Page 129 - Foundations of Marketing

96 Part 2 | Marketing Research and Target Markets

Sampling Because the time and resources available for research are limited, it is almost
population All the elements, impossible to investigate all the members of a target market or other population. A population ,
units, or individuals of interest or “universe,” includes all the elements, units, or individuals of interest to researchers for a
to researchers for a specific specific study. Consider a Gallup poll designed to predict the results of a presidential election.
study All registered voters in the United States constitute the population. By selecting a limited
sample A limited number number of units—a sample —to represent the characteristics of a total population, researchers
of units chosen to represent can predict the behaviors of the total population. Sampling in marketing research, therefore,
the characteristics of a total is the process of selecting representative units from a population. Sampling techniques allow
population marketers to predict buying behavior fairly accurately without having to collect responses
sampling The process of from a total population. Because it would be impossible in most situations to collect reactions
selecting representative units from the entire market or market segment, most types of marketing research employ sampling
from a total population techniques.
probability sampling A type There are two basic types of sampling: probability sampling and nonprobability sam-
of sampling in which every pling. With probability sampling , every element in the population being studied has a known
element in the population being chance of being selected for study. Random sampling is a form of probability sampling. When
studied has a known chance of marketers employ random sampling , all the units in a population have an equal chance of
being selected for study
appearing in the sample. Likewise, the various events that can occur have an equal or known
random sampling A form of chance of taking place. For example, a specific card in a regulation playing deck has a 1 in
probability sampling in which 52 probability of being drawn. Sample units are ordinarily chosen by selecting from a table
all units in a population have
an equal chance of appearing of random numbers statistically generated so that each digit, 0 through 9 , will have an equal
in the sample, and the various probability of occurring in each position in the sequence. The sequentially numbered elements
events that can occur have of a population are sampled randomly by selecting the units whose numbers appear in the
an equal or known chance of table of random numbers. There are random number generators available for free online, such
taking place as Random.org, that will generate lists of random numbers for this purpose.
Another type of probability sampling is stratifi ed sampling , in
which the population of interest is divided into groups according
to a common attribute, and a random sample is then chosen within
each subgroup. A stratified sample may reduce some of the error
that is a risk of a completely random sample, ensuring that a group
is not accidentally overrepresented. By segmenting a population
into groups, the researcher makes sure that each segment receives
its proportionate share of sample units and helps investigators avoid
including too many or too few sample units from each subgroup.
Samples are usually stratified when researchers believe there may
be variations among different types of respondents. For instance,
many political opinion surveys are stratified by gender, race, age,
and/or geographic location.
The second type of sampling, nonprobability sampling , is
more subjective than probability sampling because there is no
way to calculate the probability that a specific element of the
population being studied will be chosen. Quota sampling, for
example, is highly judgmental because the final choice of par-
ticipants is left to the researchers. In quota sampling , research-
Courtesy of Tobii Technology, Inc. participants from each segment. In quota sampling, researchers
ers divide the population into groups and then arbitrarily choose

impose some controls—usually limited to two or three variables,
such as age, gender, or race—over the selection of participants
to ensure that representative categories of respondents are
included. A study of people who wear eyeglasses, for example,
may be conducted by interviewing equal numbers of men and
women. However, because quota samples are not probability
Collecting Data
Tobii Technology assists clients in providing technology samples, not everyone has an equal chance of being selected
to study eye movements on screen to collect primary data and sampling error therefore cannot be measured statistically.
and gain insight into customers’ responses to marketing Quota samples are used most often in exploratory studies, when
communications. researchers have not yet generated hypotheses to test. In this

Copyright 2014 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s).
Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.