V I P T G N U 19
  • 20.
    Cluster Sampling “Population isdivided into groups called Clusters by some means such a geographic location, school, etc. Then the researcher randomly selects some of the clusters and uses all the members of the clusters as the subjects of the sample” (Bluman, 2012, p. 12). A, B, C, D E, F, G, H L, M, N, O I, J, K T, U, V P, Q, R, S W, X, Y, Z A, B, C, D W, X, Y, Z Population Cluster Random I, J, K 20
  • 21.
    Quota Sampling “Quota Samplingis entirely non-random and is often used for structured interviews as part of a survey strategy” (Saunders et al, 2012, p. 284). It can be considered as a non-probabilistic version of Stratified Sampling. To select a quota sample (Saunders et al, 2012, p. 285): 1. Divide the population into specific groups 2. Calculate a quota for each group based on relevant and available data. 3. Give each interviewer an ‘assignment’, which states the number of cases in each quota for which they must collect data. 4. Combine the data collected by the interviewer to provide the full sample 21
  • 22.
    Purposive “Based on previousknowledge of a population and specific purpose of the research, investigators use personal judgment to select a sample. Researchers assume they can use their knowledge of the population to judge whether or not a particular sample will be representative” (Fraenkel & Wallen, 2006, p. 100) Example: A teacher may select 2 High-GPA, 2 Medium-GPA, 2 Low-GPA students to find about the class’ opinion on a particular teaching strategy. 22
  • 23.
    Volunteer (Self Selection& Snowball) Self-Selection Sampling: “Non-probability sampling procedure in which the case, usually an individual is allowed to identify their desire to be part of the sample” (Saunders et al, 2012, p. 681) Snowball Sampling: “Non-probability sampling procedure in which subsequent respondents are obtained from information provided by initial respondents ” (Saunders et al, 2012, p. 682) 23
  • 24.
    Haphazard (Convenience) In aConvenience Sample the researcher uses the subjects that are convenient (Bluman, 2012, p. 12). Example: Interview subjects at the entrance of a park. This type of sample may not be a representative of the population. For example, when you are interviewing people at the entrance of the park, the subjects may change based on the time, date, day, weather, etc. 24
  • 25.
    References Bluman, A. (2012).Elementary statistics: A step by step approach (8th ed.) [International edition]. New York, NY: Mc-Graw Hill. Fraenkel, J. R., & Wallen, N. E. (2006). How to design and evaluate research in education (6th ed.). New York, NY: Mc-Graw Hill Saunders, M., Lewis, P., & Thornhill, A. (2012). Research methods for business students (6th ed.). Harlow, England: Prentice Hall. 25