Exam 3 - Chapter 8-12 Mrkt Research Flashcards
Random Sampling?
Always present unless you take a census, always improve this by increasing sample size or distribution that is tight.
Systematic Error?
Way a question was asked. Includes nonresponse error and self-selection bias(People who feel strongly are more likely to respond to a survey)
Meaning of Self-Selection Bias?
You select yourself to be apart of a survey
Meaning of “Testing Error”?
Act of testing changes the responses that you have. Feel special because they’re given their opinion
Meaning of Hawthorne Effect?
Act of being watched changes productivity, not temperature changes or lighting
Acquiescence Bias?
Tendency to concur with a particular position/go with the flow
Extremity Bias?
Embracing or avoiding extreme positions in responses
Interviewer Bias?
Responses influenced by the relationship between the interviewer and the respondent
Social Desirability Bias?
Wish to create a favorable impression and or avoid embarrassment
Auspices Bias?
Responses are altered by knowledge of the sponsor of the study
Characteristics of Quantitive Research?
Numbers, structured responses, statistically significant
Characteristics of Qualitative Research?
Words and phrases, interpreted, difficult to administer, no numbers, small sample size
Cross Sectional Studies?
Data collected at a single point in time, generally segmented in some way
Longitudinal Study?
Respondents questioned at multiple points in time. (tend to be repeated over a period of time over two ways. Attractive study, you change the sample so you get out of testing error. Get a panel and keep asking the same questions. Panel, you don’t change the sample. Do have testing error because they become sensitized to studies.)
Types of Longitudinal Studies?
Tracking Study(Using successive waves of similar samples); Consumer Panel(Using the same sample over and over again to track attitudes and/or behaviors); Large omnibus panels are often put together by contractors(Thousands of people)
TQM Concepts?
Quality(A Bundle of satisfactions defined by the customer’s perception as to how the satisfaction compared with expectations.
Internal and External Customers: Concept that every person has at least one customer, usually many, inside and/or outside the organization.
TQM requires considerable surveying inside and outside of the company in order to measure customer satisfaction, employee motivation and satisfaction, standards to utilize benchmarks
Choices among Personal Interview?
Telephone Interview, Self-Administered Mail, Door-to-Door,Fax, Internet(E-mails, Internet Survey)
Characteristics of Telephone Interview?
Telephone Interview(Best for monitoring, people overseeing can monitor) Fast, has its own issues
Characteristics of Mail Questionnaire?
Very good for sensitive information because its anonymous. Response rate very low. Offer incentives, following up, making it sound important, leaving a message on the phone that a mail survey is coming, donation to charity
Characteristics of Door-To-Door?
High Response rate, may miss certain groups(working moms, secure apartments)
Survey Pretesting?
Send to experts if you want them to look at how the questions are asked. Trial Run may be done to see if they understand the questions, problems didn’t anticipate
How To Increase Response Rates?
Cover letter(Personalize), Money to charity, interesting questions, follow ups, advance notification, survey sponsorship
Should you Survey Pretest?
Always do it if you have the time
Pretesting by experts?
Experts can find logic errors or how questions are asked wrong(systematic errors)
Pretesting run by a small group of participants?
Understand what problems are, fatigue
What Can Be Observed in Observational Research?
Physical Actions, Verbal Behavior, Non-verbal expession(Tone of voice, body language) How close people stand to another, Temporal(How long they take to do something), Physical disturbance(Trash)
Meaning of Complementary Evidence?
Reinforcing or illuminating a communicated response
Response Latency?
The time it takes to decide to do something
Observer Error?
The more complicated the survey is, the more likely an error is to happen. An observer sees part of themselves in their subject; Common Cause: fatigue, halo effect(dressed a certain way)
Halo effect?
Dressed a certain way.
Contrived Observation?
Artificial creation of an observable moment(Bike crashing, women dropping books)
Guidelines for ethics of observation?
Consent, Expectation of observation, Can’t make out the faces or identities - thats ok too
Types of Observation of Objects?
Trash Analysis, Inventory analysis, Patterns of wear or usage,
Meaning/Examples of Content Analysis?
Content analysis; Basically, looking at the number of times words are repeated.
Counting letters to editor, knowing which one are in favor or against, not concerned about content.
# of people murdered on news a day.
Problems in TV Monitoring?
1) Users compliance,
2) So many channels, difficult to get a sample size that is meaningful.
3) People just turn the TV on and forget about it
Advantages of Observation Research?
Not talking to people, getting rid of observer bias.
Dis-advantage of Observation Research?
Long time, Don’t know the motives of the people.
Meaning of click through rate?
Proportion of people who are exposed to an ad and enter the website
Extraneous(Confounding) Variable?
Extraneous variable, something you can’t control. Weather
Moderating Variable?
Moderating, One you can test for. Tells the relationship between independent and dependent variable
Meaning of Demand Characteristics?
Giving away the purpose of the study when you give instructions on how they are supposed to take a survey
Placebo effect?
seen advertising even though they haven’t. Taking sugar pill, feeling better after.
Means of Overcoming Potential Errors in Sample Selection?
Randomization, Matching, & Blocking
Randomization?
Make your sample size large enough to ensure that things are random enough
Matching?
Splitting your sample groups up. Same number of Males and Females.
Blocking?
Take some of the variables out, and make subsets out of them. Men and Women; Income; Age; Creating a bunch of moderating variables
Randomization?
Make your sample size large enough to ensure that things are random enough
Matching?
Splitting your sample groups up. Same number of Males and Females.
Blocking?
Take some of the variables out, and make subsets out of them. Men and Women; Income; Age; Creating a bunch of moderating variables
Fundamental Experimental Design Questions?
One independent variables, one dependent - Basic Experiment
Factorial, Two independent, two dependent variables
Within Subjects?
Within Subjects: Subjects see all of the treatments, three advertising programs see all of them 3 of 3
Between Subjects?
Between Subjects: See one of three ads. 1 of 3
Field Experiments?
conducted in stores
Maturation?
People Change over time, fatigued
Lab Experiments
Things can be proven in a lab setting, but not necessarily generalized for the rest of the world
Internal Validity?
Independent variables causes effect in dependent variable
Maturation?
Something happens to the subjects, fatigue, sickness
Manipulation?
large enough difference between treatments. Have a difference between 1.79 to 2.59. Not 1.69 to 1.59
History?
History: 9/11, Russia, Evil(Crimea)
Maturation?
Something happens to the subjects, fatigue, sickness
Instrumentation Error?
They Change the phrasing of the question inadvertently
Mortality Error?
people leave experiment, go out of town, sickness, die, etc..
O1 O2 O3 X O4 O5 O6
Cash Register Receipts. Change Music. Monitor Receipts Again
O Means?
Survey
X Means?
Treatment
When to test Market?
Reputation is at stake, Building new plant, Fear of competitive copying
Uses of test marketing?
Forecast new products, test elements of marketing mix, identifies product weakness, storage condition,
Advantage Of Test Marketing?
Real World Setting, Simplicity of Analysis and Communication
When to test Market?
Reputation is at stake, Building new plant, Fear of competitive copying
Dis-advantage Of Test Marketing?
Time, money, competitive reaction
Advantage Of Test Marketing?
Real World Setting, Simplicity of Analysis and Communication
Alternatives - Controlled Test Market?
Company does everything for you, don’t get the distribution reaction out of it.
Criteria for Selecting Test Markets?
Not too large of population Representative of Population Media Coverage and Isolation Distribution Control Overuse of one town