Multimedia Marketing/Marketing.iso/market/chapter4.8p

Text File | 1996-08-22 | 15KB | 330 lines

SECTION 4.8  Data Acquisition and Analysis

INSTRUCTIONS Try to answer this question--What are the best ways of acquiring
information from subjects and of analyzing it so that it is useful in solving marketing problems?

EXAMPLE

A life insurance company decided to conduct a survey to determine
the image of the company among recently-graduated college students.
It hired six students to interview target customers during three
summer months. The results were not satisfactory.

Two of the students decided that they would not work very hard that
summer. They spent the bulk of it in their rooms, filling out fake
interview questionnaires without actually interviewing target consumers.
In this way, they could falsify a day's work in less than a half hour's
work.

Unfortunately, it was only after several months that company managers
stumbled upon the cheating and terminated the guilty parties. By this
time, a large amount of the data which had been acquired was useless,
and the company decided to terminate the entire project.

DETAILS

In one of the later phases of marketing research, data are acquired.
This can be very time consuming. A cigarette company test marketed a
new cigarette for eight months, for instance. Direct mail surveys
typically take up to two months for responses to be returned.

The research instrument, the sample, and the information needs
of management dictate the procedure to follow in this stage. Some
firms use part-time help to collect the data because of cost
economies. Students, homemakers , and retirees are relatively
inexpensive to hire for this task, in most cases.

On the other hand, part-timers usually require considerable training
and supervision to assure that procedures are followed accurately
and consistently. For this reason, many companies hire good field
research firms who maintain professional staffs of competent and
fully-trained interviewers to collect data.

Nevertheless, problems still may emerge when companies contract out
personal interview work to a specialty firm. Field personnel
often receive their pay on a piecework basis, sometimes without
close field supervision, and interviewers may cut corners; violate
instructions in order to speed up the survey process; or even
fabricate ficticious interviews and turn in counterfeit measuring
instruments. Most reputable field research firms verify collected
information by having auditors recontact a sample of a survey's
respondents (often ten per cent) to assure that they were in fact
surveyed and that instructions were followed.

PROBLEM 1

A marketer of pickles is preparing to conduct a study of consumer
preferences for pickles of various sizes. It will hire a number
of interviewers to collect the data. Which of the following is a
good procedure to minimize or prevent counterfeit interviews:

A. Pay interviewers by the hour instead of by piecework.
B. Hire college students and retirees.
C. Hire homemakers.
D. Select interviewers who are highly educated.

WORKED

If a pickle marketer is doing a survey and wants to minimize or
prevent counterfeit interviews, a good strategy is to pay interviewers
by the hour instead of by piecework. If they receive money for the
latter, they are motivated to turn in as many completed questionnaires
and is feasible, because they receive direct compensation for this.
So, there is motivation to cut corners or to fabricate fake interviews.
Paying by the hour eliminates this incentive.

ANSWER  A
SECTION 4.8  Data Acquisition and Analysis

INSTRUCTIONS Try to answer this question--What are the best ways of acquiring
information from subjects and of analyzing it so that it is useful in solving marketing problems?

EXAMPLE

A life insurance company decided to conduct a survey to determine
the image of the company among recently-graduated college students.
It hired six students to interview target customers during three
summer months. The results were not satisfactory.

Two of the students decided that they would not work very hard that
summer. They spent the bulk of it in their rooms, filling out fake
interview questionnaires without actually interviewing target consumers.
In this way, they could falsify a day's work in less than a half hour's
work.

Unfortunately, it was only after several months that company managers
stumbled upon the cheating and terminated the guilty parties. By this
time, a large amount of the data which had been acquired was useless,
and the company decided to terminate the entire project.

DETAILS

After acquiring the data, researchers analyze and interpret it.
Analysis involves breaking data down into meaningful categories
and studying differences and relationships between them. It includes
selecting and applying some type of mathematical model to test a
study's hypotheses. The objective is to uncover relevant variable
parameters (such as means and medians) and associations between
variables.

Raw data, standing alone, are not very meaningful to marketing
managers. It would not be helpful to a health food store manager
to know the annual volume of purchases by each of the store's
many customers, for instance. But the mean annual purchase figure
probably would be useful. The manager might want to compare the
mean figure for various market segments, such as various income
groups. This might indicate that the store is appealing primarily
to middle income groups.

Another type of analysis that could be useful would be to determine
the relationship between purchases in the store and age. Regression
analysis is useful for this purpose. This method analyzes the
relationship between two or more variables. It might indicate that, as
consumers grow older, they tend to patronize the store more. This
would be useful to the manager in developing a segmentation strategy.
He could aim at older middle income groups.

Interpretation means making the results of research meaningful to
managers. The latter do not want to be confronted with a vast
sea of numbers. Rather, they want summary statements of what the
data are, what they mean to the company, and what action prescriptions
they suggest. A well-written research report sets forth clearly-
articulated interpretations to managers. The writen report should
be followed by an oral presentation of the major findings.

PROBLEM 2

A clothing wholesaler has collected data on purchases by retailers
and wants to analyze it. This involves:

A. Breaking data down into meaningful categories and studying
   differences and relationships between them.
B. Determining and reporting on the significance of the data for
   marketing decisions.
C. Determining the sample size that will yield both valid and
   reliable research results.
D. Making recommendations to management on future marketing strategy,
   based on research results.

WORKED

A clothing wholesaler has collected data on purchases by retailers
and wants to analyze it. This involves breaking data down into
meaningful categories and studying differences and relationships
between them. The wholesaler, for instance, might break down the
sales data between large and small retailers. Then it could determine
the mean value of purchases of these two classes and compare the
two. This might reveal that small retailers, while their number is
large, are not buying much, on average, compared to large retailers.
This information might convince the retailer to concentrate its
marketing effort on large retailers.

ANSWER  A
SECTION 4.8  Data Acquisition and Analysis

INSTRUCTIONS Try to answer this question--What are the best ways of acquiring
information from subjects and of analyzing it so that it is useful in solving marketing problems?

EXAMPLE

A life insurance company decided to conduct a survey to determine
the image of the company among recently-graduated college students.
It hired six students to interview target customers during three
summer months. The results were not satisfactory.

Two of the students decided that they would not work very hard that
summer. They spent the bulk of it in their rooms, filling out fake
interview questionnaires without actually interviewing target consumers.
In this way, they could falsify a day's work in less than a half hour's
work.

Unfortunately, it was only after several months that company managers
stumbled upon the cheating and terminated the guilty parties. By this
time, a large amount of the data which had been acquired was useless,
and the company decided to terminate the entire project.

DETAILS

There are two major categories of mathematical models used in marketing.
These are statistical models and management science.

Statistical models are used most often in marketing. Multivariate
statistical analysis is used extensively with survey information
in order to calculate how much of a dependent variable's fluctuation
can be explained by fluctuations in the values of independent
variables. The statistical procedures which are most often employed
are called  multiple regression analysis, discriminant analysis,
cluster analysis, and factor analysis.

These multivariate methods are useful because they can show the
relationships between groups of variables. A multivariate study
might show, for instance, that the heaviest customers of a health
food store are age 55-64, have incomes between ,30,000 and ,40,000,
live in tract homes in suburbs, own personal computers, and are
members of the lower middle social class. This example illustrates
how multivariate methods can simulataneously study the operation
of several variables (in this case age, income, housing, computer
ownership, and social class). This kind of analysis is much more
meaningful than only studying one or two variables at a time.

With the passage of time, marketing researchers are also increasingly
called upon to solve problems of optimization. Determining optimal
allocations of advertising dollars among various media, the best
sales force territory assignments, and the most efficient shipping
routes are examples of problems that cannot be solved using statistical
procedures. Linear programming, integer programming, and a group of
similar specialized techniques are used for such purposes.

PROBLEM 3

A marketing researcher for a company that produces paints, stains,
and adhesives becomes involved in a number of data analysis
projects. Which of the following is an example of statistical
analysis?

A. Making comparisons of the average sales of company products in
   hardware stores with average sales in discount stores.
B. Determining how much space each product should occupy in company
   warehouses.
C. Planning the most efficient travel itineraries for sales
   representatives.
D. Determining how much research and development time should be
   devoted to each product line.

WORKED

A marketing researcher for a company that produces paints, stains,
and adhesives is engaging in statistical analysis when she compares
average sales of company products in hardware stores with average
sales in discount stores. In this case, the researcher is taking
measurements of two segments of the population (hardware and
discount stores) and comparing them. If there are statistically
significant differences between the two, the analyst will note
this and report the differences to management. This may lead to
new decisions by management, such as having sales representatives
spend more time with discount store buyers than they did in the
past.

ANSWER  A
SECTION 4.8  Data Acquisition and Analysis

INSTRUCTIONS Try to answer this question--What are the best ways of acquiring
information from subjects and of analyzing it so that it is useful in solving marketing problems?

EXAMPLE

A life insurance company decided to conduct a survey to determine
the image of the company among recently-graduated college students.
It hired six students to interview target customers during three
summer months. The results were not satisfactory.

Two of the students decided that they would not work very hard that
summer. They spent the bulk of it in their rooms, filling out fake
interview questionnaires without actually interviewing target consumers.
In this way, they could falsify a day's work in less than a half hour's
work.

Unfortunately, it was only after several months that company managers
stumbled upon the cheating and terminated the guilty parties. By this
time, a large amount of the data which had been acquired was useless,
and the company decided to terminate the entire project.

DETAILS

After the analysis, it is the responsibility of the research team
to interpret the data for management. This consists of informing
management, usually in both oral and written reports, what the study
uncovered and the meaning of the research to management. Included
should be a discussion of both accepted and rejected hypotheses,
assessment of their meaning, and a clear explanation of the
assumptions of the analysis. The statement of assumptions is extremely
important, and unfortunately it is sometimes forgotten by naive and
technique-oriented researchers.

It should be kept in mind that the purpose of practical research is
to assist management in decision making and nothing more. Esoteric
research and interpretations are not needed.

A research project should not be forgotten upon completion; in the
final stage, the analysts should specify follow-up procedures.
Perhaps they should perform a future study to determine if conditions
change over time or test some of the assumptions to learn if they
were appropriate. Proper follow-up procedures anticipate changing
conditions or incorrect conclusions that may have been drawn because
of erroneous assumptions or other reasons. By performing follow-up
studies, possible trouble spots may be uncovered before it is too
late to take corrective action.

PROBLEM 4

A company that produces pillows and comforters is conducting a
consumer study and a research analyst is preparing a statement
of interpretations. Which one of the following is not necessary
in this statement?

A. An explanation of the assumptions of the analysis.
B. A discussion of rejected and accepted hypotheses.
C. An assessment of the meaning of the hypotheses.
D. A discussion of the sample which was tested in the research.

WORKED

A company that produces pillows and comforters should include in
its statement of interpretations an explanation of the assumptions
of the analysis, a discussion of rejected and accepted hypotheses,
and an assessment of the meaning of the hypotheses. In interpretation,
the researchers are summing up the gist of the results of the study
to management. They are informing management what the study uncovered
and what this means to managerial decision making. The statement of
interpretation is that part of the research report that usually
most interests marketing managers because it provides them with
action prescriptions.

ANSWER  D