Exam 1 Studyguide

What are the problem words which you want to avoid and be careful?

All, Always, Never, Occasionally, Often, Rarely, Regularly, Sometimes, Usually

What kind of table is often used to highlight the information needed in the survey design process?

Dummy Tables

What is multivariate testing?

a. A multivariate test (MVT) is a design-optimization method in which multiple variants of specified variables are tested in a user interface, with the goal of maximizing conversions. b. 2 or more design elements (the variables) are tested. Each of these variables could have multiple variants.

What are the advantages and disadvantages of focus groups/online focus groups/social media focus groups?

a. Advantages: i. "Snowballing" effect: One person's experience or feelings often gets others to contribute ii. Good for collecting in depth information iii. Allows for spontaneity and enables complex issues to be discussed. iv. Process highlights differences between consumers! b. Disadvantages: i. Results are not directly quantifiable. ii. Not representative and difficult to get attendees. iii. Group think, minority viewpoints may not be heard. iv. Moderator must be experienced. v. Presence of marketing researcher may bias the results.

What are the practical tips for holding focus groups?

a. Allow one speaker at a time b. Encourage everyone to participate c. Stick with an idea when it arises, until the group finishes it d. Encourage participants to disagree, if they do e. Follow the moderator guide

What is experimentation?

a. An experiment is a procedure in which: i. One or more causal variables (treatments) are systematically manipulated. ii. Data on the outcome are gathered iii. Other variables are controlled that may influence the effect variable (Age, Income, etc...)

How to make sure that the respondents have the information you need?

a. Ask questions respondents know about! b. Fact: Most questions will be answered, but will they mean anything? c. Use filter (screening, qualifying) questions

What are the things you want to avoid when you design a survey?

a. Avoid Unstated Alternatives i. An alternative answer not expressed in a question's options b. Avoid leading questions c. Avoid assumed consequences (clearly state the possible consequences) d. Avoid generalizations and estimates

What are the Five misconceptions of marketing research?

"Big decision" myth: i. Turn to marketing research only when you have a major decision to make "Survey myopia" myth: i. Marketing research is the same as field survey research "Big buck" myth: i. Marketing research is too expensive "Sophisticated researched" myth: i. Only trained experts can and should pursue it "Most research is not read" myth: i. Marketing research is irrelevant to managers or simply confirms what they already know.

What is the rank-order of different marketing measurement scales?

Highest Level of Measurement: Ratio Second Highest Level of Measurement: Interval Third Highest Level of Measurement: Ordinal Lowest Level of Measurement: Nominal

What is Spurious Correlation?

Spurious: Artificial, Fake

What is "backward" marketing research? What is the process?

The basic idea is that marketing research should be done from the end product forward. There are 8 steps described (in order): 1. Determine how the research results will be implemented 2. "Diagram out" the final report 3. Specify the analyses necessary to do the final report 4. Determine necessary data 5. See if secondary data is available 6. If not, collect primary data 7. Carry out field work 8. Do analysis, write report In this way, the results will be action‐oriented, and the data will be focused on the appropriate problem. The worst type of research is the hope‐and‐pray approach.

What are the two major types of experiments in marketing?

a. Between-subjects Design: Each participant receives only one treatment; comparisons are made between groups of different participants. b. Within-subjects Design: Participant receives more than one treatment, comparisons are made across multiple measures on the same participant.

What is the big data? What are the 5 Vs of big data?

a. Big Data: Huge volumes of data, generated at a very high speed. 1. Volume 2. Velocity 3. Variety 4. Veracity 5. Value

1. What are the three requirements to establish causality?

a. Correlation: (Evidence of association between X and Y, must occur together or vary together) b. Temporal Antecedence: X must occur before Y c. No third factor driving both: Eliminate factors other than X as the cause of Y

Relationship between correlation and causation?

a. Correlation: Relationship between two variables b. Causation: One variable producing an effect in another variable

When will focus groups or depth interviews be preferred?

a. DI: i. Focused on individual narrative and one-on-one conversation ii. When researchers are probing for specific areas such as processes, ideas, and thoughts more deeply than in FG's b. FG: i. When researchers want to generate new ideas and diverse thoughts (Difficult for DI) ii. When researchers want more spontaneous or unconventional ideas

What is the laddering technique?

a. Depth Interview Laddering Technique: i. Typically starts with product characteristics and moves to user characteristics: "Laddering".

Why is it difficult to do field experiments properly?

a. Difficult to control contamination-conditions remaining independent. b. Difficult to control adherence and attrition. c. Difficult to generalize the results- depends on the context.

How to handle sensitive questions?

a. Don't ask unless absolutely necessary b. Guarantee anonymity c. Place sensitive Questions near the end!

What is exploratory research? What are the major types of exploratory research?

a. Exploratory Research: Research conducted to gain ideas and insights to better define the problem or opportunity confronting a manager. b. Types of Exploratory Research: i. Projective Methods ii. Depth Interviews iii. Focus groups

When is exploratory research preferred? What is the goal of exploratory research?

a. Exploratory research is preferred when there is an ambiguous problem. b. The goal of exploratory research is to provide a better understanding of the situation, but not designed to come up with final answers and decisions!

What is a focus group? What are the major characteristics of focus groups?

a. Focus Group: An interview conducted among a small number of individuals simultaneously, that relies more on group discussion than direct questions to generate data. (A form of exploratory research) b. Major Characteristics: i. Typically, 6-10 people (carefully screened) ii. Homogeneous within group, heterogeneity introduced across groups! iii. 1-2 Hours long iv. Topics introduced by the moderator, sessions are recorder and transcribed

What is meaning of heterogeneity for causal research?

a. Heterogeneity: Causal effects may be different in different groups.

What is individual-level data/aggregate-level data/cross-sectional data/longitudinal data/panel data?

a. Individual-Level Data: The data that can be traced back to the individual (consumers) b. Aggregate-Level Data: Individual purchase behaviors are aggregated - the purchase information is no longer traced to any individual consumer. EX: Weekly sales of Starbucks Latte c. Cross-Sectional Data: Example: Sales of Starbucks Coffee during the first week of January 2014 in all Starbucks stores in the US. (Time and Locations) d. Longitudinal Data: Same phenomenon repeatedly measured again and again over time. EX: Macy's sales in January, February, March... e. Panel Data: A combination of Cross Sectional (across individuals), and Longitude (over time) combined.

To calculate the sample size, what information do you need to know? What is the magic number of the survey sample size?

a. Information needed to calculate sample size: Confidence Level, Population Size, Margin of Error (Confidence Interval) b. Magic number: The magic number is the ideal sample size given the required information.

What are the types of secondary data?

a. Internal: Data that originate within the organization for which the research is being done. i. Transaction Data ii. Customer Communications iii. Marketing Research Tracking Studies b. External: Data that originated outside the organization for which the research is being done. i. Published Data ii. Commercial Data

What is the key difference between A/B testing and multivariate testing?

a. Key Difference: A/B testing provides data on comparing multiple versions. Multivariate testing has an advantage, because every possible combination of the UI elements on the page are tested.

What is marketing research?

a. Marketing research is about obtaining Quality information for making better marketing decisions. b. Marketing Research: Principles for systematically collecting, analyzing, and interpreting data that can aid decision makers.

What are the typical problems of scanner data?

a. May not cover the entire population. b. Information is not always detailed enough. c. Doesn't provide measures of emotions and psychographic characteristics. d. Don't know the exact consideration set faced by the consumer. e. May be expensive!

What is the primary difference between focus groups and nominal groups?

a. Nominal Groups: A group interview technique that initially limits respondent interaction while attempting to maximize input from individual group members. (EX: People write answers before sharing them) i. Focus Groups on the other hand, encourage group participation and do not limit interaction in any way.

What is nominal/ordinal/interval/ratio scale?

a. Nominal Scale: Measurement in which numbers are assigned to objects solely for the purpose of identification. Ex: Zip Codes, SSN b. Ordinal Scale: Measurement in which numbers are assigned to the data on the basis of some order of the objects. Ex: Quality rankings, preference rankings. c. Interval Scale: The assigned numbers allow the comparison of the size of the difference among and between members. Ex: Temperature Scale, Attitude towards brands d. Ratio Scale: Measurement has a natural, or absolute zero, and therefore allows the comparison of absolute magnitudes of the numbers. Ex: Length, Weight

What are permissible statistics of nominal/ordinal/interval/ratio scale?

a. Nominal: Count/Frequency/Percentage/Mode b. Ordinal: Count/Frequency/Mode/Median c. Interval: Count/Frequency/Mode/Median/Mean d. Ratio: Count/Frequency/Mode/Median/Mean/Geometric mean

What are the major differences between the nonprobability sample and the probability sample?

a. Nonprobability Sample: Neither sampling error, nor margin of sampling error can be calculated. Inferences cannot be made about the population; therefore results are not generalizable from the sample to the population. b. Probability Sample: Sampling error can be calculated; therefore results are generalizable from the sample to the population.

What are the nonprobability samples/probability samples?

a. Nonprobability Samples: A sample that relies on personal judgement in the element selection process. (Sampling error cannot be estimated) b. Probability Samples: A sample in which each target population element has a known, nonzero chance of being included in the sample. (Selected randomly, Sampling error can be estimated!)

What is the pretest? Why is pretest important?

a. Pretest: Use of a questionnaire on a trial basis, in a small pilot study. b. Why Pretest is important: Pretesting is the least expensive way to make sure the questionnaire research is a success. It is also the last chance the researcher has to ensure the data collection form is working properly.

What are the types of primary marketing data? How to collect primary data (the major types of primary data collection methods)?

a. Primary Data types: i.Demographic/socioeconomic characteristics (WHO) ii. Attitudes and opinions (Thoughts) iii. Awareness/Knowledge (Intended Actions) iv. Motivation/Protocol (Why) v. Behavior (Actions), observational and experimental b. Primary Data Collection Methods: i. Questioning: (Focus groups, depth interviews, surveys) ii. Observation: (Direct observation, Field experiments)

What are the differences between primary and secondary data?

a. Primary Data: Collected by you, Surveys, focus groups, exct.., better fit and accuracy, costs more and takes longer. b. Secondary Data: Collected by someone else, Internal and external secondary data, data is already available, problems of fit and accuracy, costs less and can be collected quickly.

What is secondary data/primary data?

a. Primary Data: Collecting original data for the specific question at hand b. Secondary Data: Data collected (internally or externally) for a different purpose than the specific question at hand.

Why is the probability sampling preferred?

a. Probability sampling is preferred because the analyst can statistically assess the level of sampling error and make projections to the population.

What are projective techniques? Advantages and disadvantages of projective techniques?

a. Projective Techniques: Methods that encourage respondents to reveal their own feelings, thoughts and behaviors by shifting focus away from the individual through the use of indirect tasks (Word association, Sentence completions, Storytelling Role playing) b. Advantages: i. The amount, richness, and accuracy of the information. ii. Enable the researcher to access presumably unreachable beliefs, attitudes, and behaviors. iii. Generate, supplement, and verify hypotheses. c. Disadvantages: i. Interpretation requires a sophisticated skill set. ii. Difficult for some respondents to fully immerse themselves in the exercise. iii. Reliability of instruments may be a major issue.

What are the pros and cons of secondary data? How to assess the accuracy of secondary data?

a. Pros: Time and cost savings b. Cons: i. Problems of fit (different units of measurement, different class definitions) ii. Problems of accuracy (originating source, purpose of publication, general quality) c. Assessing the Accuracy of Secondary Data: i. Primacy of Source: 1. Primary Source: The originating source of secondary data 2. Secondary Source: A source of secondary data that did not originate the data but rather secured them from another source. 3. Sponsor of the Research 4. Evidence of Quality 5. Always use the primary source of secondary data!!!!

1. What are quasi experiments? Differences between true experiments and quasi experiments?

a. Quasi Experiments: A nonrandom method, where researchers often do not have control over the treatment, but instead study pre-existing groups that received different treatments. i. Useful when manipulation is difficult! b. True Experiments: A random assignment method, researcher designs the treatment, requires the use of control and treatment groups. i. True experiments are the only sure way to assess causality!

When asking about the frequency of behaviors, how to make the questions easier for the respondents to answer?

a. Questions should always be asked in specific, rather than general terms b. Use an appropriate time frame c. Provide some ranges (Ex: How many ___, last year, in a typical week, etc?)

1. Why do we need randomization when we run experiments?

a. Randomization is the foundation of experimentation! i. Eliminates self-selection into the treatment ii. Avoids systematic differences between treatment and control. iii. With large enough samples, results at the end of the experiment can be attributed to the thing that is manipulated!

What is response order bias? What is the idea of the split-ballot technique?

a. Response Order Bias: An error that occurs when the response to a question is influenced by the order in which the alternatives are presented. b. Split Ballot Technique: A technique for combatting response bias, in which researchers use multiple versions of a survey, with different wordings of an item or different order of response options.

What is the sampling error? Will increasing sample size help reduce the sampling error?

a. Sampling error: The difference between results obtained from a sample and results that would have been obtained from the whole population. b. Increasing sample size will help reduce sampling error!

How to organize a questionnaire

a. Screeners: Qualifying Questions (Goal is to identify target respondents) b. First Few Questions: Warm-ups (Shows the respondent that the survey is simple) c. First Third of Questions: Transitions (Questions related to research objectives, require slightly more effort) d. Second Third: Difficult and complicated questions (The respondent has committed to finishing the questionnaire) e. Last Third: Classifying and demographic questions (The respondent may leave some "personal" questions blank, but they are at the end of the survey.

What are the advantages of exploratory research?

a. Small Scale b. Flexible

What are the steps to design a survey?

a. Specify what info will be sought b. Determine method of administration c. Determine content of Individual Questions d. Determine form of response to each question e. Determine wording of each question f. Prepare Dummy Tables g. Determine Question Sequence h. Determine Appearance of Questionnaire i. Develop Recruiting message or script j. Reexamine steps 1-9, pretest, and revise if necessary

What are mistakes to avoid while you run A/B testing?

a. Stopping tests too early b. Sending too little or unbalanced traffic to variations c. Using too many elements together d. Failing to consider external factors e. Running multiple A/B tests at the same time f. Ignoring statistical significance /cultural differences

What is syndicated data?

a. Syndicated Data: Commercial data that isn't specific to any one client.

What is A/B testing?

a. Two (or more) versions of an ad, email, webpage. b. Measure the results (ratings, open rate, click through rate) c. Test for significant differences