Sampling method in marketing research

1. Sampling method in marketing research

2.

When conducting marketing research, it is
almost always impossible to interview all
consumers. Even if it were possible, such a
study turned out to be too expensive.
Therefore, when conducting marketing
research, the selective method is widely
used. A sample study is a type of non—
continuous study based on the principle of
random (probabilistic) selection, conducted
with the aim of subsequent dissemination
of the data obtained to the general
population.

3.

The general
population
A sample
population
• is a complete list of units of the studied
population, from the composition of
which some part of the units that are
subject to direct research is selected.
• is a part of the general population that
is directly included in the study.

4.

The sampling method provides the following advantages in comparison
with a continuous survey:
— significant savings in financial and material
costs;
— reducing the time for conducting research and
obtaining results;
— obtaining deeper data by expanding the
research program and questionnaire.

5.

One of the main requirements for
the sample is its representativeness.
Representativeness (or
representativeness) means that the
results obtained from the sample
can be extended to the entire
general population with a certain
error. Moreover, this error can be
calculated.

6.

When designing a sample in a marketing study, the following problems should be solved.
1. Definition of the unit of the general population.
2. Defining the boundaries of the general population.
3. Making a list of the general population.
4. Choosing the method of selectin

7.

Definition of the unit of the general population.
The general population covers all persons who can potentially participate in
marketing research. The unit of the general population in most cases is an individual
(for example, for the cigarette market), in some cases it may be a family (for example,
for the refrigerator market) or an enterprise (for industrial goods markets).
It is very important to distinguish between a unit of the general population and a
respondent who will act as a direct source of information. A similar situation occurs
when the unit of the general population is a family or an organization. So, when
researching food for family consumption (for example, sausages), it is most expedient
to interview the person responsible for the purchases in the family. When marketing
research concerns household appliances, it is more expedient to interview both spouses
at the same time, and in some cases also children. At enterprises, the respondent is
usually the person responsible for procurement, or if the enterprise is small, its head.

8.

Defining the boundaries of the general population.
There are three main options here. The general population consists of:
1) consumers of the
investigated product produced
by our company (i.e. our
customers). For example, if the
degree of customer satisfaction
with service is being
investigated, then it does not
make sense to interview those
who use competitors' products,
since they do not have the
necessary information;
2) all consumers of the product
under study, regardless of the
manufacturer (i.e. our customers
and buyers of competing firms).
To assess the level of quality of
our goods and services, it is not
enough to interview only our
consumers, in this case the
assessment will be too unreliable.
After all, the opinion of our
customers and buyers of
competitors' products is radically
different;
3) everyone who can potentially
consume the product under
study. So, it is possible to
investigate not only car owners,
but also all those who, for
medical reasons, have the
opportunity to obtain a driver's
license. Such citizens are
potential consumers.

9.

3. Making a list of the general population.
In marketing research, it is very important to make a list of members of the general
population. However, this is usually a big problem. In this regard, some marketers even
believe that the formation of a random sample in marketing research is impossible.
If a survey is conducted on the street or in a shopping center, then such a sample is a rough
approximation to a random selection of respondents. Consumers who do not walk along
these streets or do not visit these stores will not be able to get into the sample. It is simply
impossible to compile and use a list of members of the general population in such a study.
When interviewing by phone or at the place of residence, the marketer has a database. This
base can be considered as a general set. Of course, it is usually incomplete and contains
errors, but this option is much closer to the selective method from the point of view of
statistics.
In any case, the marketer should be prepared for the fact that the error of the sampling
method in real marketing research will be much greater than that obtained by calculating
standard statistical procedures.

10.

4. Choosing the method of selecting respondents.
The following methods of selecting units from the general population in the sample are most often used:
1) The actual random
method. The sample is
formed as a result of random
selection of individual units
from the general population.
Here you can use the
Random function in
Microsoft Excel or tables of
random numbers;

11.

2.Mechanical method. Here, the selection of units in the sample is made from the general population,
divided into equal groups. One respondent is selected from each group, randomly from the first, and
from the next with a step equal to the size of the group. Suppose the general population is 500 thousand
people (the adult population of a certain city), and the sample size was assumed to be 500 respondents.
It turns out that one respondent must be selected from every thousand people. From the first thousand
we select randomly. For example, the 135th person was selected, then we further need to interview the
1135th person, 2135th, etc.;
3. Stratified method. Initially, the general population is divided into homogeneous typical groups
(strata), then a certain number of units are selected from each group using a purely random or
mechanical method. Such a selection scheme ensures the presence of representatives of all typical
groups in the sample, i.e. a high representativeness of the sample in terms of structure, taking into
account the considered trait. Thus, when using a typical sample, compared with the actual random
sample, the error is reduced.

12.

4) Serial, or "nest" method. In this case, not individual respondents are selected, but
whole groups at once, all participants of which are included in the sample. For
example, if a survey of students is conducted at a university, it is more expedient to
select not individual students, but entire groups at once. Further, all students are
examined within these groups;
5) The "snowball" method. Such a selection of respondents is used in rare cases when
a small percentage of the population falls under the conditions of the general
population, and it is impossible to compile a list of the general population. Suppose it is
necessary to conduct an expert assessment in the field of marketing with the
involvement of a very wide range of specialists. Since it is impossible to get a list of
marketing experts anywhere, it is effective to use the "snowball" method. At the
beginning there is a certain number of marketers who answer the expert questionnaire.
After that, they are asked to name their acquaintances who fit the required conditions.
After several such operations, the sample grows rapidly like a snowball.