Creating a Research Design

1
Creating a Research Design

• Observation

2
Observation Research Defined
Observation research can be defined as the
systematic process of recording the behavioral
patterns of people, objects, and occurrences
without questioning or communicating with them.
3
Conditions for Using Observation
Three conditions must be met before observation
can be successfully used as a data collection
tool for marketing research.
observable or inferable
repetitive, frequent, or predictable
relatively short duration
4
Observational Situations
Situation Example
People watching people Observers stationed in
supermarkets watch consumers select frozen
Mexican dinners. The purpose is to see how
much comparison shopping people do at the
point of purchase. People watching
phenomena Observer stationed at an intersection
counts traffic moving in various
directions. Machines watching people Move or
videotape cameras record behavior as in
people-watching-people example. Machines
watching phenomena Traffic-counting machines
monitor traffic flow.
5
Dimensions Along with Observational Approaches
Vary

• Natural versus Contrived Situations
• A natural situation is just that, natural. The
  observer plays no role in the behavior of
  interest.
• A contrived situation is set up as a simulation
  by the observer.

6
Dimensions Along with Observational Approaches
Vary

• Open versus Disguised Observation
• In an open observation, the person knows that he
  or she is being observed. It is well known that
  the presence of an observer may have an influence
  on the phenomena being observed.
• A disguised observation is the process of
  monitoring people, objects, or occurrences that
  do not know they are being watched.

7
Dimensions Along With Observational Approaches
Vary

• Structured Verses Unstructured
• In the structured observation, the observer fills
  out a questionnaire-like form on each person
  observed.
• In the totally unstructured observation, the
  observer simply makes notes on the behavior being
  observed.

8
Dimensions Along with Observational Approaches
Vary

• Human versus Machine Observers
• In some situations it may be possible and even
  desirable to replace human observers with
  machines.
• In certain situations machines may do the job
  less expensively, more accurately, or more
  readily.

9
Dimensions Along with Observational Approaches
Vary

• Direct versus Indirect Observation
• Most of the observation done in marketing
  research is direct observation that is, directly
  observing current behavior.
• Past behavior can only be observed through
  indirect observation.

10
Advantages and Disadvantages of Observation
Research

• Advantages
• Observation research provides the researcher the
  opportunity to watch what people actually do
  rather than relying on reports of what they do.
• This approach can avoid much of the biasing
  factors caused by the interviewer and question
  structure associated with the survey approach.

11
Advantages and Disadvantages of Observation
Research

• Disadvantages
• Only behavior and physical personal
  characteristics can usually be examined. The
  researcher does not learn about motives,
  attitudes, intentions, or feelings.
• Observation research can be time consuming and
  costly if the observed behavior occurs rather
  infrequently.

12
Human Observation

• Mystery Shoppers
• Are used to gather observational data about a
  store and to collect data about customer/
  employee interactions.
• One-Way Mirror Observation
• The practice of watching unseen from behind a
  one-way mirror.

13
Human Observation

• Shopper Patterns and Behavior
• Shopping patterns are used by retailers to trace
  the flow of shoppers through a store.
• Shopper behavior research involves observing, or
  perhaps filming and then watching the film, of
  shoppers or consumers in a variety of shopping
  settings.
• The direct observation and/or films are analyzed
  to try to understand customer behavioral
  tendencies.

14
Human Observation

• Content Analysis
• Is an observation technique used to analyze
  written material (usually advertising copy) into
  meaningful units using carefully applied rules.
• Humanistic Inquiry
• A research method in which the researcher is
  immersed in the system or group under study.

15
Human Observation

• Audits
• Audits are another category of human observation
  research. An audit is examination and
  verification of the sale of a product.
• Audits generally fall into two categories
• Retail audits that measure sales to final
  customers.
• Wholesale audits that determine the amount of
  product movement from warehouses to retailers.

16
Machine Observation

• Traffic Counters
• Machines used to measure vehicular flow over a
  particular stretch of roadway.
• Outdoor advertisers rely on traffic counts to
  determine the number of exposures per day to a
  specific billboard.
• Retailers use the information to ascertain where
  to place a particular type of store.

17
Machine Observation

• Physiological Measurement
• EEG
• The electroencephalograph (EEG) is a machine that
  measures rhythmic fluctuations in the electric
  potential in the brain.
• Researchers claim that EEG measures can be used
  to assess viewers attention to an advertisement
  at a specific point in time, the intensity of the
  emotional reactions elicited by specific aspects
  of the ad, and their comprehension and attention
  to the ad.

18
Machine Observation

• Physiological Measurement
• GSR
• The galvanic skin response (GSR), also known as
  the electrodermal response, measures changes in
  the electric resistance of the skin associated
  with activation responses.
• Marketing research firms use the GSR to evaluate
  viewers interest levels during commercials.

19
Machine Observation

• Physiological Measurement
• Pupilometer
• The pupilometer measures changes in pupil
  dilation.
• The basic assumption is that increased pupil size
  reflects positive attitudes, interest, and
  arousal in an advertisement.
• Voice Pitch Analysis
• Voice pitch analysis examines changes in the
  relative vibration frequency of the human voice
  to measure emotion.

20
Machine Observation

• Opinion and Behavior Measurement
• People Reader
• The people reader is a machine. The machine looks
  like a lamp and is designed so that when
  respondents sit in front of it they are not aware
  it is simultaneously recording both the reading
  material and their eyes.
• The people reader is used to document information
  concerning reading habits and the results of
  different size ads in terms of stopping power and
  brand-name recall.

21
Machine Observation

• Opinion and Behavior Measurement
• RAMS
• RAMS, or Rapid Analysis Measurement System, is a
  hand-held device with a dial in the center.
  Respondents turn the dial to the right when they
  are feeling more favorable toward a subject and
  vice versa.

22
Machine Observation

• Opinion and Behavior Measurement
• People Meter
• The people meter is a microwave computerized
  rating system that transmits demographic
  information overnight to measure national t.v.
  audiences.
• The people meter provides information on what
  t.v. shows are being watched, the number of
  households watching, and which family members are
  watching.

23
Scanner Based Research

• Scanner-Based Research
• BehaviorScan
• A single-source system that maintains a 3,000
  household panel to record consumer purchases
  based upon manipulation of the marketing mix.
• InfoScan
• A scanner-based tracking service for consumer
  packaged goods.

24
Scanner Based Research

• Scanner-Based Research (continued)
• IRIs Software
• IRIs Apollo Space Management software with
  advanced digitized imaging can produce
  photo-quality schematics of optimal shelf
  allocations of products.
• Other Important IRI Software Programs Include
• DataServe Analyzer, DataServe Partners, DataServe
  Targeter, and QuickStart.

25
The Future of Scanning
The next generation of scanners, to be known as
Scanner Plus, will have abilities far beyond
those of todays machines. These scanners will be
able to communicate with personal computers in
homes. One function could be to analyze an
individual households consumption based on its
prior purchase patterns and offer menu
projections or product use suggestions with an
associated shopping list.
26
Creating a Research Design

• Experimentation

27
What is an Experiment?
An experiment is a research approach where one
variable is manipulated and the effect on another
variable is observed.
28
Manipulation Examples
29
Manipulation Examples

• You order your semester textbooks for college
  classes during the first week of school from an
  online bookstore. Because you have many
  assignments to complete in the coming week, you
  request overnight delivery at an additional cost
  of 20.00. The books arrived three days later.

• You order several outfits, a pair of sandals, a
  new purse and a pair of expensive sunglasses
  online from a well-known company. Since you will
  travel on vacation to the beach that weekend, you
  choose to pay an additional 25 for second-day
  delivery. The package arrives on the third day
  just as you were packing to leave. One of the
  outfits was not the one you ordered and the
  sunglasses were not included in the package.

30
Demonstrating Causation

• What is causal research?
• Demonstrating Causation
• Concomitant Variation
• Appropriate Time Order of Occurrence
• Elimination of Other Possible Factors

31
The Experimental Setting - Laboratory or Field
Laboratory Experiments
Field Experiments
Tests conducted outside the laboratory in an
actual market environment.
Experiments conducted in a controlled setting.
32
Experimental Validity

• Validity
• The validity of a measure refers to the extent to
  which the measure is free from both systematic
  and random error.
• In addition to the general concept of validity,
  in experimentation we also are interested to two
  specific kinds of validity
• Internal validity
• External validity

33
Internal and External Validity
Internal Validity
External Validity
The extent to which causal relationships measured
in an experiment can be generalized to outside
people, settings, and times.
The extent to which competing explanations for
the experimental results can be avoided.
34
Experimental Notation
X Is used to indicate the exposure of an
individual or a group to an experimental
treatment. The experimental treatment is the
factor whose effects we want to measure and
compare. O (for observation) is used to refer to
the process of taking measurement on the test
units. Test units are individuals or groups of
individuals or entities (retail stores) whose
response to the experimental treatment is being
tested.
35
Experimental Notation
Different time periods are represented by the
horizontal arrangement of the Xs and Os. For
example,
36
Threats to Experimental Validity

• History
• Maturation
• Instrument Variation
• Selection Bias
• Mortality
• Testing Effect
• Regression to the Mean

37
Experimentation Summary of Basic Issues

• Experimental Design
• In an experimental design, the researcher has
  control over one or more independent variables
  and manipulates one or more independent
  variables.
• Nonexperimental Designs
• Involve no manipulation and typically are
  referred to as ex post facto (after the fact)
  research.

38
Experimentation Summary of Basic Issues
An experimental design includes four factors
The treatment or experimental variable to be
manipulated
The subjects to participate in the experiment
A dependent variable to measure
Some plan or procedure for dealing with
extraneous causal factors
39
Experimentation Summary of Basic Issues

• Experimental Effects
• The term experimental effect refers to the effect
  of the treatment variables on the dependent
  variable.
• The goal is to determine the effect of each
  treatment condition (level of treatment variable)
  on the dependent variable.

40
Experimentation Summary of Basic Issues
Approaches used to control extraneous factors
Randomization - involves randomly assigning
subjects to treatment conditions
Physical control - involves somehow holding the
value or level of the extraneous variable constant
Design Control - implementing specific types of
experimental designs
Statistical control - certain statistical
procedures can be used to adjust for the effects
of confounding variables
41
Why Experiments are Not Used More Often

• High Cost
• Security Issues
• Implementation Problems

42
Selected Experimental Designs

• The text gives examples of
• preexperimental
• true experimental, and
• quasi-experimental designs.

43
Test Market Defined

• A common form of experimentation used by
  marketing research practitioners is test
  marketing. The term test market is used by
  marketing researchers rather loosely to refer to
  any research that
• Involves testing a new product or any change in
  an existing marketing strategy in a single
  market, a group of markets, or a region of the
  country.
• Involves the use of experimental procedures.

44
Test Market Objectives

• Test market studies are designed to provide
  information in regard to the following issues
• market share
• effect that the new product will have on other
  products (cannibalization rate)
• consumer characteristics
• behavior of competitors

45
Costs of Test Marketing
Direct Costs

• Point-of-purchase materials
• Customized research information and associated
  data analysis
• Higher trade allowances to obtain distribution
• Coupons and sampling

• Production of commercials
• Payments to advertising agency for services
• Media time at a higher rate because of low volume
• Syndicated research information

46
Costs of Test Marketing
Indirect Costs
Diversion of sales activity from existing products

• Cost of management time spent on the test market.
• Cost of letting your competitors know what you
  are doing, allowing them to develop a better
  strategy or beat you to the national market.

• Possible negative trade reactions to your
  products if you develop a reputation of not doing
  well.
• Possible negative impact of a test market failure
  on other products with the same family brand.

47
Factors to Consider Before Conducting a Test
Market

• weigh the cost and risk of failure against the
  probability of success and associated profits.
• the likelihood and speed with which your
  competitors can copy your product and introduce
  it on a national basis also must be considered.

48
Factors to Consider Before Conducting a Test
Market

• consider the investment required to produce the
  product for the test market versus the investment
  required to produce the product in the quantities
  necessary for a national rollout
• The final consideration relates to the damage
  that an unsuccessful new product launch can
  inflict on a companys reputation.

49
Steps in a Test Market
Define the Objective
Select Markets for the Test