SADC Course in Statistics Basic Sampling Concepts (Session 02)
-
Upload
sarah-stone -
Category
Documents
-
view
225 -
download
1
Transcript of SADC Course in Statistics Basic Sampling Concepts (Session 02)
SADC Course in Statistics
Basic Sampling Concepts
(Session 02)
2To put your footer here go to View > Header and Footer
Learning ObjectivesBy the end of this session, you will be able to • describe what is meant by sample, target
population, sampled (study) population, sampling frame, sampling units
• explain what is meant by a representative sample
• discuss the importance of sampling to ensure survey results are generalisable – use of probability based samples
• describe key issues central to the sampling process
3To put your footer here go to View > Header and Footer
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?
4To put your footer here go to View > Header and Footer
More on Populations• Population sizes may be finite or infinite.
We will allow the possibility of finite populations in this module.
• Describing the population clearly is necessary before sampling is considered. Why do you think this is so?
• Need also to distinguish between the target population, as defined above, and the study (or sampled) population. Latter excludes the inaccessible members of the target group e.g. temporary international migrant workers.
5To put your footer here go to View > Header and Footer
Samples• A sample is a subset of units drawn from
the population. This module is largely on sampling procedures.
• Sampling units refer to the entities on which measurements are made during a survey
• During planning (and implementation) of a survey, check the extent to which the target population can be (or has been) covered. Recognise that survey conclusions apply only to the population (study popn) which has a chance of inclusion in the sample.
6To put your footer here go to View > Header and Footer
Issues related to sampling unitsPart of the planning of the sampling processincludes:
• Identifying the sampling unit. Is it– a village, household, farm, school, etc?
• Recognising occurrence of units at different levels.– In most surveys, sampling is done at different
levels, e.g. district, then village, then household, then household member. Need to be specific about the selection of sampling units and sampling effort at each stage and specify this in the sampling protocol.
7To put your footer here go to View > Header and Footer
More on sampling units• Observing units over time
– In longitudinal surveys, or monitoring and evaluation surveys, should the same units be chosen for measurement each year?
• Unequal sized units– Are all units, whose measurements are being
summarised, of the same size? If not, e.g. with business surveys, should measurements be weighted according to size?
• Number of units– Is the sample large enough to enable an
analysis that addresses the study objectives?
8To put your footer here go to View > Header and Footer
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
• Why do we need a sampling frame?– Tool essential for objectively selecting a sample of
units from the population of all units
• Different frames are needed if sampling at different hierarchical levels– May develop relevant sub-frames at each stage
as sampling progresses down the hierarchy
9To put your footer here go to View > Header and Footer
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– 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
– In doing so, ensure selection is such that variability of sampling units within each sub-division is captured, i.e. not just 1 unit per group.
10To put your footer here go to View > Header and Footer
Non-representativeness?• Example: in a survey of Kenya, urban Nairobi is
represented by a random sample of size 10, which, by chance, includes eight women, five being office workers and four of those employed in Ministry offices. Both the two men, by chance, are unemployed.
• Example: a random sample of two areas in Malawi happened to come up with urban Mzuzu and urban Zomba [2 of the 6 largest towns]
• In both cases, these (honest) random samples were of very small size, and by chance produced quite untypical results.
11To put your footer here go to View > Header and Footer
Generalisability• Also 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.
• Probability sampling is the general term for methods where sample selection is objectively-based on known chances of inclusion in the sample.– If the probabilities are known and non-zero, they
don’t have to be equal: corrections can be made at the data analysis stage.
12To put your footer here go to View > Header and Footer
Non-generalisability• Example – a researcher selected her sample from a
district, including one farm each with all possible combinations of (i) child-/female-/male-headed; (ii) Christian/Muslim; (iii) dambo/hillside land; (iv) purely vegetable-/mixed/purely maize-growing
• Some cases very hard to find e.g. there were few Muslim, child-headed, purely maize dambo farms.
• Sample of 36 does not relate to population proportions: not enough cases to be able to represent the commoner combinations
• Researcher’s selections not objective/random and cases only came from one smallish area, maybe untypical. Cannot generalise to country!
13To put your footer here go to View > Header and Footer
Summary of 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?
14To put your footer here go to View > Header and Footer
Summary of issues to consider when sampling - continued• 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?
Developing a good sampling plan requirestime and effort! Full plan should be welldocumented with full justification!
15To put your footer here go to View > Header and Footer
ReferencesPettersson H. (2004) “Design of master sampling frames and master samples for household surveys in developing countries”. Chapter V of the UN Publication An Analysis of Operating Characteristics of Household Surveys in Developing and Transition Countries: Survey Costs, Design Effects and Non-Sampling Errors. Pp.71-94. Available at http://unstats.un.org/unsd/hhsurveys/index.htm
De Vaus, A.D. (2001) “Research Design in Social Research”. Sage Publications, London - for discussion on generalisability
Bechhofer, F. & Paterson, L. (2000) “Principles of research design in the social sciences”. Routledge, London – for a discussion on representativeness
16To put your footer here go to View > Header and Footer
Some practical work follows …