Marketing 360 1nd Edition Lecture 6 Outline of Last Lecture 1. Ethics2. Consumerism and the Bill of Rights3. AMA Ethical Values and Norms4. Ethical Dilemmas as social Dilemmas5. Consumer/ Market Orientation6. Social Marketing Orientation7. Patagonia’s Stewardship8. Example of CBBR Research Outline of Current Lecture 1. The marketing information system2. Four sources of data in MIS3. MDSS4. Comparing MIS to MDSS5. Data Mining6. Seven steps in marketing research7. Establishing cause and effect8. Factors needed to prove cause and effect 9. True experiments vs quasi-experiments10. Illustrating confounds in quasi-experiments11. Research questions, hypotheses, theories, and models12. An example (failed service recoveries)Current Lecture1. The Marketing Information System These notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.a. Determines what information marketing managers need, then gathers, sorts, analyzes, stores, and distributes information to system usersi. Internal company dataii. Marketing intelligenceiii. Marketing researchiv. Acquired databases 1. Computer hardware and softwarea. Information for marketing decisions 2. Four sources of data in MISa. Internal company datai. Information on sales, who buys what, when, how much, back-orders, effect of marketing activitiesb. Marketing intelligencei. Much information available on web1. Company sites, government sites, trade associationsii. Futuristic scenarios (what if?)c. Marketing researchi. Collecting, analyzing, interpreting info about customers, competitors, business environment to improve marketing effectivenessd. Syndicated researchi. Collected by a different firm, firm then sells information to other businessese. Custom researchi. Research conducted for a single firm to provide specific information managers needf. Acquired databasesi. Other firms, governmentii. Misuse has led to government restrictions, do not call list3. MDSSa. Data plus analysis and interactive software allow marketing managers to acces MIS data and conduct analysisi.4. Comparing MIS to MDSSa. MISi. What were our sales last monthb. MDSSi. Increases in sales to overall increase in industry sales, or something unique about our product5. Data Mininga. Sophisticated analysis techniques that take advantage of large amounts of customer transaction information, and discover patterns for different customer groupsb. It helps toi. Acquire new customersii. Know when to hole ‘em 1. Retain existing customersiii. Know when to fold ‘em1. Firing bad customersiv. Market basket analysis1. Target customers for related products6. Seven steps in marketing researcha. Define the research problemi. Key questions, consumer groups of interest, set questions in context1. Based on customer feedbackb. Determine research designi. Secondary research (internal, external sources)1. Collected earlier for some other purposeii. Primary research (exploratory, descriptive, casual)1. Collected to make a specific decisioniii. Three types of primary research1. Exploratorya. Generates insight for future, more rigorous studiesb. Consumer interviewsc. Focus groupi. Product oriented group discussion led by moderatord. Projective techniquei. Gets at underlying feelings, especially when they are difficult to expresse. Case studyi. Of a particular firmf. Ethnographyi. “live” with people, figure out who they use products2. Descriptive (quantitative)a. Probes more systematicall into problem, bases conclusions on large number of observationsb. Ross sectionali. Survey at a single point in time, no cause and effectc. Longitudinali. Responses from same person at multiple timesii. How do preferences change over time?3. Casual researcha. Determined cause and effect relationshipsb. Independent variable= pricec. Dependent variable= salesc. Chose method for collecting datai. Typically either survey or observationii. Quality of data1. Is sample representative of broader population2. Reliabilitya. Are measures free from error?3. Validitya. Does research measure what it intends tod. Design the samplei. Probability sampling1. Simple random samplinga. Everyone has same chance of getting in2. Systematic samplinga. Pick nth person after a random start3. Stratifieda. Divide into groups, sample randomly in each group ii. Non-probability sampling1. Convenience sample2. Quota samplinga. Same in % to different groupse. Collect the datai. Careful training is essentialii. International research poses variety of problems1. Language, regulations, access, customsf. Analyze/interpret datag. Prepare report1. Executive summary, describe method/results, limitations, conclusions, recommendations for action7. Establishing cause and effecta. Why is establishing cause and effect important?i. The ability to prove cause and effect puts a researcher (and decision maker) in the best position to control outcomeii. A study that can prove cause and effect has high “internal validity’b. Correlation doesn’t prove causationi. Much marketing research produces correlational dataii. Correlational data is useful for highlighting relationships, but it is only descriptive and should be interpreted with cautioniii. Correlation does not prove causationiv. Common problems1. Reverse causalitya. We think X causes Y, but actually Y causes X2. Spurious relationships caused by a third variablea. Ice cream consumption and drowning are correlated, but itis due to the weather8. Factors needed to prove cause and effecta. Temporal sequencei. The cause must precede the effect in timeb. Concomitant variationi. The cause and effect must “covary”c. No spurious associationi. No confounds or third variable problemsd. Random assignmenti. Avoids spurious relationships and confounds 9. True experiments vs quasi-experimentsa. Experiments in generali. Researcher manipulated the casual variable (independent variable) and measures how it affects the response variable (dependent variable)b. True experimental designsi. Assignment of individual participants to the study conditions is randomii. In theory, eliminates any systematic cofound between characteristics of the participants and the study conditionsiii. Only design that allows for true cause and effect conclusionsc. Quasi-experimental designsi. Researcher randomly assigns locations to study conditions, but is unable to randomly assign individual participants to the conditionsii. Pre-existing
View Full Document