Sample Selection
Sanju Rusara Seneviratne MBPsS
Overview of Sample Selection
• Population vs. Sample
• Sampling Techniques
 Simple Random Sampling
 Stratified Random Sampling
 Convenience Sampling
 Quota Sampling
Why is sampling important in research?
• The way we select participants will determine the
population to which we can generalize our findings.
• The procedure used to assign participants to different
treatment conditions will determine whether bias exists in
our treatment groups.
• If we do a poor job at the sampling stage of the research
process, the integrity of the entire project is at risk.
Important Definitions
• Population: all members that meet a set of specifications
or a specified criterion.
 Undergraduate students at your university.
• Element: a single member of any given population.
 You.
• Sample: a selection of some elements (members) of a
population.
 This Class.
• Census: all elements (members) of a population.
 All undergraduate students at your university.
Probability vs. Non-Probability Sampling
There are two major sampling techniques:
1. Probability sampling – a researcher can specify the
probability of an element/member being included in the
sample.
2. Non-probability sampling – no way of estimating the
probability of an element/member being included in the
sample.
Given that (typically) a researcher’s interest in generalizing
findings derived from a sample to the general population, using
probability sampling is far more useful and precise. The caveat
here is that it is much more difficult and expensive to do this.
Simple Random Sampling
• Probability sampling is also called random sampling or
representative sampling.
• Random describes the procedure used to select elements from
a population.
• Each element in the population has an equal chance of being
selected (simple random sampling) or has a known probability
of being selected (stratified random sampling).
Steps in Simple Random Sampling
• Step 1. Defining the Population
 Define the population to which the findings will be generalized,
e.g. undergraduate students at CIRP.
• Step 2. Constructing a List.
 Procure a complete list of the population from which to select.
(Exhaustive list – all members of the population appear on it)
• Step 3. Drawing the Sample.
 Randomly drawing members from a sample, e.g. flipping coins,
drawing names out of a rotating drum or random number
generation.
• Step 4. Contacting Members of a Sample.
 We must contact each of those selected for our sample and
obtain the information needed. If we fail to contact all individuals
in the sample, the representativeness is lost.
Stratified Random Sampling
• To stratify means to classify or to separate people into groups
according to some characteristics, such as position, rank,
income, education, sex or ethnic background.
• These separate groupings are referred to as subsets or
subgroups.
• For a stratified random sample, the population is divided into
groups or strata, e.g. females/males, age, religion, SES.
• A random sample is selected from each stratum based upon
the percentage that each subgroup represents in the
population.
Stratified Random Sampling
• They require more effort and there is a practical limit to
the number of strata used.
• Because participants are to be chosen randomly from each
stratum, a complete list of the population within each
stratum must be constructed.
• There are two uses to this form of sampling:
 Primary interest is in the representativeness of the sample for purposes of
commenting on the population.
 The focus of interest is comparison between and among the strata.
Stratified Random Sampling
• Stratified samples are used to optimize group
comparisons.
• In this case, we are not concerned with representing the
entire population.
• Although random sampling is optimal from a
methodological point of view, it is not always possible
from a practical point of view.
Convenience Sampling
• This is a quick, inexpensive and convenient method!
• Researchers simply use participants who are available at
the moment.
• Convenience samples are non-probability samples.
• Therefore, it is not possible to specify the probability of
any population element’s being selected for the sample.
Quota Sampling
• It is not always possible or desirable to list all members of the
population and randomly select elements from that list.
• We first decide which subgroups of the population interest us.
 Localities are selected and interviewers are assigned a starting point, a
specified direction, and a goal of trying to meet quotas for subsets
(ethnic origins, political affiliations, and so on) selected from the
population.
 This, in turn, is dictated by the nature of the problem being investigated
(the question being asked). For issues of national interest (such as
abortion, drug use, or political preference), frequently used subsets are
age, race, sex, socioeconomic level, and religion.
 The intent is to select a sample whose frequency distribution of
characteristics reflects that of the population of interest.
• Within each subset, participants are not chosen randomly.
 There are no ready-made lists from which the researcher can select
randomly. Often individuals are selected based on availability.
Quota Sampling
• Often interviewers, for sampling purposes, concentrate on
areas where large numbers of people are likely to be.
• This could bias the findings.
• For example:
 samples taken in airports may overrepresent high-income
groups
 those at a bus or rail depots may overrepresent low-income
groups
 samples at either place may underrepresent those who
seldom travel.
Summary of Types of Sampling
Type of Sampling Advantages Limitations
Simple Random - Representative of the
population
- May be difficult to
obtain the list
- May be more
Stratified Random
Sampling
- Representative of the
population
- May be difficult to
obtain the list
- May be more
Convenience Sampling - Simple
- Easy
- Convenient
- No complete member
list required
- May not be
representative of the
population
Quota Sampling - Simple
- Easy
- Convenient
- No complete member
list required
- May not be
representative of the
population
Activity – Reflection
Go to an Internet site that has an online poll. (If you are not
familiar with such a site, simply conduct a search using the
keywords “online poll” or “online survey” and select one of
the sites.) Participate in the survey, and view the current
results. Describe the nature of the survey and the current
results.
• What method of sampling was used?
• What characteristics do you believe the members of the
sample possess?
• Do you believe that these characteristics influence the poll
results?
• Would the results be different if the general population of
U.S. adults were polled?

Sample Selection

  • 1.
  • 2.
    Overview of SampleSelection • Population vs. Sample • Sampling Techniques  Simple Random Sampling  Stratified Random Sampling  Convenience Sampling  Quota Sampling
  • 3.
    Why is samplingimportant in research? • The way we select participants will determine the population to which we can generalize our findings. • The procedure used to assign participants to different treatment conditions will determine whether bias exists in our treatment groups. • If we do a poor job at the sampling stage of the research process, the integrity of the entire project is at risk.
  • 4.
    Important Definitions • Population:all members that meet a set of specifications or a specified criterion.  Undergraduate students at your university. • Element: a single member of any given population.  You. • Sample: a selection of some elements (members) of a population.  This Class. • Census: all elements (members) of a population.  All undergraduate students at your university.
  • 5.
    Probability vs. Non-ProbabilitySampling There are two major sampling techniques: 1. Probability sampling – a researcher can specify the probability of an element/member being included in the sample. 2. Non-probability sampling – no way of estimating the probability of an element/member being included in the sample. Given that (typically) a researcher’s interest in generalizing findings derived from a sample to the general population, using probability sampling is far more useful and precise. The caveat here is that it is much more difficult and expensive to do this.
  • 6.
    Simple Random Sampling •Probability sampling is also called random sampling or representative sampling. • Random describes the procedure used to select elements from a population. • Each element in the population has an equal chance of being selected (simple random sampling) or has a known probability of being selected (stratified random sampling).
  • 7.
    Steps in SimpleRandom Sampling • Step 1. Defining the Population  Define the population to which the findings will be generalized, e.g. undergraduate students at CIRP. • Step 2. Constructing a List.  Procure a complete list of the population from which to select. (Exhaustive list – all members of the population appear on it) • Step 3. Drawing the Sample.  Randomly drawing members from a sample, e.g. flipping coins, drawing names out of a rotating drum or random number generation. • Step 4. Contacting Members of a Sample.  We must contact each of those selected for our sample and obtain the information needed. If we fail to contact all individuals in the sample, the representativeness is lost.
  • 8.
    Stratified Random Sampling •To stratify means to classify or to separate people into groups according to some characteristics, such as position, rank, income, education, sex or ethnic background. • These separate groupings are referred to as subsets or subgroups. • For a stratified random sample, the population is divided into groups or strata, e.g. females/males, age, religion, SES. • A random sample is selected from each stratum based upon the percentage that each subgroup represents in the population.
  • 9.
    Stratified Random Sampling •They require more effort and there is a practical limit to the number of strata used. • Because participants are to be chosen randomly from each stratum, a complete list of the population within each stratum must be constructed. • There are two uses to this form of sampling:  Primary interest is in the representativeness of the sample for purposes of commenting on the population.  The focus of interest is comparison between and among the strata.
  • 10.
    Stratified Random Sampling •Stratified samples are used to optimize group comparisons. • In this case, we are not concerned with representing the entire population. • Although random sampling is optimal from a methodological point of view, it is not always possible from a practical point of view.
  • 11.
    Convenience Sampling • Thisis a quick, inexpensive and convenient method! • Researchers simply use participants who are available at the moment. • Convenience samples are non-probability samples. • Therefore, it is not possible to specify the probability of any population element’s being selected for the sample.
  • 12.
    Quota Sampling • Itis not always possible or desirable to list all members of the population and randomly select elements from that list. • We first decide which subgroups of the population interest us.  Localities are selected and interviewers are assigned a starting point, a specified direction, and a goal of trying to meet quotas for subsets (ethnic origins, political affiliations, and so on) selected from the population.  This, in turn, is dictated by the nature of the problem being investigated (the question being asked). For issues of national interest (such as abortion, drug use, or political preference), frequently used subsets are age, race, sex, socioeconomic level, and religion.  The intent is to select a sample whose frequency distribution of characteristics reflects that of the population of interest. • Within each subset, participants are not chosen randomly.  There are no ready-made lists from which the researcher can select randomly. Often individuals are selected based on availability.
  • 13.
    Quota Sampling • Ofteninterviewers, for sampling purposes, concentrate on areas where large numbers of people are likely to be. • This could bias the findings. • For example:  samples taken in airports may overrepresent high-income groups  those at a bus or rail depots may overrepresent low-income groups  samples at either place may underrepresent those who seldom travel.
  • 14.
    Summary of Typesof Sampling Type of Sampling Advantages Limitations Simple Random - Representative of the population - May be difficult to obtain the list - May be more Stratified Random Sampling - Representative of the population - May be difficult to obtain the list - May be more Convenience Sampling - Simple - Easy - Convenient - No complete member list required - May not be representative of the population Quota Sampling - Simple - Easy - Convenient - No complete member list required - May not be representative of the population
  • 15.
    Activity – Reflection Goto an Internet site that has an online poll. (If you are not familiar with such a site, simply conduct a search using the keywords “online poll” or “online survey” and select one of the sites.) Participate in the survey, and view the current results. Describe the nature of the survey and the current results. • What method of sampling was used? • What characteristics do you believe the members of the sample possess? • Do you believe that these characteristics influence the poll results? • Would the results be different if the general population of U.S. adults were polled?