Types of Samples: Probability Sampling (Representative Samples)
Types of Samples: Probability Sampling (Representative Samples)
Types of Samples: Probability Sampling (Representative Samples)
The best sampling is probability sampling, because it increases the likelihood of obtaining
samples that are representative of the population.
Random sample
The term random has a very precise meaning. Each individual in the
population of interest has an equal likelihood of selection. This is a
very strict meaning -- you can't just collect responses on the street and
have a random sample.
The assumption of an equal chance of selection means that sources such as a telephone
book or voter registration lists are not adequate for providing a random sample of a
community. In both these cases there will be a number of residents whose names are not
listed. Telephone surveys get around this problem by random-digit dialing -- but that
assumes that everyone in the population has a telephone. The key to random selection is
that there is no bias involved in the selection of the sample. Any variation between the
sample characteristics and the population characteristics is only a matter of chance.
Stratified sample
Stratified samples are as good as or better than random samples, but they require a fairly
detailed advance knowledge of the population characteristics, and therefore are more
difficult to construct.
As they are not truly representative, non-probability samples are less desirable than
probability samples. However, a researcher may not be able to obtain a random or stratified
sample, or it may be too expensive. A researcher may not care about generalizing to a
larger population. The validity of non-probability samples can be increased by trying to
approximate random selection, and by eliminating as many sources of bias as possible.
Quota sample
Purposive sample
A subset of a purposive sample is a snowball sample -- so named because one picks up the
sample along the way, analogous to a snowball accumulating snow. A snowball sample is achieved by
asking a participant to suggest someone else who might be willing or appropriate for the study.
Snowball samples are particularly useful in hard-to-track populations, such as truants, drug users, etc.
Convenience sample
Non-probability samples are limited with regard to generalization. Because they do not truly
represent a population, we cannot make valid inferences about the larger group from which
they are drawn. Validity can be increased by approximating random selection as much as
possible, and making every attempt to avoid introducing bias into sample selection.
Sampling: Sample size
In a very general sense, the larger the sample, the better -- because larger samples tend to
be more similar to the population from which they are drawn. However, if the population of
interest is small, then the sample can be relatively small. Large samples require more time
for data collection and analysis, and are therefore more costly than smaller ones.
1. population size
2. available resources (time, money)
3. strength of effect being measured
4. refusal and spoilage rates
5. number of analyses to be performed
Three methods
Signal contingent
The participant is signaled with a beeper, cell phone call, or programmed watch at random
times within a fixed time period (e.g., between 8 AM and 9 PM). At the signal, the
participant records the behavior of interest (e.g., activity, location, mood, thoughts).
Example: The focus of this study(1) was home activity and school performance of Latino
children. Researchers randomly selected eight 10 and 11-year olds from an ongoing
longitudinal study. The children were paged randomly within a time block of 4 PM-6 PM on
Monday to Saturday for 12 weeks. About 3 calls a day were made. The children received
$.50 for each returned call. The fieldworker (on the phone) asked "What were you doing
when the pager went off?" along with a few follow-up questions. The accumulated responses
were subsequently coded into more general activity categories.
Interval contingent
Participant is assigned pre-set intervals for reporting events. For example, before going to
bed at night, the participant fills out a checklist of the day's activities.
Event contingent
The event is determined by the research project, for example, migraine headache, or
phoning a friend. The participant makes a record whenever the key event occurs. The
recording of the event depends upon (is contingent) on its occurrence.
These reports are personal and subjective. They can be quantified by coding and
counting.
The data are similar to those obtained by diaries. An advantage is that the signal
makes then less dependent upon memory.
The data tend to be a little thin -- don't have the richness of ethnography. It is only
a small sample of behavior, and little is known about the setting or context