3. By the end of the session you will be able to:
• Explain what sampling means in research
• List the different sampling methods available
• Choosing a sampling method
• Survey Bias
5. Population
• The population is the entire group of all entities about which we want
information, e.g.
• All children below 16 years of age ~ to assess their nutritional levels
• All farmers growing maize ~ to learn about total maize production in the
country
• All schools in a country ~ to learn about educational achievements
• Suppose you want to estimate the proportion of adults in rural areas
with no primary education. What do you think is your population?
6. The Specific Population types are;
• Target population is a group of individuals who meets the criteria.
• Subject or respondent population refers to a group of individuals
participating in the study.
• Strata or stratum is described as a mutually exclusive segment of a
population established by one or more characteristics.
7. Samples
• A sample is a subset of units drawn from the population.
• Sampling units refer to the entities on which measurements are
made during a survey.
• Sampling unit refers to specific entities, places or locations which can
be used during sampling process
• For example a village, household, farm, school, etc
8. Sampling frames
• The sampling frame is a list of all sampling units, for example
• list of villages in a region
• list of households within a village
• list of schools in the district
• Different frames are needed if sampling at different hierarchical levels
9. Representativeness
• The primary aim in a survey is to make inferences about the target
population. For this, need to ensure population is well represented in
the sample.
• Representativeness – means that the sample must be like the
population in as many ways as possible.
• Bring in qualitative aspects of the population to ensure there is adequate
representation of divisions of the population, e.g. rural/urban, different
wealth categories, geographical coverage, etc.
• May be achieved by sampling from these different sub-groups so that all
necessary factors likely to influence survey results are represented
10. Generalisability
• key to survey success is to ensure the sampling is such that survey results
can be generalised to the study population.
• Generalisability requires taking probability-based samples during the
sampling process.
11. Issues to consider when sampling
• Have the objectives been clearly specified?
• Has the target population been clearly defined and the possibility of survey
results being applicable to a different ‘study’ population been recognised?
• What is the geographical coverage?
• What factors are likely to influence survey results – have they been
considered in the sampling?
• What should be the sampling unit(s) for fieldwork?
12. Issues to consider when sampling
• Will the sample results lead to generalisable conclusions?
• Will the proposed sampling plan be possible within time and budget
limitations?
• Is the sampling procedure practically feasible?
• Will the adopted sampling scheme provide results that address
survey objectives with appropriate measures of precision?
14. Types of Sampling Methods
Sampling methods can be split into two distinct groups:
1. Probability samples
2. Non-probability samples
15. Probability Samples
• Probability samples offer each respondent an equal probability or
chance at being included in the sample.
• They are considered to be:
• Objective
• Empirical
• Scientific
• Quantitative
• Representative
The idea behind this type is random selection.
16. Non Probability Samples
• A non probability sample relies on the researcher selecting the
respondents.
• They are considered to be:
• Subjective
• Not scientific
• Qualitative
• Unrepresentative
17. Probability Samples Vs Non-Probability Samples
Probability Samples
• Generalize to the entire
population
• Unbiased results
• Known, non-zero probability of
Selection
Non Probability Samples
• Exploratory Research
• Convenience
• Probability of selection in
unknown
18. Probability Sampling Techniques
1. Random Sampling
2. Systematic Random Sampling
3. Stratified Random Sampling
4. Cluster Random Sampling
5. Quota Random Sampling
6. Multi-Stage Sampling
19. 1. Random Sampling
• Random means that each person within the sample frame has an
equal chance of being selected.
• In order to be random, a full list of everyone within a sample frame is
required.
• Random number tables or a computer is then used to select
respondents at random from the list.
21. 2. Systematic Random Sampling
This selection is like random sampling but rather than use random tables
or a computer to select your respondents you select them in a systematic
way. For instance, every 3rd person on the college list is selected.
22. 3. Stratified Random Sampling
• In this form of sampling, the population is first divided or organized
into two or more mutually exclusive homogenous subsets before
sampling.
• These subsets, called strata, are non-overlapping and together they
comprise the whole of the population.
• Sample is drawn random within each subsets.
23. The primary benefit of this method is to ensure that cases from smaller
strata of the population are included in sufficient numbers to allow
comparison
24. 4. Cluster Random Sampling
• In some instances the primary sampling unit is not the individual
element, but a large cluster of elements.
• Similar to stratified sampling but the groups/subsets are selected for
their geographical location.
• Constructing a complete list of population elements is difficult, costly, or
impossible.
• The population is concentrated in "natural" clusters (city blocks, schools,
hospitals, etc.
• Elements with each strata are Homogenous but are Heterogeneous
across strata.
25.
26. The Difference Between Strata and Clusters
• Although strata and clusters are both non-overlapping subsets of the
population, they differ in several ways.
• With stratified sampling, the best survey results occur when elements within
strata are internally homogeneous.
• With cluster sampling, the best results occur when elements within clusters
are internally heterogeneous.
27. 5. Quota Random Sampling
• The sampling procedure that ensures that a certain
characteristic of a population sample will be represented
to the exact extent that the investigator desires
• Rather than taking just anyone, you set quotas to ensure that the
sample you get represents certain characteristics in proportion to their
prevalence in the population.
• Need to know something about the characteristics of the population
ahead of time
30. Non-probability Sampling
• Convenience Sampling
• Snowball Sampling
• Purposive Sample
• These non-probability methods can be used in conjuncture with the
cluster, quota or stratified methods, however they will remain non-
probability samples
31. 1. Convenience Sampling
• This involves selecting the nearest and most convenient people to
participate in the research.
• This method of selection is not representative and is considered a
very unsatisfactory way to conduct research.
32. 2. Snowball Sampling
• The sampling procedure in which the initial respondents are chosen
by probability or non-probability methods, and then additional
respondents are obtained by information provided by the initial
respondents (such as being members of a gang).
• Snowball Sampling is choosing the participants to find more
participants for making a sample group.
33.
34. 3. Purposive Sampling
• Also known as judgmental, selective or subjective sampling.
• When members of the sample are purposively selected because they
possess certain traits that are critical to the study
• Limited generalizability
35.
36. Choosing a sampling method
Method Best when
Simple random sampling Whole population is available.
Stratified sampling
There are specific sub-groups to investigate (eg.
demographic groupings).
Systematic sampling
When a stream of representative people are available
(eg. in the street).
Cluster sampling
When population groups are separated and access to all
is difficult, eg. in many distant cities.
37. Choosing a sampling method
Method Best when
Snowball sampling
You are ethically and socially able to ask and seek similar
subjects. (ask for recommendations)
Convenience sampling You cannot proactively seek out subjects.
Judgment sampling You are expert and there is no other choice.
Quota sampling You have access to a wide population, including sub-groups
38. Sample Size
• When conducting probability sampling it is important to use a sample
size that is appropriate to the aims and objectives of the research.
• A common error is to assume that the sample should be a certain
percentage of the population, for example 10%. In reality there is no
such relationship and it only the size of the sample that is important.
• A probability sample size of 100+ is considered a large enough sample to
conduct statistical analysis
39.
40.
41. Result from survey is never exactly the same
as the actual value in the population
Survey Error and Survey Bias
42. Survey Error and Survey Bias
• In survey research, error can be defined as
any difference between the average values that were obtained
through a study and the true average values of the population being
targeted. For example: Age and Height of student in the class etc.
43. Survey Error and Survey Bias
• bias refers only to error that is systematic in nature.
• For example, including a question like “Do you drive recklessly?” in a public
safety survey would create systematic error and therefore be bias.
44. 1. Non-sampling bias
• Is present even if sampling and analysis done correctly
• Would still be present if survey measured outcome in ENTIRE
sampling frame
you have either sampled the wrong people or screwed up your
measurements!
45. 1. Non-sampling bias
Source of bias
Sampling frame out of date
Non-response
Measurement error
• Use current sampling frame
• Limit generalizations
• Minimize non-response
• Use various statistical methods to weight data
• Standardize instruments
• Write clear & simple questions
• Train and supervise survey workers
Prevention or cure
46. 2. Sampling bias
• Selection of non-representative sample, i.e., the likelihood of
selection not equal for each sampling unit
• Failure to weight analysis of unequal probability sample
you have not sampled people with equal probability
and you have not accounted for this in your analysis!
47. Sampling error
• Difference between survey result and population value due to
random selection of sample
• Influenced by:
• Sample size
• Sampling scheme
Unlike non-sampling bias and sampling bias, it can be
predicted, calculated, and accounted for.
48. Sampling error
• Measures of sampling error:
• Confidence limits
• Standard error
• Coefficient of variance
• P values
• Others
• Use these measures to:
• Calculate sample size prior to sampling
• Determine how sure we are of result after analysis
49. External Validity
External validity (sampling for representation) is the condition permitting
the generalization or inference of the sample findings to the population
from which the sample was selected.