Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
107 Cards in this Set
- Front
- Back
Casual relationships are determined by...
|
the use of experiments
|
|
Regarding Casual Inference, surveys can help you infer ____, but not strong enough for ____ ____
|
1. Causality
2. Scientific evidence |
|
Casual research design
|
(X -> Y)
|
|
Independent Variable
|
X
AKA: casual variable, treatment, or stimulus |
|
Dependent Variable
|
Y
AKA: Outcome, caused variable |
|
May be thought of as understanding a phenomenon in terms of conditional statements of the form "If X, then Y.
|
Causality
|
|
Describe the sample (hence the name descriptive study) - after running an ad, you may ask a # of questions.
|
Survey
|
|
2 groups, one group sees the new ad, other group doesnt. Outcome could be attitude toward the brand.
|
Experiment
|
|
If treatment X is given, the outcome Y occurs.
If treatment X is not given, then the outcome Y should not occur. Which two groups? |
Experimental group
Control group |
|
Variables that may have some effect on a dependent variable but yet are not independent variables.
|
Extraneous Variables
|
|
Methods for controlling Extraneous Variables:
(3) |
Randomization
Physical Control Design Control |
|
The random assignment of subjects to treatment conditions to ensure equal representation of subject characteristics
|
Randomization
|
|
Holding constant the extraneous variables throughout the course of an experiment (Ex. matching participants on personal demographic characteristics)
|
Physical Control
(used when randomization is NOT possible) |
|
Using the experimental design itself to control extraneous casual factors
|
Design Control
|
|
An ___ ___ 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.
|
Experimental design
|
|
When a measurement of the dependent variable is taken prior to changing the independent variable, the measurement is sometimes called a ____
|
Pretest
|
|
When a measurement of the dependent variable is taken after changing the independent variable, the measurement is sometimes called a ____
|
Posttest
|
|
A "___" ___ design is one that truely isolates the effects of the independent variable on the dependent variable while controlling for effects of any extraneous variables
|
"True" Experimental Design
|
|
Designs that do not properly control for the effects of extraneous variables on our dependent variable are known as ___ ____
|
quasi-experimental designs
|
|
There is no ___ ____ in quasi-experimental designs
|
experimental effect
|
|
Internal Validity
|
the observed change in the dependent variable is, in fact, due to the independent variable. (establish casual relationship X-Y)
|
|
External Validity
(responsiveness of your sample) |
the results of the experiment apply to the "real world" outside the experimental setting.
|
|
___ experiments are those in which the independent variable is manipulated and measures of the dependent variable are taken in a contrived, artifical setting for the purpose of controlling the many possible extraneous variables that may affect the dependent variable.
|
Laboratory Experiments
(adantages: quick, less expensive) (disadantage: artifical setting) |
|
____experiments are those in which the independent variables are manipulated and the measurements of the dependent variable are made on test units in their natural setting.
|
Field Experiments
(adv: should hold true in real world) (disadv: expensive, time-consuming) |
|
8 Threats to Internal Validity in Experiments
(name 6) |
Selection Bias
History Maturation Testing Instrumentation Martality |
|
___ ___ may have been present if the experiment was only given to those who showed interest.
|
Selection Bias
|
|
AC Nielsen's ___ Index
|
Television
|
|
A form of field experiment for assessing the market's reactions to new/existing product and its associated marketing mix.
|
Test Marketing
|
|
Tests a products marketing mix variables through existing (normal) distribution channels.
|
Standard test Market
|
|
Performed by an outside firm that guarantees distribution of the test product through outlets of selected cities.
|
Controlled Test Market
|
|
___ ___ markets are those in which a limited amount of data on consumer response to a new product is fed into a model containing certain assumptions regarding planned marketing programs, which generates likely product sales volume.
|
Simulated Test Markets (Virtual)
|
|
Advantages of Simulated Test Markets
(and provide example) |
Convienence, Cost Savings (dont have to buy shelf space), Don't have to have physical product ready.
Example: CyberScan.com |
|
___ ____ markets are those in which a panel of consumers has agreed to carry identification cards that each consumer presents when buying goods and services.
(example and advantage) |
Electronic Test Markets
Example: Kroger Card Advantage: Gives individual specific data |
|
- multiple versions of semi-finished product are tested by consumers
|
B2C
|
|
- they key technology is presented to select clients
|
B2B (Industrial Market)
|
|
Possible negatives with industrial test market..
|
Confidence agreement is not very effective
~must give prototype to clients you trust, like long term clients |
|
A ___ ___ ___ market is test marketing conducted in specific foreign countries that seem to be good predictors for an entire continent.
Examples: 2 |
Lead Country Test Maket
Colgate-Palmolive's new shampoo and conditioner Swiffer - first in Japan. |
|
Colgate-Palmolive tested their new shampoo/conditioner in what countries first?
(4) |
Philipines, Autralia, Mexico, and Hong Kong
for 1 year |
|
Test Marketing Examples:
McDonald's Tests McPizza in __ stores in ___ ___ |
24 stores in Evansville, Indiana
|
|
Subway Pizza started in a handful of stores in __, then expanded to __ resturants, then went ___
|
New York City
15 Nationwide |
|
Startbucks $1 coffee started in _____
|
Seatle, Washington
|
|
Nature of manipulation in test markets
|
4 P's
|
|
Top 5 test market cities are:
|
Albany (NY), Charlotte(NC), Rochester(NY), Nashville(TN), and Eugene(OR)
|
|
Advantages of Test Marketing
|
Most accurate method for forcasting future sales, allows opportunity to pretest marketing mix variables
|
|
Disadvantages of Test Marketing
|
Results may be wrong (not guaranteed), competitors can sabatoge test marketing efforts (ex. mountain dew sport), expensive, and time-consuming.
|
|
Total Error =
|
Sampling Error + Non-sampling Error
|
|
Sampling error is the same thing as ___ validity
|
External
|
|
Non Sampling (systematic error) error is the same thing as ___ validity.
|
Internal
|
|
4 Types of Internal Validity issues.
|
Respondent error (no response, refusal, hostility)
Researcher Biases, Measurement (bad questionaire), Design Errors (bad experimental design) Problem with definitional Errors (bad MRP's) Researchers Administrative errors (loosing data, mistakes processing data) |
|
___ ___ is any error in a survey that occurs because a sample is used.
Caused by what 2 factors? |
Sampling Error
1. Method of sample selection 2. Size of the sample |
|
"Large Sample Size Bias"
|
People think larger sample size is always better, which is not the case.
1000 frat guys vs 500 randoms |
|
Statistical consideration for survey.
|
minimum of 50 people in sample. Stat analysis reuires at least that.
|
|
The only perfectly accurate sample is a ____
|
Census
|
|
The ___ ___ ___ is the correct method by which to determine sample size, which applies the concepts of accuracy (sample error), variability, and confidence interval to create a "correct" sample size.
|
Confidence Interval Approach
|
|
____.com uses national opinion polls.
|
Gallop.com
+ or - 3% sample error (which is standard) |
|
National opinion polls tend to use sample sizes ranging from _____-____.
Why? |
1000-1200
With a sample size of 1000 or more, very little gain in accuracy occurs even with doubling or trippling the sample to 2000 or 3000. |
|
____refers to how similar or dissimilar responses are to a given question.
Represented by _ and _ |
Variability
p and q |
|
The larger the size of the (probability) sample, the less is its sample error.
T or F? |
True
|
|
3 Types of Probability Sampling:
Simple Random - Stratified Random - Cluster - |
Simple Random - each element (individual) in the population has a known and equal chance of being selected
Stratified Random - population is partitioned into sub-populations, then elements are randomly selected from each strata at random. Cluster - Population is partitioned into sub-populations, or clusters, then clusters are selected at random, then either all or some of each selected cluster taken as part of the sample. |
|
Advantages of Simple Random Sampling (2)
|
Representative and generalizable.
|
|
Types of Non-Probability Sampling:
|
Convenience
Quota Snowball |
|
____ sampling attempts to obtain a sample of convenient elements.
Adv / Disadv.? |
Convenience Sampling
Adv: Least expensive, and time consuming Disadv: findings generalizability |
|
___ sampling is a two stage, restricted judgmental sampling. First stage, develop control categories or quotas of population elements. Second, sample within categories.
|
Quota Sampling
Non-equal chance of being selected! |
|
___ Sampling: initial random sample selected, then each respondent identifies others in target population.
|
Snowball Sampling
|
|
Sampling techniques in which each element has a fixed probability of being selected.
|
Probability Sampling
|
|
Sampling techniques that do not use chance selection procedures. Rather they rely on the personal judgement of the researchers.
|
Non-Probability Sampling
|
|
a master lsit of all the smaple units in the population is called a ___ ___
|
sample frame
|
|
The degree to which the sample frame fails to account for all of the population
|
Sample frame error
|
|
A structured questionaire given to a sample of a population (sampling frame) and designed to elicit specific information from respondents
|
Survey
|
|
Descriptive research is undertaken to describe answers to questions of __, ___, ___, __, and ___.
Not ___ |
Who, What, Where, When and how
Not: Why |
|
Surveys tend to be:
(2) |
Based on large samples
Structured (everyone asked same questions) |
|
Advantages of Surveys
|
Standardization
Easy to Administer and record Ability to use advanced statistical analysis, esy to analyze Can detect small differences among subgroups. |
|
Disadvantages of Surveys
|
Difficult to design good questionaire
Limits to the in-depth detail of data Limited control over timeliness, and potentially low response rates Participants may lie |
|
All scales have this characteristic. Discriptors such as Male/Female, Married/Single, etc.
|
Description
|
|
___ refers to the relative sizes of the descriptors, such as great than, less than, and equal to.
Which scales have this characteristic? |
Order
Ordinal, Interval, and Ratio scales |
|
___ refers to the absolute differences between the descriptors which are expressed in units.
Which scales have these characteristics? Exception? |
Distance
Interval and Ratio scales Exception: True Class Interval |
|
A scale is said to have the characteristic of ____ if there is a unique beginning or true zero point for the scale.
Which scale has this characteristic? |
Origin
Ratio scale |
|
A ___ ____ is not a ___ ___ value for a scale
|
Neutral Category
Zero Value |
|
A ___ ___ is one that is typically composed of small number of distant values or categories.
Two Types? |
Categorical Scale
Nominal and Ordinal |
|
A ___ ___ is composed of numbers or labels that have an underlying measurement continuum.
Two Types? |
Metric Scale
Interval and Ratio scales |
|
A scaled-response for commonly used by marketing researches that measures intensity of agreement or disagreement.
|
Likert Scale
|
|
The ___ ___ scale is a good way to measure a brand, company, or store by listing a series of bi-polar adjectives to choose from.
|
Semantic Differential scale
|
|
___scales simply label objects
|
Nominal
|
|
___scales indicate only relative size differences between objects by rank-ordering
|
Ordinal
|
|
___scales use descriptors that are equal distances apart for construct measurement.
Adjectives are ___descriptors Numbers are ____ descriptors |
Interval
Primary Secondary |
|
Scale that uses unequal descriptor distances.
|
True Class Interval Scale
|
|
___scales have a true zero origin exists, variable can be measured directly.
|
Ratio scale
|
|
Use ___ and ___ scales as much as possible because ___ and ___ can be derived
|
Ratio and Interval
Mean and Standard Deviation |
|
A ___ is an abstract idea inferred from specific instances that are thought to be related.
Can only be measured ____ |
Construct
Indirectly |
|
The first step in constructing a scale is to ___ your construct.
|
Operationalize
|
|
The mental vision of how each construct will be measure is called an ___ ___
|
Operational Definition
|
|
___is the truthfulness of responses to a measure
|
Validity
|
|
___ ___ determines if the question "looks like" it measures what it is suppose to measure.
|
Face Validity
|
|
When using semantic differential scale, avoid Halo Effect by ...
|
Flipping adjectives randomly.
|
|
___ ___ is a systematic process in which the researcher contemplates various question formats, considers a number of factors characterizing the survey at hand, words the questions carefully, and organizes the questionaires layout.
|
Questionaire Design
|
|
___ ___ is defined as the ability of a question's wording or format to influence respondents' answers.
|
Question Bias
|
|
In questionaire organization, the introduction includes what: (4)
|
Disguised/Undisguised (purpose of study)
Incentives Anonymity/confidentiality Screening questions |
|
The researcher uses question ____ to scrutinize a possible question for its question bias
|
Question evaluation
|
|
Survey questions should be:
(4) |
Focused
Brief Simple Crystal Clear |
|
10 Words to avoid in question development
|
All
Always Any Anybody Best Ever Every Most Never Worst |
|
Survey questions should not be: (4)
|
Leading
Loaded Double-barreled Overstated (know what each of these is) |
|
Question flow is important in a survey. __ and ___ should be used to accomlish this.
|
Warm up Questions
Transitions |
|
___ __ refers to the introduction and the actual flow of questions on the questionaire
|
Questionaire Organization
|
|
The use of numbers associated with the question responses to facilitate data entry during data collection and data analysis after the survey has been conducted.
|
Coding
|
|
3 types of Non-Probability Sampling:
Convienence Sampling Quota Sampling Snowball Sampling |
Convenience - obtain sample of convenient elements. (people nearby)
Quota - A two-stage, restricted judgemental sampling. First stage, develop control categories or quotas of population elements. Second stage, sample within categories Snowball - Initial random sample selected. Then respondent identifies others in target population. |
|
Convenience Sampling is used most often because its:
Limitation? |
Easy
Cheap Saves time Findings generalizability |
|
Establishing internal validity means....
|
your observations are correct, at least within your sample.
|