MKTG 470 Experiments: Quiz2

Experimental Design

is a procedure for devising an experimental setting such that a change in a dependent variable may be attributed solely to the change in an independent variable.

Internal Validity:

is concerned with the extent to which the change in the dependent variable was actually due to the independent variable

Experiment

is defined as manipulating an independent variable to see how it affects a dependent variable, while also controlling the effects of aditional extraneous variables.

Factorial design

A design in which all levels of each independent variable are combined with all levels of the other independent variables. Allows investigation of the separate main effects and interactions of two or more independent variables.

ADVANTAGES OF WITHIN SUBJECTS DESIGNS

-ADVANTAGES OF WITHIN SUBJECTS DESIGNS:Lower error variance due to individual person differences Greatly reduced sample size requirements -DISADVANTAGES: Carry-over effects

Dependent variables come in three flavors

-Consumer response:attitude change, knowledge change, satisfaction -Marketing Performance indicators:attitude change, knowledge change, satisfaction. -Firm Performance: Tobin Q, Share Price.

What are the minimum requirements of a typical experiment?

-Independent Variables -Dependent Variables -Extraneous Variables

Laboratory Experiment

-Marketers have more control of experiment conditions. -Tends to have greater internal validity.

Field Experiment (Test Marketing)

-Marketers tend to have less control of experimental conditions. -Tends to have a greater external validity.

Three key factors for appropriate test market

-Representativeness .Demographics .Psychographics/Lifestyle -Competitive Characteristics -Ability to Control Distribution and Promotion .Partnerships with retailers/distributors .Sufficient media channels available .Access to Point of Sales data (retail partners)

Laboratory Experiment Risk & Problems

-Scientists cannot exercise laboratory control in natural world settings or randomly assign subjects. -Digital marketing allow thousands of IV combinations at little or no cost.

Threats to Validity of Field Experiment

-Selection-maturation threat:The two groups naturally evolve in a different way with respect to the DV. -Selection-mortality threat:There is a different, nonrandom dropout between pre and posttest of two different test markets. -Resentful demoralization: A test market not receiving treatment backlashes against the marketer.

How is an ecological fallacy a potential problem when interpreting marketing experiments?

...

How do you calculate effect size

...Experimental Effect = E = (O2 - O1) - (O4 - O3)

Why do marketing experiments usually NOT have a traditional control group?

...In marketing, we may not have a control group in the strict sense, but we can just as easily compare our experimental group to any other experimental group (for example, showing Group 1 a new advertisement, and showing Group 2 a differen new advertisement)

Why is random assignment so important?

...It reduced Extraneous variables

How is it different from a post-only design?

...Without a pre-test, we assume the two groups started at the same baseline... in other words:

Standard test market

: is one in which the firm tests the product and / or marketing mix variables through the company's normal distribution channels.

Pros for Test Marketing

Allows more accurate method of forecasting future sales Allows firms the opportunity to pretest marketing mix variables Compared to laboratory experiments, much higher on external validity

Random Assignment

Assigning participants to experimental and control conditions by chance, thus minimizing preexisting differences between those assigned to the different groups

Dependent Variables:

Brand awareness, Brand knowledge, attitude change, attention to add.

Experiments are used to study____ research questions

Causal

Why do direct marketers often use a fractional factorial design?

Commonly used in marketing research especially in direct marketing, software helps us calculate the correct design/

Running Example

Did our banner advertisement cause more people to sign up for our e-mail newsletter?

Cons Test Marketing

Does not yield infallible results... "can't step into the same river twice" Are expensive Exposes the new product to competitors Takes time to conduct Competitive Meddling: when competitor wants to screw you.

Ecological fallacy

Erroneously drawing conclusions about individuals solely from the observation of groups

Experimentl with more the one IV

Factors

Difference between full factorial and fractional factorial design?

Full factorial lists all the possible combinations while fractional only computes partial interesting combinations.

Warnings about Marketing experiments

Impractical in many cases Can be somewhat costly & time consuming, but tools like Qualtrics makes it more cost effective in some cases.

Most experiments will have:

Independent variable levels randomly assigned to different experimental groups Random assignment is a fair way to ensure that the two different groups are likely to be the same. However, it does not guarantee that the two groups are equivalent. A control group This is in order to compare the effect of the independent variable on the dependent variable to a group that did not receive any treatment In marketing we often don't have a classic control group (like a placebo pill). Instead, our control group is some meaningful baseline, like "currently used advertising campaign." Experiments must have some way to compare the different levels of independent variables treatment on the dependent variable.

Test Marketing2:

Is the phrase commonly applied to an experiment, study, or test that is conducted in a field setting. Essentially a Field Expeiment

What is the fundamental characteristic of experiments that makes it uniquely capable of answering 'why' questions?

It determines a causal relationship between two or more variables. If X then Y. Causal studies are conducted through the use of experiments.

Fractional Factorial Design

It's a fraction of the runs in a full factorial design

What is an ecological fallacy?

Misinterpret result. Erroneously drawing conclusions about individuals solely from the observation of groups.

VIRTUAL TEST MARKETS

Sometimes called pre-test-market models because they use a simulated shopping experience to estimate actual market place performance (Urban and Katz, 1983, JMR)

Why is causal research so important to marketing research and marketing management?

When a marketer wants to understand what causes a consumer or business to make a decision, we require causal research to adequately answer the question.

What is a Posttest group experiment

When a measurement of the dependent variable is taken after changing the independent variable.

What is a Pretest group experiment

When a measurement of the dependent variable is taken prior to changing the independent variable.

Laboratory Experiment

are those in which the independent variable is manipulated and measures of the dependent variable are taken in a contrived, artificial setting for the purpose of controlling the many possible extraneous variables that ma affect the dependent variable.

Extraneous variable

are those variables that may have some effect on a dependent variable yet are not independent variables.

What is an DV

are those variables that we have little or no direct control over, yet we have a strong interest in. Ex. Examples would be customer purchase, brand attitude, return on investment, net profits, market share, customer satisfaction, etc.

What is an IV

are those variables which the researcher has control over and wishes to manipulate. Ex. as usually being factors within our control as an organization

Controlled test markets

are used by outside(external supplier) research firms that guarantee distribution of the products through prespecified types and numbers of distribution

Usually requires translating_______ into actionable and measurable variables

constructs

Within subject design

each individual person receive each IV treatment

Experimental Effect

effect, usually measured numerically, of the independent variable on the dependent variable. E = (O2 - O1) - (O4 - O3)

Full Factorial Design

every level of every factor appears with every level of every other factor

Laboratory experiments are high on____ often at the expense of ______

internal validity, external validity

Experimental Group

is a group that has been exposed to a change in the independent variable

Group Control

is a group whose subjects have not been exposed to the change in the independent variable

External Validity

refers to the extent to which the relationship observed between the independent and dependent variables during the experiment is generalizable to the real world.

Since "laboratory experiments" are impractical or impossible for many marketing research objectives...

so marketers use Test Markets or Virtual Test Markets instead

Marketing experiment are a____, is is easy to misinterpret results and make an ____

social science experiments, Ecological fallacy

Test Marketing

the limited introduction of a product and a marketing program to determine the reactions of potential customers in a market situation

Operalization

the movement of variables from the rather abstract conceptual level to the very real measured level . ex: a survey

Experiment Design

usually requires translating constructs into actionable & measurable variables

All experiments will have

≥ 1 independent variable Marketers must have control over administering the different levels of the independent variable(s) (aka - treatments) ≥ 1 dependent variable One or more appropriate ways to measure the dependent variable -Efforts to control extraneous variables The design of the experiment can do this; it can also be done to an extent by measuring the extraneous variable then statistically controlling for it later.