VI.NONPROBABILITY SAMPLING
Convenience Sampling
Convenience sampling refers to sampling by obtaining people or units that are
conveniently available.
Researchers generally use convenience samples to obtain a large number of completed
questionnaires quickly and economically, or when obtaining a sample through other
means is impractical.
Research looking for cross-cultural differences in organizational or consumer behavior
typically uses convenience samples.
Judgment Sampling
Judgment (purposive) sampling is a nonprobability technique in which an experienced
individual selects the sample based on his or her judgment about some appropriate
characteristics required of the sample members.
The consumer price index (CPI) is based on a judgment sample of market-basket items,
housing costs, and other selected goods and services expected to reflect a representative
sample of items consumed by most Americans.
Test-market cities often are selected because they are viewed as typical cities whose
demographic profiles closely match the national profile.
Often used in attempts to forecast election results.
Political and sampling experts judge which small voting districts approximate overall
state returns from previous election years.
Then, these bellwether precincts are selected as the sampling units.
The assumption is that the past voting nature of these districts is still representative of
the state’s political behavior.
Quota Sampling
The purpose of quota sampling is to ensure that the various subgroups in a population
are represented on pertinent sample characteristics to the exact extent that the
investigators desire.
In quota sampling, the interviewer has a quota to achieve.
Aggregating the various interview quotas yields a sample representing the desired
proportion of the subgroups.
Possible Sources of Bias
The logic of classifying the population by pertinent subgroups is essentially
sound.
However, because respondents are selected according to a convenience sampling
procedure rather than on a probability basis (as in stratified sampling) the
haphazard selection of subjects may introduce bias.
Quota samples tend to include people who are easily found, willing to be
interviewed, and middle class.
Advantages of Quota Sampling
The major advantages over probability sampling are
speed of data collection
lower costs
convenience
Although there are many problems with this method, careful supervision of the
data collection may provide a representative sample for analyzing the various
subgroups within a population.
May be appropriate when the researcher knows that a certain demographic group
is more likely to refuse to cooperate with a survey (e.g., older men).
Snowball Sampling
Snowball sampling refers to a variety of procedures in which initial respondents are
selected by probability methods, but additional respondents are then obtained from
information provided by the initial respondents.
This technique is used to locate members of rare populations by referrals.
Reduced costs and sample sizes are clear-cut advantages of snowball sampling.
However, bias is likely to enter into the study because a person suggested by
someone also in the sample has a higher probability of being similar to the first
person.
If there are major differences between those who are widely known by others and
those who are not, there may be some serious problems with this technique.
Since the focus group is not expected to be a generalized sample, snowball sampling
may be very appropriate.
VII. PROBABILITY SAMPLING
All probability samples are based on chance selection procedure, which eliminates the bias
inherent in nonprobability sampling procedures because the probability sampling process is
random.
Randomness characterizes a procedure whose outcome cannot be predicted because it
depends on chance.
oIt should not be thought of as unplanned or unscientific—it is the basis of all
probability sampling techniques.
There are several probability sampling techniques discussed below.
Simple Random Sampling
Simple random sampling is a sampling procedure that assures that each element in the
population will have an equal chance of being included in the sample.
Drawing names from a hat is a typical example of simple random sampling; each person
has an equal chance of being selected.
This process is simple because it requires only one stage of sample selection, in contrast
to other, more complex probability samples.
When populations consist of large numbers of elements, tables of random numbers or
computer-generated random numbers are utilized for sample selection.
Selecting a random sample: to use a table of random numbers, a serial number is
assigned to each element of the population. Then, assuming a population of 99,999 or
less, five-digit numbers are selected from the table of random numbers merely by reading
the numbers in any column or row, by moving upward, downward, left, or right. A
random starting point should be selected at the outset.
The random digit dialing technique of sample selection requires that the researcher
identify the exchange or exchanges of interest (the first three numbers) and then use a
table of numbers to select the next four numbers.
Systematic Sampling
Extremely simple.
An initial starting point is selected by a random process; then every nth number on the list
is selected.
To illustrate this procedure, suppose one wishes to take a sample of 1,000 from a list
consisting of 200,000 names. Using systematic selection every 200th name from the list
will be drawn. In this example, the sampling interval is 200.
While this procedure is not actually a random selection procedure, it does yield random
results if the arrangement of the items in the list is random in character.
The problem of periodicity occurs if a list has a systematic pattern, that is, if it is not
random in character.
Stratified Sampling
The first step of choosing strata on the basis of existing information is the same for both
stratified and quota sampling.
However, the process of selecting sampling units within the strata differs substantially.
In stratified sampling, a subsample is drawn using simple random sampling within each
stratum. This is not true with quota sampling.
The reason for taking a stratified sample is to have a more efficient sample than would be
taken on the basis of simple random sampling.
Random sampling error will be reduced because each group is internally homogeneous
but there are comparative differences between groups.
More technically, a smaller standard error may result from this sampling because the
groups will be adequately represented when strata are combined.
Another reason for selecting a stratified sample is to ensure that the sample will
accurately reflect the population on the basis of the criterion or criteria used for
stratification.
Occasionally a simple random sample yields a disproportionate number of one
group or another and the representativeness of the sample could be improved.
A researcher selecting a stratified sample will proceed as follows:
A variable (sometimes several variables) is identified as an efficient basis for
stratification.
The variable chosen should increase the homogeneity within each stratum
and increase the heterogeneity between strata.
The stratification variable is usually a categorical variable or one easily
converted into categories, that is, subgroups.
For each separate subgroup or strata, a list of population elements must be obtained,
but if a complete listing is not available, a true stratified probability sample cannot be
selected.
A table of random numbers or some other device is then used to take a
separate random sample within each stratum.
The researcher must determine how large a sample must be drawn for each
stratum.
Proportional versus Disproportional Sampling
If the number of sampling units drawn from each stratum is in proportion to the relative
population size of the stratum, the sample is a proportional stratified sample.
In a disproportional stratified sample the sample size for each stratum is not allocated
in proportion to the population size but is dictated by analytical considerations (i.e.,
variability in store sales volume).
Cluster Sampling
The purpose of cluster sampling is to sample economically while retaining the
characteristics of a probability sample.
In a cluster sample, the primary sampling unit is no longer the individual element in the
population (e.g., grocery stores) but a larger cluster of elements located in proximity to
one another (e.g., cities).
The area sample is the most popular type of cluster sample.
Cluster sampling is classified as a probability sampling technique because of either the
random selection of clusters or the random selection of elements within each cluster.
Cluster samples frequently are used when lists of the sample population are not available.
Ideally a cluster should be as heterogeneous as the population itself—a mirror image of
the population.
A problem may arise with cluster sampling if the characteristics and attitudes of the
elements within the cluster are too similar.
This problem may be mitigated by constructing clusters composed of diverse
elements and by selecting a large number of sampled clusters.
Multistage Area Sampling
Multistage area sampling involves two or more steps that combine some of the
probability techniques already described.
Typically, geographic areas are randomly selected in progressively smaller
(lower-population) units.
The U.S. Bureau of the Census provides maps, population information, demographic
characteristics for population statistics, and so on, by several small geographical areas
that may be useful in sampling.
VIII. WHAT IS THE APPROPRIATE SAMPLE DESIGN?
A researcher who must decide on the most appropriate sample design for a specific project
will identify a number of sampling criteria and evaluate the relative importance of each
criterion before selecting a sampling design.
Exhibits 16.8 and 16.9 summarize the advantages and disadvantages of each nonprobability
sampling technique and for the probability sampling techniques, respectively.
Degree of Accuracy
The degree of accuracy required or the researcher’s tolerance for sampling and
nonsampling error may vary from project to project, especially when cost savings or
other considerations may be a trade-off for a reduction in accuracy.
Resources
The cost associated with the different sampling techniques varies tremendously.
If the researcher’s financial and human resources are restricted, certain options will have
to be eliminated.
Managers concerned with the cost of the research versus the value of the information
often will opt for cost savings from a certain nonprobability sample design rather than
make the decision to conduct no research at all.
Time
Researchers who need to meet a deadline or complete a project quickly will be more
likely to select simple, less time-consuming sample designs.
Advance Knowledge of the Population
In many cases, a list of population elements will not be available to the researcher.
A lack of adequate lists may automatically rule out systematic sampling, stratified
sampling, or other sampling designs, or it may dictate that a preliminary study, such as a
short telephone survey using random digit dialing, be conducted to generate information
to build a sampling frame for the primary study.
National versus Local Project
Geographic proximity of population elements will influence sample design.
When population elements are unequally distributed geographically, a cluster sample may
become much more attractive.
IX. INTERNET SAMPLING IS UNIQUE
Internet surveys allow researchers to reach a large sample rapidly—both an advantage and a
disadvantage.
If rapid response rates are expected, the sample for an Internet survey should be metered out
across all time zones.
In addition, for some populations, people are more likely to go online during the weekend
than on a weekday.
If the researcher can anticipate a day-of-the-week effect, the survey should be keep open
long enough so all sample units have the opportunity to participate in the research
project.
The ease and low cost of an Internet survey also has contributed to a flood of online
questionnaires, some more formal than others.
Another disadvantage of Internet surveys is the lack of computer ownership and Internet
access among certain segments of the population.
A sample of Internet users is only representative of Internet users, who tend to be
younger, better educated, and more affluent.
This is not to say that all Internet samples are unrepresentative of all target populations.
Nevertheless, when using Internet surveys, researchers should be keenly aware of
potential sampling problems because some members of target populations do not have
Internet access.
Website Visitors
Many Internet surveys are conducted with volunteer respondents who visit an
organization’s website intentionally or by happenstance.
These unrestricted samples are clearly convenience samples.
They may not be representative because of the haphazard manner by which many
respondents arrived at a particular website or because of self-selection bias.
A better technique for sampling website visitors is to randomly select sampling units.
Randomly selecting website visitors can cause a problem.
It is possible to over-represent the more frequent visitors to the site, and thus
represents site visits rather than visitors.
There are several programming techniques and technologies that can help accomplish
more representative sampling based on site traffic (i.e., cookies, registration data, or
pre-screening).
This type of sampling is most valuable if the target population is defined as visitors to a
particular website.
Evaluation and analysis of the visitors’ perceptions and experiences of the website
would be a typical survey objective with this type of sample.
Panel Samples
Drawing a probability sample from an established consumer panel or other pre-recruited
membership panel is a popular, scientific, and effective method for drawing Internet
samples.
Typically samples from a panel yield a high response rate because panel members have
already agreed to cooperate with the research organization via e-mail and the Internet.
Because the panel has already supplied demographic characteristics and other
information from previous questionnaires, researchers have the ability to select panelists
based on product ownership, lifestyle, or other characteristics.
Consider Harris Interactive Inc. Harris Interactive is an Internet survey research
organization that maintains a United States panel of more than 6.5 million individuals.
Because Harris Interactive knows that all demographic groups are not fully accessible via
the Internet, it uses a propensity-weighting scheme to ensure that survey results are
representative. The research company does parallel studies – phone as well as Internet – to
test the accuracy of its Internet data gathering capabilities.
Recruited Ad Hoc Samples
Another means for obtaining an Internet sample is to obtain or create a sampling frame
of e-mail addresses on an ad hoc basis.
Databases containing e-mail addresses can be compiled from many sources including
customer/client lists, advertising banner recruiting survey participants, online
sweepstakes, pop-up windows and registration forms that must be filled out in order to
gain access to a particular website.
Researchers may contact respondents by “snail mail” or by telephone to ask for their
email addresses and obtain permission for an Internet survey.
Using offline techniques, such as random-digit dialing and a short telephone screening
interview, to recruit respondents can be a very practical way to get a representative
sample for an Internet survey.
For companies anticipating future Internet research, adding an optional e-mail
registration into customer relationship databases (product registration cards, telephone
interactions, on-site registration, etc.) can prove to be a valuable database for sample
recruitment.
Opt-in Lists
Another means for obtaining an Internet sample is to obtain list of e-mail addresses from
individuals who opt-in, that is given permission to receive e-mail messages related to a
particular topic of interest.
It is important not to send unauthorized e-mail to respondents.
Spamming is not tolerated by experienced Internet users and can backfire, creating a host
of problems—the most extreme being complaints to the Internet service provider (ISP),
which may shut down the survey site.