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;
30 Cards in this Set
- Front
- Back
Primary sources (3 general things)
|
1. Generally more expensive (time and money) than secondary data-have to do data collection
2. Allows for customization 3. Must first decide HOW to collect data from users |
|
Sampling decisions-probability samples (3)
|
Utilizes some form of random selection of participants (subset through US)
1. Simple random sample-everyone in a given population has an equal probability of being selection (lower costs) 2. Stratified random sample-derive sub samples based on cultural background...simple from homogeneous groups 3. Cluster (area) sample-Divide geographic clusters and randomly choose-talk to everyone in cluster-could also cluster for time periods |
|
Nonprobability samples (1)
|
Convenience samples-draw participants that is weightable-close to us-like MKT 300 students as population-but difficult to generalize to entire population of US students--shoppers in shopping mall on Monday morning..not representative
|
|
One example of crowdsourcing
|
Amazon Mturk-cheap fast way-paying amount of money--is it relevant? would have to take results with caution
|
|
2 different approaches to collecting data
|
1. Qualitative (in depth, lots of info): useful for hypothesis and model building, can use projective tests (like sentence completion), observational research, ethnographic research, focus groups
2. Quantitative: more common in later stages, to test hypotheses, surveys and experiments--once you have good feeling of research, do this..more common for later stages in marketing |
|
Examples of Projective tests
|
TAT Thematic Apperception Test--fill in story
inkplot tests..confronted with pics |
|
Qualitative example (Zaltman Metaphor Elicitation Technique)
|
DuPont investigated attitudes toward pantyhose
Some distressed images Some appealing images Overall, images suggested a surprising love/hate relationship with pantyhouse--call out set of pics-come up with pics-make collage-the interview |
|
Observational research
|
Excellent source of primary, exploratory data (don't ask but observe)
Observe shoppers, equip shoppers with pagers, ask them to note what they're doing when paged (mystery shoppers-random customer and then work toward management)----problem: don't know why doing these things |
|
More on observational research (ethnographic research)
|
Observational research in which researcher immerses self into consumers' lives, culture (become a customer!)
Does not observe, but also ask questions of subjects MTV teenagers example |
|
Focus groups
|
Widely used in most kinds of mkt research (relativily fast and not too expensive)
Emphasis on group interaction when focused on a topic introduced by a moderator-like new running shoe Open ended discussion with moderator ensuring that relevant issues are brought up 3 Uses: Generate new ideas, uncovering basic customer needs, attitudes, priorities Testing concepts (often for new products) |
|
Group dynamics in focus groups (3 things to consider)
|
1. Peer pressure can often lead to misleading results-other group members may be reluctant to oppose passionate minority
2. Such dynamics produce especially unreliable results for goods largely consumed in private-headache med, prep H 3. But those dynamics could be relevant for goods consumed and discussed publicly..New Coke *small minority against New Coke-find out why minorities feel a certain way* |
|
Purpose of experiments
|
Isolate effects of independent variables on dependent variables
|
|
Experiments allow us to draw causal conclusions (4)
|
1. The essential feature of experiments: random assignment
2. Experiments differ in their internal and external validity 3. Internal validity (ability to confidently draw causal conclusions--how clean experiment, free of confounds?) 4. Confounded variable: Can't use both coke and pepsi when looking the shape of glass and how much consumed |
|
External validity
|
Ability to generalize from research setting to other contexts in real world-is he research environment context free or results useless?
Difficult to achieve both high internal and external validity |
|
Causation
|
one variable PRODUCES movement in other variable
|
|
Correlation ____, but not ____ for causality
|
necessary but not sufficient
|
|
3 factors necessary for casusal conclusion
|
1. Correlation
2. Temporal antecedence (a before b)--cause before effect 3. No third factor driving both-rule out all kind of confounds |
|
Implicit egotism
|
Notion that name similarity influences major life decisions
but maybe matches reversed (causality like ford created ford) |
|
Questions can ____ _____
|
shape answers
|
|
Rating scales bias
|
Respondents infer the researcher's intent based on numbers used in rating scale
different kinds of 11 point scales |
|
Estimating behavioral frequency
|
Respondents assume researcher constructs a meaningful scales based on knowledge of expectations about distribution of behavior in real world (scale is creating bias for distribution of frequencies)
|
|
Better way to measure behavioral frequency
|
How many hours a day you watch TV (open ended)
don't use vague quantifiers (like occasionally and frequently) |
|
Question order
|
Supposed we want to cleanly isolate life satisfaction from marital satisfaction:
1. joint lead in can be helpful (we will now ask you 2 questions about your well being) 2. Rewording life satisfaction Q can also be helpful.."aside from your maarriage, which you already told us about..) |
|
3 other common survey wording errors
|
1. Ambiguously worded questions (what is your church pref...Sex?)
2. Double barreled questions (do you eat Wend'ys hamburgers and Chili?) 3. Leading questions: you agree that Subway is great? |
|
What do we value? Dictionaries and comparisons
|
WTP for A, B or A and B
If consumers will be comparing your product to close competitors, asking them to evaluate your product in isolation may not be that informative |
|
What is better..direct inquiry or indirect inquiry?
|
indirect
Admission and denial with having done something before..maybe ask zaglia about this? |
|
fMRI
|
Functional Magnetic Resonance Imaging
Measures neural activity by tracking changes in blood oxygenation (useful when hard time saying what feel about something)--if having difficulty articulating Problem: reverse inference...often hear "because area 1 active, person must have been engaging in process Z"..however, area 1 also associated with processes of A, B, C, D, Z---different processes |
|
Comparing self report to fMRI
|
B most effective on self report, but FMRI had clear outcome for C in results
|
|
rTMS (repetitive transcranial Magnetic Stimulation)
|
Manipulative brain stuff..don't know if activation drives behavior
|
|
Non-informative neuroscience data partially rescues bad explanations
|
curse of knowlege and brain data..people like it even if not very helpful no real support
|