Chapter 6: Sampling
Applied Social ResearchA Tool for the Human Services
Terms in this set (14)
Sampling is the process by which a researcher selects one or more cases out of some larger grouping for study.
Why Researchers Use Sampling?
Feasibility: the whole group is sometimes too large to study everyone.
Data quality: information based on carefully drawn samples can actually be better than information from an entire group.
A sample consists of one or more elements or cases selected from some larger grouping or population.
A representative sample accurately reflects the distribution of
in the target population.
A population refers to all possible cases of what we are interested in studying.
Key features of a population and examples:
Content: Survivors of childhood sexual abuse
Units: Michigan families with developmentally disabled children
Extent: Combat veterans
Time: Agency clients treated for addiction 2000-2005
Survivors of childhood sexual abuse
A sampling frame is a
of all the elements in a population.
Does the sampling frame include all members of the population? (Yes)
Common sampling frames:
Telephone directories - Coverage, Unlisted numbers, Random digit dialing (rdd)
Utility subscribers - Multiple family dwellings
City directories - Includes information on occupation
Probability Theory and Sampling Distributions
Probability theory focuses on determining the likelihood or probability that certain events will occur.
A sampling distribution is a distribution of sample statistics. *Bell curve
When we draw a random sample, the most likely outcome is a representative sample or one that is very close to representative.
Probability sample: each element in the population has some chance of inclusion.
Each element's probability of inclusion is:
Sampling error: an estimate of the extent to which the values of the sample differ from those of the population.
Simple random sampling (SRS): each element in the population has an equal probability of inclusion in the sample.
Systematic sampling: variation on simple random sampling involves taking every kth element listed in a sampling frame.
Stratified Random Sampling
Stratified sampling involves dividing the population into smaller subgroups, called strata, and then drawing separate random or systematic samples from each of the strata.
Needs to be a reflective ratio of the population
Types of Stratified Sampling
%-->Proportionate Sampling: the size of the sample from each stratum is proportionate to occurrence in the population.
Goal is to reduce sampling error
Disproportionate Sampling, that is over sample: sufficient proportion is selected from each sample to make statistical comparisons
Goal is to have representative sub-sample for each stratum.
Area sampling is also called "cluster sampling" or "multi-stage sampling."
-First stage: select a sample of areas, e.g. census tracts.
-Intermediate stage(s): select sample of smaller areas within each area selected in first stage.
-Final stage: select sample of units from each area selected in previous step.
1. Randomly select states
2. Identify census tracts in each selected state
3. Select random sample of Census Tracts
4. Make random selection of blocks within each tract
5. Take Random Sample of Households from each block
How Large a Sample?
How many cases are needed for the research hypotheses?
Precision: how much error can we accept?
Population homogeneity: the more variability in the population to be sampled the larger the sample required.
Sampling fraction: the number of elements in the sample relative to the number of elements in the population.
*Note: bigger spoon for the chili; smaller spoon for tomato soup (homogenous)
Nonprobability samples are those in which the investigator does not know the probability of each population element's being included in the sample.
-No intent to generalize
-Qualitative study focused on process of phenomena
-Impossible to develop sampling frame
-Cannot specify representativeness
-Degree of sampling error is unknown
-Inferential statistical test may assume probability sample<--only with probability sample
Nonprobability Sample Types (5)
1. Availability sampling (convenience or accidental sampling)
2. Snowball sampling (interactive sampling) rely on interaction of persons to generate sample
3. Quota sampling
4. Purposive (or judgmental) sampling
5. Dimensional sampling is technique for selecting small samples to enhance representativeness
-Specify important dimensions
-Choose a sample that includes at least one case representing each possible combination of dimensions