Descriptive & Causal

Research Designs
Descriptive Research Design
• Surveys (most common method of collecting primary data)
– method of obtaining information based on the questioning of
respondents
– verbally, in writing or via computer
- Questioning is structured

• Observation
Classification of Survey Methods
Some Outgoing Envelope
Outgoing envelope: size, color, return address
Blank
Blank
Decisions Postage
Method of addressing
Blank
Blank
Related to Cover Letter Blank

the Mail Sponsorship

Personalization
Signature
Postscript
Interview Type of appeal
Questionnaire
Blank
Blank
Package Length Layout
Content Color
Size Format
Reproduction Respondent anonymity
Return Envelope Blank
Type of envelope Blank
Postage Blank
Incentives Blank
Monetary versus nonmonetary Blank
Prepaid versus promised amount Blank
Classification of Survey Methods
 Criteria for Evaluating Survey Methods (1 of 6)
TASK FACTORS (tasks that have to be performed to collect data)

1. Diversity of Questions and Flexibility of Data Collection

• The flexibility of data collection is determined primarily by the extent to which the respondent
can interact with the interviewer and the survey questionnaire. The diversity of questions that
can be asked in a survey depends upon the degree of interaction the respondent has with the
interviewer and the questionnaire, as well as the ability to actually see the questions.
2. Use of Physical Stimuli
• The ability to use physical stimuli such as the product, a product prototype, commercials, or
promotional displays during the interview.
Criteria for Evaluating Survey Methods (2 of 6)

3. Sample Control
• Sample control is the ability of the
survey mode to reach the units specified
in the sample effectively and efficiently.
4. Quantity of Data
• The ability to collect large amounts of
data.
5. Response Rate
• Survey response rate is broadly defined
as the percentage of the total attempted
interviews that are completed.
 Criteria for Evaluating Survey Methods (3 of 6)

SITUATIONAL FACTORS (control of externalities in the data collection process)

1. Control of the Data Collection Environment

• The degree of control a researcher has over the environment in which the
respondent answers the questionnaire.
2. Control of Field Force
• The ability to control the interviewers and supervisors involved in data
collection.
 Criteria for Evaluating Survey Methods (4 of
6)

3. Potential for Interviewer Bias

• The extent of the interviewer's role determines the potential for bias
4. Speed
• The total time taken for administering the survey to the entire sample
5. Cost
• The total cost of administering the survey and collecting the data
 Criteria for Evaluating Survey Methods (5 of 6)
RESPONDENT FACTORS (factors pertaining to survey respondents)

1. Perceived Anonymity
• Perceived anonymity refers to the respondents' perceptions that their
identities will not be discerned by the interviewer or the researcher.
2. Social Desirability
• Social desirability is the tendency of the respondents to give answers
that are socially acceptable, whether or not they are true.
 Criteria for Evaluating Survey Methods (6 of 6)
3. Low Incidence Rate
• Incidence rate refers to rate of occurrence of persons eligible to
participate in the study a given sample size requirement
• Incidence Rate determines How many contacts need to be screened for
• Generally in cases of Niche market or highly targeted market
• E.g. Maybelline study requires female respondents, between 18-27, who
have used foundation at least twice in the last week
4. Respondent Control
• Methods that allow respondents control over the interviewing process
will solicit greater cooperation and are therefore desirable.
• When to answer/ flexibility of answering in parts
 Comparative Evaluation of Survey Methods (1 of 3)
Telephone In-Home Mall-Intercept Mail
Criteria CATI Interviews Interviews CAPI Surveys Mail Panels Email Internet Mobile
Task Factors blank blank blank blank blank blank blank blank blank

Diversity of Low to High High Moderate Moderate Moderate Moderate Moderate Low
questions and moderate to high to high
flexibility

Use of physical Low Moderate to High High Moderate Moderate Low Moderate Low to
stimuli high to high moderate

Sample control Moderate to Potentially Moderate Moderate Low Moderate to Low Low to Low to
high high high moderate moderate

Quantity of data Low High Moderate Moderate Moderate High Moderate Moderate Low

Response rate Moderate High High High Low High Low Very low Moderate
 Comparative Evaluation of Survey Methods (2 of 3)
Telephone In-Home Mall-Intercept Mail Mail
Criteria CATI Interviews Interviews CAPI Surveys Panels Email Internet Mobile
Situational blank blank blank blank blank blank blank blank blank
Factors
Control of data Moderate Moderate to High High Low Low Low Low Low
collection high
environment

Control of field force Moderate Low Moderate Moderate High High High High High

Potential for Moderate High High Low None None None None None
interviewer bias

Speed High Moderate Moderate to Moderate Low Low to High Very high Very high
high to high moderate

Cost Moderate High Moderate to Moderate Low Low to Low Low Low to
high to high moderate moderate
 Comparative Evaluation of Survey Methods (3 of 3)
Telephone In-Home Mall-Intercept Mail Mail
Criteria CATI Interviews Interviews CAPI Surveys Panels Email Internet Mobile
Respondent blank blank blank blank blank blank blank blank blank
Factors
Perceived anonymity of Moderate Low Low Low High High Moderate High Moderate
the respondent

Social desirability Moderate High High Moderate Low Low Moderate Low Low
to high

Obtaining sensitive High Low Low Low to High Moderate Moderate High Moderate
information moderate to high to high

Low incidence rate High Low Low Low Moderate Moderate Moderate High High

Respondent control Low to Low Low Low High High High Moderate High
moderate to high
OBSERVATION
Mar
keti
ng
Rese
arch
:
The OWN SALES SUBSIDIARY
Japa  Increased market share

nese 
Sales of its digital cameras in 2000s
By 2018, operating in 125 countries
Way  80% sales generated outside Japan
Observation Methods - Structured Versus
Unstructured Observation
• For structured observation, the researcher specifies in detail what is
to be observed and how the measurements are to be recorded, e.g.,
an auditor performing inventory analysis in a store.
• In unstructured observation, the observer monitors all aspects of the
phenomenon that seem relevant to the problem at hand, e.g.,
observing children playing with new toys.
Observation Methods - Disguised Versus
Undisguised Observation
• In disguised observation, the respondents are unaware that they are
being observed. Disguise may be accomplished by using one-way
mirrors, hidden cameras, or inconspicuous mechanical devices.
Observers may be disguised as shoppers or sales clerks.
• In undisguised observation, the respondents are aware that they are
under observation.
Observation Methods - Natural Versus
Contrived Observation
• Natural observation involves observing behavior as it takes places in
the environment. For example, one could observe the behavior of
respondents eating fast food at Burger King.
• In contrived observation, respondents' behavior is observed in an
artificial environment, such as a test kitchen.

Big Boss?!?

Canon?!?
A Classification of Observation Methods
Observation Methods: Personal Observation
• A researcher observes actual behavior as it occurs.
• The observer does not attempt to manipulate the phenomenon
being observed but merely records what takes place.
• For example, a researcher might record traffic counts and observe
traffic flows in a department store.
 Observation Methods: Mechanical Observation
Rather than human observers, record the phenomenon being observed

Do not require respondents' direct participation.

• The AC Nielsen’s people meter – attached to a television set to record what’s playing & who’s watching
• Turnstiles that record the number of people entering or leaving a building
• On-site cameras (still, motion picture, or video) – used by retailers
• Optical scanners in supermarkets
Do require respondent involvement.
• Eye-tracking monitors – records gaze movements of the eye – assess advertising effectiveness
• Pupilometers – records change in the diameter of the pupils - cognitive thinking
• Voice pitch analyzers – record change in pitch to gauge emotional reaction
• Devices measuring response latency – Response time is directly proportionate to uncertainty
Observation Methods: Audit
• The researcher collects data by examining physical records or
performing inventory analysis.
• Data are collected personally by the researcher.
• The data are based upon counts, usually of physical objects.
• E.g. Retail and wholesale audits conducted by marketing research
suppliers
Observation Methods: Content Analysis
• The objective, systematic, and quantitative description of the
manifest content of a communication/ advertisement.
• The unit of analysis may be words, characters (individuals or objects),
themes (propositions), space and time measures (length or duration
of the message), or topics (subject of the message).
• E.g. the frequency of appearance of blacks, women and members of
minority groups in print media between 2011-2019 can be studied
using content analysis
Observation Methods: Trace Analysis
Data collection is based on physical traces, or evidence, of past behavior.

 The number of different fingerprints on a page was used to gauge the readership
of various advertisements in a magazine.
 The position of the radio dials in cars brought in for service was used to estimate
share of listening audience of various radio stations.
 The magazines people donated to charity were used to determine people's
favorite magazines.
 Internet visitors leave traces which can be analyzed to examine browsing and
usage behavior by using cookies.
A Comparative Evaluation of Observation Methods

Personal Mechanical Content Trace

Criteria Observation Observation Audit Analysis Analysis
Degree of structure Low Low to high High High Medium

Degree of disguise Medium Low to high Low High High

Ability to observe High Low to high High Medium Low

in natural setting

Observation bias High Low Low Medium Medium

Analysis bias High Low to medium Low Low Medium

General remarks Most flexible Can be intrusive Expensive Limited to Method of

communications last resort
Relative Advantages of Observation
• They permit measurement of actual behavior rather than reports of
intended or preferred behavior.
• There is no reporting bias, and potential bias caused by the
interviewer and the interviewing process is eliminated or reduced.
• Certain types of data can be collected only by observation.
• If the observed phenomenon occurs frequently or is of short duration,
observational methods may be cheaper and faster than survey
methods.
Relative Disadvantages of Observation
• The reasons for the observed behavior may not be determined since little is
known about the underlying motives, beliefs, attitudes, and preferences.
• Selective perception (bias in the researcher's perception) can bias the data.
• Observational data are often time-consuming and expensive, and it is difficult to
observe certain forms of behavior.
• In some cases, the use of observational methods may be unethical, as in
observing people without their knowledge or consent.

It is best to view observation as a complement to survey methods, rather than

as being in competition with them.
Ethnographic Research
• Study of human behaviour in its natural context
• Involves observation of behaviour and setting, along with depth
interviews
• Sometimes Audio and video recordings are also obtained
Causal Research Design : Experimentation
Concept of Causality
A statement such as "X causes Y" will have the following
meaning to an ordinary person and to a scientist.

Ordinary Meaning Scientific Meaning

X is the only cause of Y.
X must always lead to Y (X is a
deterministic cause of Y).
It is possible to prove that X is a
cause of Y.
X is only one of a number of possible
causes of Y.
The occurrence of X makes the
occurrence of Y more probable
(X is a probabilistic cause of Y).
We can never prove that X is a cause of Y.
At best, we can infer that X is a cause of
Y.
Conditions for Causality
1. Concomitant variation is the extent to which a cause, X, and an
effect, Y, occur together or vary together in the way predicted by
the hypothesis under consideration.
Y – purchase of fashion clothing
X – Education
 Evidence of Concomitant Variation between
Purchase of Fashion Clothing and Education

Purchase of Fashion Clothing – Y

Education – X High Low Total

High 363 (73%) 137 (27%) 500 (100%)

Low 322 (64%) 178 (36%) 500 (100%)

Respondents with high education are more likely to purchase more fashion clothing
 Purchase of Fashion Clothing By
Income and Education

Low Income High Income

Purchase Purchase

Education High Low Total Education High Low Total

High 122 (61%) 78 (39%) 200 (100%) High 241 (80%) 59 (20%) 300 (100%)

Low 171 (57%) 129 (43%) 300 (100%) Low 151 (76%) 49 (24%) 200 (100%)
Conditions for Causality
1. Concomitant variation is the extent to which a cause, X, and an
effect, Y, occur together or vary together in the way predicted by
the hypothesis under consideration.
2. The time order of occurrence condition states that the causing
event must occur either before or simultaneously with the effect; it
cannot occur afterwards.
3. The absence of other possible causal factors means that the factor
or variable being investigated should be the only possible causal
explanation. All other factors are constant or controlled.
 Definitions and Concepts
• Independent variables are variables or alternatives that are manipulated and whose
effects are measured and compared, e.g., price levels.
• Test units are individuals, organizations, or other entities whose response to the
independent variables or treatments is being examined, e.g., consumers or stores.
• Dependent variables are the variables which measure the effect of the independent
variables on the test units, e.g., sales, profits, and market shares.
• Extraneous variables are all variables other than the independent variables that
affect the response of the test units, e.g., store size, store location, and competitive
effort.
 Experimental Design
An experimental design is a set of procedures specifying:

 the test units and how these units are to be divided into homogeneous
subsamples,
 what independent variables or treatments are to be manipulated,
 what dependent variables are to be measured; and
 how the extraneous variables are to be controlled.

Symbols:
X = the exposure of a group to independent variable/treatments
O = process of observation/measurement of dependent variable
R = the random assignment of test units to separate treatments
Validity in Experimentation – Internal & External
• Internal validity (measures the accuracy of the experiment) refers to
whether the manipulation of the independent variables or treatments
actually caused the observed effects on the dependent variables.
Control of extraneous variables is a necessary condition for
establishing internal validity.

How well a study is conducted.

• External validity refers to whether the cause-and-effect relationships

found in the experiment can be generalized. To what populations,
settings, times, independent variables, and dependent variables can
the results be projected?

How applicable the findings are to the real world.

Classification of Extraneous Variables (1 of 2)
• History (H)refers to specific events that are external to the experiment but occur
at the same time as the experiment.
• Maturation (MA) refers to changes in the test units themselves that occur with
the passage of time. Older/ experienced/ bored/ tired/ unintrested
• Testing effects are caused by the process of experimentation. Typically, these are
the effects on the experiment of taking a measure on the dependent variable
before and after the presentation of the treatment.
• The main testing effect (MT) occurs when a prior observation affects a latter
observation.
• In the interactive testing effect (IT), a prior measurement affects the test unit's
response to the independent variable.
Classification of Extraneous Variables (2 of 2)
• Instrumentation (I) refers to changes in the measuring instrument, in the
observers, or in the scores themselves. E.g. Change of questionnaire
• Statistical regression effects (SR) occur when test units with extreme
scores move closer to the average score during the course of the
experiment – confounding effect on the experimental results
• Selection bias (SB) refers to the improper assignment of test units to
treatment conditions. E.g. merchandising display experiment in two
different stores
• Mortality (MO) refers to the loss of test units while the experiment is in
progress. Respondents’ beliefs & attitudes change over time
Controlling Extraneous Variables
• Randomization refers to the random assignment of test units to experimental
groups by using random numbers. Treatment conditions are also randomly assigned
to experimental groups.
As a result of random assignment, extraneous factors can be represented equally

• Matching involves comparing test units on a set of key background variables before
assigning them to the treatment conditions. Takes care of the selection bias

• Statistical control involves measuring the extraneous variables and adjusting for
their effects through statistical analysis. (ANCOVA)

• Design control involves the use of experiments designed to control specific

extraneous variables.
A Classification of Experimental Designs
 A Classification of Experimental Designs
• Pre-experimental designs do not employ randomization procedures to control
for extraneous factors: the one-shot case study, the one-group pretest-posttest
design, and the static-group.
• In true experimental designs, the researcher can randomly assign test units to
experimental groups and treatments to experimental groups: the pretest-
posttest control group design, the posttest-only control group design, and the
Solomon four-group design.
 A Classification of Experimental Designs
• Quasi-experimental designs result when the researcher is unable to achieve full
manipulation of scheduling or allocation of treatments to test units but can still
apply part of the apparatus of true experimentation: time series and multiple
time series designs.
• A statistical design is a series of basic experiments that allows for statistical
control and analysis of external variables: randomized block design, Latin square
design, and factorial designs.
 A Classification of Experimental Designs

Source of Invalidity
blank History Maturation Testing Internal Regression Selection Mortality External
Instrumentation Interaction
of testing
and X

Design blank blank blank blank blank blank blank blank

Preexperimental designs: blank blank blank blank blank blank blank blank
One-shot case study − − blank blank blank − − blank
X O blank blank blank blank blank blank blank blank
One-group pretest–posttest − − − − ? blank blank −
design
O X O blank blank blank blank blank blank blank blank
Static group comparison + ? + + + − −
X O blank blank blank blank blank blank blank blank
True experimental designs: blank blank blank blank blank blank blank blank

Pretest–posttest control + + + + + + + −
R O X O blank blank blank blank blank blank blank blank
R O O blank blank blank blank blank blank blank blank
 A Classification of Experimental Designs

blank History Maturation Testing Internal Regression Selection Mortality External

Instrumentation Interaction
of testing
and X
Posttest-only control group + + + + + + + +
design

R X O blank blank blank blank blank blank blank blank

R O blank blank blank blank blank blank blank blank

Quasi-experimental designs: blank blank blank blank blank blank blank blank

Time series − + + ? + + + −

O O O X O O O blank blank blank blank blank blank blank blank

Multiple time series + + + + + + + −

O O O X O O O blank blank blank blank blank blank blank blank

O O O O O O blank blank blank blank blank blank blank blank

Note: A + indicates a relative advantage, whereas a − indicates a relative disadvantage.

 One-Shot Case Study
X 01
• A single group of test units is exposed to a treatment X
• A single measurement on the dependent variable is taken (01)
• There is no random assignment of test units
• The one-shot case study is more appropriate for exploratory than for conclusive research
• Example: To measure the effectiveness of a test commercial for Wal-Mart

Unaided: “Do you

Telephone interviews recall seeing a
are conducted with a Aided: “Do you recall seeing a
commercial for a commercial for Wal-Mart last
national sample of The dependent department store last
respondents who night?
 variables (Os) are  night?” 
report watching a unaided and aided The results of aided and unaided
particular TV program recall If they recall, details recall are compared to norm
(contains Walmart ad) about commercial scores to develop an index for
the previous night content are solicited interpreting the scores
 One-Group Pretest-Posttest Design
01 X 02
• A group of test units is measured twice.
• There is no control group.
• The treatment effect is computed as 02 – 01.
• The validity of this conclusion is questionable since extraneous variables are
largely uncontrolled.

The respondents are

Respondents are They watch a TV again administered a
interviewed about program containing personal interview to
 
their attitude towards Wal-Mart commercial measure attitudes
Wal-Mart store (O1) (X) towards the store
(O2)
Static Group Design
Attitude towards Wal-
EG: X 01 Two groups of Only EG watches a TV
Mart store of EG &
respondents would program containing
  CG respondents
CG: 02 be recruited based on Wal-Mart commercial
would be measured
convenience (X)
(O1 – O2)

• A two-group experimental design.

• The experimental group (EG) is exposed to the treatment, and the control group
(CG) is not.
• Measurements on both groups are made only after the treatment.
• Test units are not assigned at random.
• The treatment effect would be measured as 01 - 02.
True Experimental: Pretest-Posttest Control Group Design
EG: R 01 X 02
CG: R 03 04
• Test units are randomly assigned to either the experimental or the control group.
• A pretreatment measure is taken on each group.
• The treatment effect (TE) is measured as: (02 − 01) − (04 − 03).
• Selection bias is eliminated by randomization.
• The other extraneous effects are controlled as follows:
02 – 01= TE + H + MA + MT + IT + I + SR + MO
04 – 03= H + MA + MT + I + SR + MO = EV (Extraneous Variables)
• The experimental result is obtained by:
(02 − 01) − (04 − 03) = TE + IT
• Interactive testing effect is not controlled.
True Experimental: Posttest-Only Control Group Design

EG : R X 01
CG : R 02

• The treatment effect is obtained by:

TE = 01 – 02
• Except for pre-measurement, the implementation of this design is
very similar to that of the pretest-posttest control group design.
 Quasi-Experimental Designs: Time Series Design

• Researcher can control: when measurements are taken & on whom they are taken
• Researcher can’t control: scheduling of the treatments (might be naturally occurring)
& randomization of test units to treatments (cant control who will be exposed to the
treatment)

01 02 03 04 05 X 06 07 08 09 010

• Example: The respondents’ purchases before, during and after the campaign are
examined to determine whether the test commercial has a short term, long term or no
effect.
 Multiple Time Series Design
EG : 01 02 03 04 05 X 06 07 08 09 010
CG : 01 02 03 04 05 06 07 08 09 010

• If the control group is carefully selected, this design can be an improvement over
the simple time series experiment
• Can test the treatment effect twice: against the pretreatment measurements in
the experimental group and against the control group
• Example: The test commercial would be shown in only a few of the test cities.
Respondents in these cities would comprise the experimental group.
Respondents in cities where the commercial was not shown would constitute the
control group
Statistical Designs
Statistical designs consist of a series of basic experiments that allow for
statistical control and analysis of external variables and offer the following
advantages:
• The effects of more than one independent variable can be measured.
• Specific extraneous variables can be statistically controlled.

The most common statistical designs are:

the randomized block design (RBD),
the Latin square design(LSD)
the factorial design(FD)
 Randomized Block Design
• Is useful when there is only one major external variable, such as store patronage, that might
influence the dependent variable
Test units are
Blank Blank
Blocking Variable Blank Treatment Groups Blank
randomly
assigned to
Impact of Block No. Store Patronage Commercial A Commercial B Commercial C
Treatment
humor on the 1 Heavy A B C groups
effectiveness
of advertising 2 Medium A B C

3 Low A B C

4 None A B C

Note: A, B, and C denote three test commercials, which have, respectively, no humor,
some humor, and high humor.

• The test units are blocked, or grouped, on the basis of the external variable
• By blocking, the researcher ensures that the various experimental and control groups are matched
closely on the external variable
• Finally, compare the results within each block
 Latin Square Design
• Allows the researcher to statistically control two non-interacting external variables as well as to
manipulate the independent variable.

Blank Blank Variable 2  Interest in the Store

Blocking Blank

Blocking Variable 1  Store Patronage High Medium Low

High B A C

Medium C B A

Low and none A C B

Note: A, B, and C denote the three test commercials, which have, respectively, no
humor, some humor, and high humor.

• 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
• RULE: Each level of independent variable should appear only once in each row and each column
Factorial Design (FD)
• Two or more independent variables are investigated simultaneously
• X and Y are 2 independent variables

• 2 x 2 design
(Each variable has 2 levels)

• Unlike RBD and LSD, FD allow 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
• For example, Mr.A’s favorite drink might be coffee. His favorite temperature level – cold;
But he might not prefer cold coffee, leading to interaction
 Factorial Design (FD)
• Example: Whether the new drug reduces migraine attacks?

1 independent variable 
 Factorial Design (FD)
• 2-way interaction

2-way
interaction

Main effects Interaction effect

2 independent variables  6 conditions

 Factorial Design (FD) Higher-order interaction

3 independent variables  12 conditions

 Factorial Design (FD)Higher-order interaction
3-way interaction
2-way interaction

Main effect

3 independent variables  12 conditions

Limitations of Experimentation
• Experiments can be time consuming, particularly if the researcher is interested in
measuring the long-term effects.
• Experiments are often expensive. The requirements of experimental group,
control group, and multiple measurements significantly add to the cost of
research.
• Experiments can be difficult to administer. It may be impossible to control for the
effects of the extraneous variables, particularly in a field environment.
• Competitors may deliberately contaminate the results of a field experiment.