pratik meshram -Unit 3 contemporary marketing research full notes pune university semister 3
1. Unit 3
Experimental Designs
and Sampling
TOPICS
3.1. Experimentation in Marketing Research
3.2.Classical Designs
3.3. Statistical Designs
3.4. Test Marketing
3.5. Advertising & Sales Promotion Campaigns
3.6. Sampling
3.7. Case Studies
2. • Experimentation is widely used in marketing research.
• Marketing experiments have been conducted in such diverse activities as
evaluating new products, selecting advertising copy themes, determining
the frequency of salespeople's calls, and evaluating all aspects of a movie
(including ending, pacing, music, and even the story line).
• Two general types of experimental designs existnatural and controlled.
• A natural experiment is one in which the investigator intervenes only to the
extent required for measurement, and there is no deliberate manipulation
of an assumed causal variable.
• “Nature” produces the changes.
3. Prior to 1960 experimentation in marketing research was very seldom. Thereafter there has
been a steady growth in this area. A major field where experimentation in marketing is
performed happens to be test marketing or market test. A controlled experiment is planned
carefully in a selected part of the market place. In most of the cases the objective of such an
experiment is to ascertain the probable sale of say a new product.
A) Meaning:
Experiments are performed to study the effects of the factor levels on the dependent
variable. An experimental design is a plan for running an experiment. The factors of an
experimental design are variables that have two or more fixed values, or levels. In a conjoint
or discrete-choice study, the factors are the attributes of the hypothetical products or
services, and the response is preference or choice
4. B) Experimental Designs:
A number of experimental designs have been developed to overcome and reduce the various
sources of invalidity. Experimental designs can be categorized into two broad groups classical
and statistical. Classical designs consider the impact of only one dependent variable at a time,
whereas statistical designs allow for examining the impact of two or more independent
variables.
The symbols used in experimental designs are as follows:
O = The measurement of a dependent variable.
X =The manipulation, or change, of an independent variable.
R = Random assignment of subjects (consumers, stores. and so on to experimental and
control groups.
E = Experimental effect, that is, the change in the dependent variable due to the
independent variable.
5. The following designs will be used in the discussion of classical experimental designs:
1) After-Only Design:
This design consists of measuring the dependent variable after exposing the test unit to the
experiment variable. The after-only design is achieved by changing the independent variable
and, after some period of time, measuring the dependent variable. It is diagrammed as
follows:
X O1
where X represents the change in the independent variable and the distance between X and
O represents the passage of some time period, O1 represents the measurement, a posttest, of
the dependent variable.
2)After-Only With Control Group:
This is the simplest of all the controlled experimental designs. In this design, only one
treatment is given and then both the experimental and the control groups are measured.
Symbolically it can be shown as follows:
It has been criticized on the ground that it does not concern itself with the pre-test.
6. The following designs will be used in the discussion of classical experimental designs:
3) Before-After Design:
The designs considered thus far have been "after-only“ designs, because there were "no-
before" measures. Another approach to improving the control is to add a before measure. It is
represented symbolically as follows:
O1 X O2
The before measure can be added to any design already patented.
4) Before-After With Control Group:
This design provides for pre-testing or before measurements. It can be shown symbolically as
follows:
This design provides for the selection of the experimental and control groups through the
random method. The design is able to control most of the sources of systematic error.
7. The following designs will be used in the discussion of classical experimental designs:
5 )Four-Group Six Study Design:
When the investigator has to obtain data from respondents in an undisguised manner the
before-after with control group design such as the preceding one is not suitable,
Symbolically, the design can be shown as follows:
This is a combination of designs above. The effect of the treatment can be measured in
8. Introduction:
Statistical designs consist of a series of bask experiments that allow for statistical
control and analysis of external variables. In other words several basic experiments
are conducted simultaneously. Thus, statistical designs are influenced by the same
sources of invalidity that affect the basic designs being used. Statistical designs offer
the following advantages:
1) The effects of more than one independent variable can be measured.
2) Specific extraneous variables can be statistically controlled
3) Economical designs can be formulated when each test unit is measured more
than once.
The most common statistical designs are the randomized block design.
9. The Latin square design and the factorial design discussed as below:
1) Randomized Block Design:
A randomized block design is useful when there is only one major external variable such as
sales, store size or income of the respondent that might influence the dependent variable.
The test units are blocked or grouped on the basis of the external variable. The researcher
must be able to identify and measure the blocking variable. By blocking, the researcher
ensures that the various experimental and control groups are matched closely on the external
variable.
2) Latin Square Design:
A Latin square design allows the researcher to statistically control two non interacting
external variables as well as to manipulate the independent variable. Each external or
blocking variable is divided into an equal number of blocks or levels. The independent
variable is also divided into the same number of levels.
3) Factorial Design:
A factorial design is used to measure the effects of two or more independent variables at
various levels. Unlike the randomized block design and the Latin square, factorial design
sallow for interactions between variables. An interaction is said to take place when the
simultaneous effect of two or more variables is different from the sum of their separate
effects.
10. Test marketing is a technique used during product development to determine how people
respond to a product. It can be used at many different phases of development to see whether or
not the public will buy the product, how the product may need to be adjusted to make it
appealing to the public, and how members of the public interact with the product. Using
information from test marketing, product developers can refine products to make them more
commercially viable before embarking on a widespread project launch.
A) Applications Related to Test Marketing:
Test marketing, also called market testing is an application of a controlled experiment done in
limited but carefully selected parts of the marketplace called test markets.
The two major objectives of test marketing are:
1) To determine market acceptance of the product,
2) To test alternative levels of marketing mix variables.
11. A) Applications Related to Test Marketing:
Test-marketing procedures may be classified as Standard test markets, Controlled and
Minimarket tests and Simulated test marketing.
1) Standard Test Market:
In a standard test market, test markets are selected and the product is sold through regular
distribution channels Typically, the company's own sales form is responsible for distributing
the product.
a) Design:
b) Duration:
2) Controlled Test Market:
In a controlled test market, the entire test-marketing program is conducted by an outside
research company. The research company guarantees distribution of the product in retail
outlets that represent predetermined percentage of the market.
3) Simulated Test Market:
It is also called a laboratory test or test market simulation simulated test market yields
mathematical estimates of market share based on initial reaction of consumers to anew
product.
12. A) Study of Effectiveness of Advertising:
In order to study the effectiveness of advertising with any degree of accuracy it is essential to
divide the whole process into various stages. The effectiveness of advertising is assessed at
each stage of the advertising campaign.
1) Analysis of Advertising Experience:
A continuous analysis of past advertising experience is a very useful first step except incase of
new advertisers or products such an analysis can provide a valuable basis for reviewing and
developing advertising strategy.
2) Surveys of Buying Behaviour of Consumers:
Surveys of buyer behaviour and consumer preferences are helpful in developing advertising
objectives and strategy. Such research will also be useful in monitoring changes in the target
audience.
3) Pre-Test Research:
The third area involves a pre-testing of the advertisements before they are released. Pre-
testing is very important as it provides an indication as lo the likely acceptance of the
advertisement or campaign by the target market.
4) Post-Test Research
The last stage of measuring the effectiveness of an advertisement or advertising campaign is
post-test research. Post-test research is concerned with finding out information about how
well the advertisement has succeeded in achieving its objective.
13. B) Effect on Sales Campaigns:
Advertising by itself has no direct impact on the sales. This is because advertising together
with many other factors affects the sales of the advertised product. Sales are affected by
factors such as expansion of the sales force, improved distribution lessened efforts on the
part of the competitors, general improvements in the business conditions and so on.
1) Direct Response by Mall or Telephone Orders:
The measurement of the effectiveness of advertising through an increase in Sales is possible
when the advertisement is so designed as to complete the entire sales transaction. This is
what happens in mail order selling.
2) Controlled Field Experiments:
A means for measuring the effectiveness of advertising involves the setting up of a controlled
experiment within a limited area. An advertising programme is undertaken in one or more
test cities. A tabulation of the sales volume is made for the advertised product in the retail
stores of these cities before, during and after the test ads are run. At the same time a check
is also kept on the sales in one of the controlled citied where no advertisement campaign is
conducted.
14. A) Meaning:
Sampling is based on the law of statistical regularity and the law of inertia of large numbers.
It is a process of selecting a small representative group from the whole for intensive study
purpose. A sample is a part of the whole and is representative in character.
B) Definition:
(a) Crisp R. D:
The fundamental idea of sampling is that "of a small number of items or parts (called a
sample) are chosen at random from a large number of items or a whole (called a universe or
population) the sample will tend to have the same characteristics and to have them in
approximately the same proportion as the universe.”
16. C) Features of a Good Reliable/Sample:
1) Goal Oriented:
An ideal sample design should be goal oriented. It is a means and should be oriented to the
research objectives and fitted to the survey conditions.
2) Accurate Representative of the Universe:
A good sample should be an accurate representative of the universe from which it is taken.
There are different methods for selecting a sample. A representative sample can be chosen
by using any of these methods. The sample will truly be a representative only when it
represents all types of units of groups in the total population in fair proportions.
3) Random Selection:
An ideal sample should be selected at random. This means that any item in the group has a
full and equal chance of being selected and included in the sample. This makes the selected
sample truly representative in character.
17. C) Features of a Good Reliable/Sample:
4) Proportional:
An ideal sample should be proportional. It should be large enough to represent the universe
properly. If the total population is larger, the units selected should also be larger. The sample
size should be simple i.e. it should be capable of being understood and followed in the field
work.
5) Actual Information Provider:
An ideal sample should be designed so as to provide actual information required for the
study and also provide an adequate basis for the measurement of its own reliability.
6) Economical:
A good sample should be economical. The objectives of the survey should be achieved with
minimum cost and effort.
18. D) Determining Sample Size for Probability and Non-Probability Samples:
There are two broad categories of sample designs probability sample and non-probability
sample designs. Within these broad categories there are a number of sample designs. The
first discussion lays emphasis on probability sample designs.
Probability
Sampling
Non-
Probability
Sampling
19. D) Determining Sample Size for Probability and Non-Probability Samples:
a) Probability Samples:
In probability sampling methods, the sample units are selected at random. This means the
selection is arbitrary. Every member in the universe has equal chance of being selected as
the representative.
I) Types of Probability Sampling to Determine Sample Size:
Types of Probability
sampling
Random
Sampling
Systematic
Sampling
Stratified
Random
Sampling
Disproportionate
Random Sampling
Cluster
Sampling
Area
Sampling
Systematic
Random
Sampling
20. D) Determining Sample Size for Probability and Non-Probability Samples:
a) Probability Samples:
I) Types of Probability Sampling to Determine Sample Size:
1) Random Sampling:
A random sample gives every unit of the population a known and non-zero probability of
being selected. Since random sampling implies equal probability to every unit in the
population, it is necessary that the selection of the sample must be free from human
judgment.
2) Systematic Sampling:
In practice, the method followed in systematic sampling is simpler than that explained earlier
First, a sampling fraction is calculated. For instance, in the foregoing example, a sample of 50
out of 500 units was chosen.
3) Stratified Random Sampling:
A stratified random sample is one where the population is divided into mutually exclusive and
mutually exhaustive strata or sub-groups and then a simple random sample is selected within
each of the strata or sub-groups.
21. D) Determining Sample Size for Probability and Non-Probability Samples:
a) Probability Samples:
I) Types of Probability Sampling to Determine Sample Size:
4) Disproportionate Random Sampling:
The preceding section describes stratified sampling which involves the use of the uniform
sampling fraction over different strata of the population. At times, it may be preferable to use
variable sampling fractions, resulting in disproportionate stratified sampling.
5) Cluster Sampling:
In cluster sampling, the individual units are not selected as sample but are grouped together
and are selected group-wise for inclusion in the sample. Thus, groups are selected on random
basis as sample.
6) Area Sampling:
Area sampling is one aspect of cluster sampling method in which specific area is selected as a
sample in place of individual units out of the population. Area sampling is one probability
method of selection of sample.
7) Systematic Random Sampling:
In systematic random sampling method, the units of a population are first listed and the
sample is selected as per well defined system.
22. D) Determining Sample Size for Probability and Non-Probability Samples:
b) Non-Probability Samples:
Non-probability samples are those in which the participants are chosen or choose themselves
so that the chance of being selected is not known. The selection may be purposive. It may be
based on the convenience or the judgment of the researcher. The selection is deliberate and
not random.
I) Types of Non-Probability Sampling Methods to Determine Sample Size:
Types of
non
probability
sampling
Convenience
Sampling
Judgmental
Sampling
Quota
Sampling
23. D) Determining Sample Size for Probability and Non-Probability Samples:
b) Non-Probability Samples:
I) Types of Non-Probability Sampling Methods to Determine Sample Size:
1) Convenience Sampling:
In convenience sampling, the convenience of the researcher is given importance while
selecting the sample. Inclusion of units in the sample is decided by the researcher as per his
convenience. The items which are easily accessible or easily measurable are included in the
sample.
2) Judgmental Sampling:
In judgmental sampling method, inclusion of items in the sample is as per the expert
judgment given. An expert in the subject of research will be appointed to suggest the units
which are best representative of the total population for inclusion in the sample.
3) Quota Sampling:
Quota Sampling is similar to stratified random sampling. This non-probability sampling
method is commonly used in consumer research surveys. In this method, the universe is sub-
divided into identical groups on the basis of certain well defined norms like age, sex, income,
education and so on.