MKT305 Quiz 2
Why is it important to ensure fieldworkers adhere to sampling procedure specified in project?
A great plan, if not executed properly, may fail to achieve its objectives. If fieldworker doesn't follow scientifically selected plan, may be considerable bias in probject
What are the 3 components of an attitude?
Affective, cognitive, behavioral
Attitude
An enduring disposition to consistently respond in given manner to various aspects of the world (people, events, objects)
Attitude: cognitive
An individual's awareness of & knowledge about an object
Attitude: affective
An individual's general feelings/emotions toward an object Reflects emotional character of attitudes
Develop checklist of things to consider in questionnaire construction
Are simple words used? Is the language vague? Are the questions short? Are the questions easy to understand? Are the questions specific or general? Are the questions leading?
How can a market researcher try to be confident that the target population involved in some research situation is defined properly?
Ask & answer questions about crucial population characteristics: Is a list available that matches our population? Who are we not interested in? What are the relevant market segment characteristics? Are we only interested in a regional population? If so, how to determine the borders? Should the study include multiple populations?
Rating
Asks respondent to estimate magnitude or extent to which some characteristic exists Quantitative score results Involves marking response indicating one's position using 1+ attitudinal/cognitive scales
What are the reasons why a sampling process is so often used in place of a census? Why is it that a population can sometimes be as accurate as, or more accurate than, a census?
Budget/time constraints, sampling accuracy/reliability, interviewer mistakes / tabulation errors during census can cause poor results Because it is more accurate to draw conclusions about populations by measuring sample rather than taking a census
Difference between leading & loaded question?
Difficult to draw fine line, BUT Leading question wording influences answers by providing hint about purpose of question / clue that hints at something Loaded question is emotionally charged
Does this question do good job of avoiding ambiguity? We are going to ask you to classify type of fan you consider self to be for diff sports and sports programs...
Excellent way of defining terms that respondent may not formally use For example, a Diehard Fan could mean lifelong Chicago Cubs fan; however, the definition here makes it clear what this term means in context of this study
Comment on: Department store that wishes to examine whether store is losing or gaining customers draws a sample from its list of credit card holders by selecting every tenth name
Idea of selecting every 10th name systematically must be questioned, whether or not the credit card customers are representative of all the store's customers. If not, there will be problem
Identify issues researcher should consider when choosing measurement scale
Is ranking, sorting, rating, or choice technique best? Should a monadic or comparative scale be used? What type of category labels, if any, will be used for rating scale? How many scale categories or response positions are needed to accurately measure an attitude?
Type of scale: Psychographic statement asking respondent to circle appropriate response
Likert scale item Includes several scale items to form composite scale Large number of statements is generated and assessed for reliability and validity
Is rating or ranking a better attitude measurement technique?
Neither is better -- depends on objectives of research
A lobbying organization designs short questionnaire about political position. Also includes membership solicitation with questionnaire. Ethical?
Not ethical -- purpose of research is research Researcher has obligation to respondent to keep everything above board b/c respondent donates their own time
An interviewer finds when potential respondents ask how much time the survey will take most refuse if they are told 15 minutes. Interviewer now says 10 & finds most enjoy answering. Is this right?
One hand: deceptive -- shouldn't lie about time to complete Other hand: no harm -- respondents seem to enjoy answering the questions Perhaps other ways, such as monetary incentive, could be used
Comment on: Advertising executive suggests that advertising effectiveness be tested in real world. A 1-page ad is placed in magazine. 1-half of space used for ad itself -- other half, short questionnaire requests that readers comment on ad. Incentive will be given for first thousand responses
Poor sampling design. Only those who notice can be in sample -- sample is highly biased. Also, those who "self-select" to respond are in sample -- even more bias
Comment on: Motorcycle manufacturer decided to research consumer characteristics by sending 100 questionnaires to each of its dealers. The dealers would then use their sales records to track down buyers of this brand of motorcycle & distribute the questionnaires.
Poor sampling technique. Delegating responsibility of sampling to dealers who may not know anything about marketing research. Some may have very accurate sales records & randomly process, whereas some would select some most convenient method
Attitude: behavioral
Predisposition to action by reflecting a consumer's buying/purchasing intentions
Differences between random and systematic sampling error?
Random sampling: difference between sample result and result of census conducted using identical procedures. Occurs b/c of chance variation in selection of sampling units -- which may not perfectly rep population b/c of chance variations Systematic results from nonsampling factors, primarily nature of study's design & correctness of execution. These errors systematic in some way & not due to chance fluctuations, so should not be classified as random sampling errors
Ranking
Requires respondent to rank order a smaller # of items on basis of overall preference or some characteristic of the stimulus
Difference between population and sample?
Sample: subset, or some part, of a larger population, from which population characteristics are estimated Population: any complete group whose members share some common set of characteristics; a census is investigation of all individual elements that make up population
What is nonprobability sample? Some examples?
Sampling technique in which units of sample are selected on basis of personal judgment/convenience. Probability of any particular member being chosen: unknown Examples: convenience samples, judgment sampling, quota sampling, snowball sampling
Type of scale: How favorable are you toward Fair Tax Proposal?
Semantic differential item Actually a series of attitude scales Many marketing researchers find it desirable to assume it provides internal data, which is considered acceptable because fairly small amount of error
Type of scale: U.S. rep's questionnare sent to constituents
Simple rating scale w/ 2 categories However, a non-forced choice scale that provides a "no opinion" category
Comment on: Research company obtains sample for focus group thru organized groups such as church groups, clubs, schools. Organizations are paid for securing respondents; no individual directly compensated
Technique is often used. Convenient b/c often generates group session for lively convo. HOWEVER, if groups unrepresentative of population, should not be used.
When would researcher use judgment or purposive sample?
Used when a researcher knows some characteristics of population element required to fulfill research's purpose. Test markets typically selected based on investigation of demographic characteristics & execs' judgment
How should fieldworkers terminate interview?
Wait to close until have secured all pertinent information Avoid hasty departures to be courteous Should answer any respondent questions concerning nature & purpose of study to best of ability Always thank respondents for time & cooperation -- leave respondent with positive feeling
What should interviewer do if question is misunderstood? If respondent answers before encountered?
When misunderstood, instructed to repeat question. If skilled interviewers used, when respondent answers question before encountered, fieldworker will record the response where it belongs, if possible
When should interviewers probe? Give examples of how should be done
When respondents: Give no answer Give incomplete answers Give answers that require clarification Examples: 1. Repeat question 2. Use silent probe 3. Repeat respondent's reply 4. Ask neutral question
Comment on: Researcher suggests replacing an online consumer panel w/ sample of customers intercepted @ Wal-Mart stores in Atlanta, Jacksonville, St. Louis, Seattle. Goal is to understand basic consumers opinions on products sold by GM & P&G
Will be several flaws & biases from a sample of shoppers from one particular store in 4 cities in country Store in each city may not represent basic consumer opinions regarding these brands. To improve, shoppers from all types of store in which these products are sold should be considered
Comment on: Banner ad on business-oriented Web site reads "are you a large company sr. exec? Qualified execs receive $50 for under 10 minutes of time. Take survey now!" Is this appropriate?
Will not yield a probability sample as business execs are a hard to reach population. HOWEVER, may be cost-effective way to conduct Internet survey with execs.
Is convenience sampling ever appropriate?
Yes, it is appropriate for researchers to use when having to obtain a large number of completed questionnaires quickly/economically or when obtaining sample through other means is impractical. However, must remember that projecting results beyond that specific sample: inappropriate Best used for exploratory research when additional research will be subsequently conducted with probability sample
List decisions that researchers face in trying to construct valid & reliable questionnare.
1. What should be asked? 2. How should questions be phrased? 3. In what sequence should questions be arranged? What questionnaire layout will best serve research objectives? 4. How should questionnaire be pretested? Does questionnaire need to be revised?
Net Promoter Score (NPS)
The score based on customer responses to a question asking how likely they would be to recommend the company to friend/colleague. Aimed at forecasting future based on valence & strength of someone's current attitude
How might the wording of a question about income influence respondents' answers?
The wording of questions on sensitive issues, such as income, may provide respondent with a tendency to answer more heavily in a particular category
Why do marketing managers place great deal of importance on attitudes?
They have been shown to predict behavior with some accuracy If the knowledge about a product can be changed, the product can be repositioned to capture some new market
A fieldworker asks respondents whether they will answer a few questions about shopping center & neighboring community. However, interviewer also observes respondent's race & approx. age as well as other visible characteristics including manner of dress. Is this ethical?
This is common tactic in marketing research. This observation does no harm to respondent.
Comment on: Citizen's group interested in generating public & financial support for new university basketball arena prints questionnaire in area newspapers. Readers return questionnaires by mail
Those most interested in basketball or those most opposed to financial program will be most likely to spot & return questionnaire. Self-selection will probably cause sample to bias toward 2 extreme groups