Marketing research questions

1 Discuss what it means for a measurement to be valid, and give a specific example of steps that might be taken to test the validity of a measure.

A valid measure is one that measures what it claims to measure. Examples of tests for validity should involve corroborating the result with comparisons to related constructs, expert judgments, different measures taken at the same time or a predicted marketing phenomenon at a future time.

A major disadvantage of the focus group interview is that A.the decision maker cannot use the evidence as part of conclusive research.B.participants feel uncomfortable in groups and are therefore less willing to express their views C.data is not provided for analysis D.scientific scrutiny is not possible

A.the decision maker cannot use the evidence as part of conclusive research.

Research aimed at making predictions regarding the occurrence of marketing phenomena based on association between marketing variables is a form of:a descriptive research b psychographic analysisccausal researchd market potential study

Ans: aRationale: Descriptive research includes research with the objectives of portraying the characteristics of marketing phenomena and determining the frequency of occurrence, determining the degree to which marketing variables are associated, or making predictions regarding the occurrence of marketing phenomena. Causal research requires rigor in determining a cause-and-effect relationship rather than mere association, usually through experimentation. A market potential study does not necessarily make predictions

Examples of inputs (independent variables) and outputs (dependent variables) studied in marketing research are, respectively:a marketing mix; profits.b performance measures; pricing.cbehavioral responses; situational factors.d government regulation; economic climate.

Ans: aRationale: Inputs/independent variables include situational factors (such as government regulation and economic climate) and the marketing mix; outputs/dependent variables include behavioral responses and performance measures. Marketers wish to relate variables they can either control or measure to those which indicate favorable marketing outcomes, like sales and profit.

4In which of the following ways would initial questions be most likely to bias the sample (i.e., those who agree to provide information)?aif questions are too uninteresting to the respondentbif questions are partisan or highly specificcif questions are too simplistic or obviousdif responses are not mutually exclusive

Ans: aRationale: Option b could bias the results of later questions, but not necessarily the sample. Option c would not bias the sample, unless the level of simplicity was so extreme as to be offensive. Option d would typically have no effect on sample selection at all. Option a is the best choice; if the initial questions are too uninteresting, respondents may simply refuse to answer, believing that the survey is boring or fails to pertain to them. This is particularly so for phone and mail surveys.

1 How do qualitative techniques aid exploratory research?a by providing primary sources of data on consumer behavior and attitudesb by providing background information on a management problem and a historical perspective from which to assess the nature of a problem situationcby providing longitudinal or recent accurate information about a decision situationd all of the above

Ans: aRationale: Qualitative techniques provide primary sources of data, usually geared toward helping refine and direct quantitative research. An explicitly historical perspective is ordinarily supplied by secondary data, often collected by a third-party provider over a period of time. Longitudinal data may render exploratory research unnecessary for the decision situation and is rarely collected using qualitative techniques alone.

Which of the following is a primary advantage of the use of scanner panel data over other methods of consumer data collection?a accurate longitudinal histories of household purchasesb allowance for shelf usage optimizationcbetter tracking of consumer awareness and attitudesd full demographics of panel households

Ans: aRationale: The main purpose of scanner panel data is to generate accurate histories for each household, including all purchases made and the store environment for each. Shelf usage refers to store data, which does not require a panel; attitudes cannot be tracked via household-level purchases alone; demographics would be part of any panel, not only scanner panels.

Consider the following question put to respondents:How do you rate product B regarding its daily usefulness?Not useful Only slightly useful Somewhat useful Fairly usefulVery usefulWhat characteristics does this scale have?a odd number of categories, forced, non-comparative, unbalanced, asymmetricb odd number, non-forced, non-comparative, balanced, symmetric codd number, non-forced, comparative, unbalanced, asymmetricd odd number, forced, non-comparative, balanced, symmetric

Ans: aRationale: There are five categories (odd); there is no option for "don't know" or "won't say," so the respondent is forced to take a position; there is no explicit or ideal alternative product mentioned, so it is noncomparative; the scale is unbalanced, with three categories in the positive range; and it is asymmetric, with no neutral midpoint relating no opinion on the usefulness of the product (i.e., "somewhat" is not exactly halfway between "not" and "very" useful)

1 What is the nature of the error generated by a non-probability sampling procedure?a The error cannot be measured in terms of degree or direction.b The error can be sufficiently minimized when the number of control factors is maximized (e.g., at least four).cThe error does not fall within the purview of the central limit theorem.d Non-probability samples are selected by knowledgeable researchers and therefore the error is typically negligible.

Ans: aRationale: We cannot measure sampling error when we use non-probability sampling. This means that we cannot identify the degree to which the sample is representative of the population. And we therefore cannot make conclusive statements about the sample.

You are screening people for a focus group on the market for ladies' razors among working women. Which of these candidates would make acceptable participants?a a woman who currently works as a bank teller and who recently emigrated from Chinab a woman of European ancestry with three children who works in her own home and has proven herself a competent focus group participantca Native American woman who works for Gillette and has never participated in a focus group befored a woman from the Caribbean who used to work as a store clerk and is currently training to become a dentist

Ans: aRationale: Woman b has participated in focus groups before and so is not a good candidate for a new focus group. Woman c works for a company that makes and is widely associated with razors. Woman d is not currently working, although she did recently and is training for a new profession. Recent emigration is not part of the screening protocol for this group, even if linguistic issues may arise; working as a bank teller presumably implies enough fluency to participate.

2Systematic error leads directly to which of the following in resulting measurements?arandom fluctuationsbconstant biascminor reductions of validityd major reductions of reliability

Ans: bRationale: An error that is systematic will affect measurements in the same manner (direction), leading to a consistent bias.

3 What is periodicity?a the regular pattern of error falloff in a natural distribution, as determined by the central limit theoremb error in systematic sampling due to the sampling pattern coinciding with a periodic pattern in the populationcerror in stratified sampling when the strata being sampled exhibit a periodic patternd the necessity in sampling for using a periodic pattern in order to achieve a truly random sample

Ans: bRationale: Periodicity occurs in systematic sampling when the period selected for achieving the random sample coincides with a periodic pattern in the population; for instance, every fourth house is a corner house and is more expensive, so a systematic sample of every eighth house does not randomly sample the population. Periodicity can be corrected for by rearranging the elements or choosing a different sampling procedure.

Put the following questions in the best order for a survey:w Please fill in: male/female ethnic group __________ yearly income __________xWould you consider a capital investment in your farm, if it resulted in greater profits?yWould you consider converting your egg farm to a factory for chicken pies, if it resulted in greater profits?zAre you frustrated by lower profits than you are capable of achieving?ax, y, w, zbz, x, y, wcz, w, y, xdw, z, x, y

Ans: bRationale: Question z builds rapport, is designed to interest the respondent, and is almost rhetorical in nature. Question x is more general than y, so it should come before it. Question w is standard demographic information, and so should be placed last.

2Which of the following is an improvement on this question: "What type of cat food do you prepare for your cat?" (assuming the respondent has a cat)a What brand of cat food do you prepare for your cat?b What do you feed your cat?cWhat does your cat like to eat?d none of the above

Ans: bRationale: The original question uses the word "prepare," which may be interpreted differently by different people (e.g., "cook," "purchase," "serve," "get ready"); option a shares this wording problem. Option c assumes that the respondent knows what the cat likes. Option b refers to the behavior of the respondent and so is more appropriate.

6 Which of these are advantages of depth interviews in marketing research relative to focus groups?a The success of depth interviews depends on the skill of the interviewer, whereas the success of focus groups does not.b It is easier to associate a participant's feeling with related beliefs in a depth interview.cDepth interviews are more like group therapy and allow participants to connect with their inner values, beliefs, and feelings.d Focus groups involve sampling the target segment and therefore give more accurate information on the market.

Ans: bRationale: The success of both depth interviews and focus groups depends in great measure on the skill of the moderator. Depth interviews allow better identification of the source of an attitude and exploration of the associated feelings and underlying reasons on an individual level, because the respondent's views are less affected by what others may have to say. Depth interviews are more analogous to one-on-one counseling than to group therapy, and connecting with inner values isn't always as helpful as having respondents react to one another. Focus groups do not involve methodical sampling of a target segment and therefore the results can never be projected to the market as a whole; moreover, one-on-one interview participants are typically selected in a similar fashion.

5Which of the following is commonly the largest source of preventable (by the researcher) error in carrying out marketing research?apoor sample designbmeasurement errorcnon-response errordsample bias

Ans: bRationale: While poor sample design, sample bias, and non-response can each lead to error, far greater errors can come about when a question does not measure the attitude it was designed to assess. Note as well that non-response errors and some elements of both sample bias and poor design are all difficult to prevent, since they are all contingent on the sample itself, while measurement error deals only with the question(s) being asked.

5 Which of the following abilities would be useful for a focus group moderator to convey in the context of running a group?a the ability to display encyclopedic knowledge of the product, its uses, and market perceptions of itb the ability to encourage groups to focus only on the specified topic by enforcing a disciplined structure for communicationcthe ability to notice when a few people are dominating and others are reserved or silent and to encourage better balance in discreet waysd all of the above

Ans: c Rationale: A moderator usually needs to convey a lack of information, not knowing so much about the product and its uses that participants are cowed, intimidated, or made to feel superfluous. An open, flexible conversation is needed, not an overly structured one in which participants may feel reluctant to share or are actively dissuaded from veering into new, useful discussion areas. Drawing out people who have not spoken much, rather than allowing a few people to dominate, is among the keys skill for a focus group moderator to possess and clearly convey

8 In selecting a communication method for collecting respondent data for quantitative research, which of the following is not always among the main considerations?a ability to reach the target sampleb potential bias of using specific mediacinterviewer skills

Ans: c Rationale: Reaching the correct target sample, attempting to minimize or eliminate various sources of bias, and achieving a suitable response rate to ensure statistical significance are each crucial elements of communication (e.g., survey, mail, and Internet) research. By contrast, with most of main forms of communication research (except for the telephone), interviewers are not involved

2 Why is the central limit theorem critical to assessing sampling error?a It allows us to identify whether the error is within 68 percent, 95 percent, or 99 percent confidence intervals.b It tells us that the error can be corrected if the actual population is known.cIt allows us to calculate confidence intervals for quantities of interest, such as the population mean.d It tells us that the error varies inversely with the sample size.

Ans: c Rationale: Using the central limit theorem, we can calculate confidence intervals for many quantities of interests, in particular population means, for any sample of sufficient size. And sampling error is an essential element in computing confidence intervals based on the formula: mean ± some number of standard errors

8 A mock-up of research findings using simulated data is valuable because:a it prepares managers for the experience.b it reduces the expense of collecting actual data.cit helps managers to notice omissions and develop decision criteria.d marketing researchers tend to be better at visualization than managers.

Ans: cRationale: Although mock-ups are helpful in many ways, they are most useful in noticing what might have been done differently, or information needed that wasn't collected, so that decision criteria can be developed in advance.

5 A causal model does what in marketing research?a It allows a cataloguing of sources of systematic response error.b It facilitates effective marketing choices by gathering evidence for conclusive research.cIt provides a basis for hypotheses and a means of incorporating evidence toward arriving at a sound final decision.d It is the statistical analysis input into an MDSS.

Ans: cRationale: Causal models allow researchers to assess which variables (under managerial control or not) affect quantities of interest. Note that answer b is not correct because it suggests that the causal model gathers evidence.

The variable cost of a new anti-depressant is $0.50 per pill. The market size is 600,000 pills, and a bottle of 20 pills sells for $35. If the firm decides that marketing research is justified up to 5%, what is the maximum available for marketing research?a $1,050,000b $51,750c$37,500d $17,143

Ans: cRationale: Contribution margin = ($35/20) - $0.50 = $1.25 / pill 5%*600,000*$1.25 = $37,500

Cross-sectional design relates to longitudinal design in what way?a Sampling across one dimension of data gives a different result than across an orthogonal (i.e., uncorrelated) dimension.b Cross-sectional research cannot avail itself of certain statistical methods, such as clustering or ordinal regression.cLongitudinal samples (e.g., panels) provide information about the same group of units over a period of time.d Survey research focuses more on attitudinal issues than longitudinal panel research.

Ans: cRationale: Longitudinal or panel research studies the same sample over time, whereas a cross-sectional design studies a sample at a single point in time.

Which of the following survey questions is most likely to achieve meaningful research results?aDo you support using $15 per taxpayer to fund the campaigns of political candidates?bHow long ago did you purchase your refrigerator?cHave you purchased canned vegetables from the supermarket in the last month?dWhen buying ice cream, how do you weigh the caloric content and healthfulness relative to the experiential quality of the flavor?

Ans: cRationale: Option a does not provide trade-off of valued options—for example, that tax revenues allocated to political campaigns reduce funds usable elsewhere. Option b presumes that the respondent not only knows the answer, but was in fact the one who did the purchasing. Option d uses ambiguous words and overly complex language and presumes that the respondent even makes this sort of trade-off. Option c has an objective, yes/no answer, based on fully specified behavior, and provides a time referent.

Which of the following objectives would be best served by a focus group?a to assess the likelihood of success of a new marketing (pricing, promotions, distribution) planb to test hypotheses regarding the demographics of good target markets for a productcto gather further information to interpret the results of prior quantitative researchd to decrease uncertainty about the demand forecast before the release of a new product

Ans: cRationale: Options a, b, and d all deal with specific, objective goals or hypotheses and, as such, call for quantitative techniques in a conclusive research project. Gathering information about the reasons behind an established conclusion (e.g., to generate hypotheses regarding why a certain brand of product is preferred by consumers) is a good use of a focus group; although focus groups are usually conducted before quantitative research is carried out, it is acceptable to use them to help interpret motivations, rationales, and "human factor" reasons left unaddressed by formal quantitative research.

3 What is the difference between the main testing effect and the interactive testing effect?a The main testing effect is the effect of the treatment being measured; the interactive testing effect is the distortion of this by communication among respondents.b The results of the treatment are the main and interactive testing effects, after they have been separated from distortion by extraneous variables.cThe main testing effect is the effect of one test on a subsequent test; the interactive testing effect is the effect of a test on the subsequent behavior being measured.d The main testing effect is the effect of instrumentation on the results; the interactive testing effect is the effect of selection bias.

Ans: cRationale: These are the definitions of these effects.

2 Which of the following is an accurate statement regarding the scientific notion of causation?a Concomitant variation, time order of occurrence, and elimination of other possible causal factors are required to prove causation.b Rigorous examination of data under specific conditions is necessary to substantiate scientific statements of deterministic causation.cAn observed effect is never caused by only one factor.d Causal inferences demand that the cause and effect be observed to vary together, the effect be observed subsequent to the cause, and other possible causes be ruled out.

Ans: d Rationale: Causation can never be proven, only suggested or supported. Deterministic causation is a nonscientific view that one factor always leads to an effect, rather than the scientific notion of probabilistic causation. It is no more scientific to say that causal relationships are impossible than to say that they have been proven. Causal inference requires that the criteria in option d be met, although they are insufficient to demonstrate causation on their own.

Which of the following is an example of a research design?a using syndicated secondary datab setting up a consumer panel using UPC scannerscstatistical analysis and modeling through an MDSSd identifying a plausible hypothesis regarding a marketing problem by interviewing experts

Ans: dRationale: A research design must describe the information to be collected, as well as the data sources. Answers a, b, and c only specify a data source or collection method, not the information to be collected. Answer d is an example of exploratory research.

6 A response bias isa a bias a respondent has about the surveyor.b an error caused by people who are too mobile or well-off to be interested in panel inducements not being selected for the study.calso called a "mortality rate," resulting from members moving or losing interest or other forms of attrition.d a tendency, conscious or not, to mis-report the degree of the behavior being monitored.

Ans: dRationale: A response bias can come about for many reasons (which the other choices conjecture), but deals with inaccurate reporting on the part of the respondent.

2 In which of the following ways are qualitative techniques distinguished from quantitative techniques used in marketing research?a Qualitative techniques give more "human" information regarding consumer attitudes and thinking.b Quantitative techniques are necessary to help achieve the decision-oriented rigor required of conclusive research.cQualitative techniques can aid the initial development of hypotheses to be tested later by quantitative methods.d all of the above

Ans: dRationale: All of the statements are accurate

A consumer packaged goods company runs many studies to test the flavor of various products and collects rankings from many people. It then averages the ranks across many people. For product class A, the averaged flavor rankings five brands came out in decreasing preference as follows: A3 > A5 > A2 > A1 > A4. For another product class, five other brands came out as follows: B2 > B5 > B1 > B4 > B3. Based on these averaged rankings, what can you conclude about relative flavor preferences?aA3 is liked about as well as B2.b A3 is liked better than B3.cThe overall liking for the five A brands is about the same as for the B brands.d none of these

Ans: dRationale: Averaged rankings can never be compared across groups. If category A is some form of medicine and B is premium ice creams, it is easy to see that none of these statements needs to be true.

Characteristics of conclusive research are that ita is concerned with making explicit predictions.b often seeks to objectively quantify and accurately measure attitudes, beliefs, and feelings.cinvolves communication with or observation of respondents and their behavior.d all of the above

Ans: dRationale: Conclusive research is concerned with making good decisions based on accurate measurements of marketing-relevant variables. All of these exemplify that overarching goal

1 How does a time-series experiment allow for the control of extraneous variables?a by keeping the extraneous variables uniform over periodic observationsb by randomizing each of the extraneous variables over timecby making the extraneous variables irrelevant due to the history effectd by taking multiple measures and statistically correcting for the extraneous variables over time

Ans: dRationale: It is impossible to ensure that extraneous variables are kept uniform, or nearly so, over time, because we have no control over them; nor can we randomize them. The history effect can complicate time-series experiments, not help in controlling extraneous variables. A time-series experiment attempts to control extraneous variables by measuring the same test units through periodic observations, thus allowing for statistical measurement (e.g., regression) of extraneous effects over time.

9 Which is a valid reason for undertaking a marketing research project?a The contribution margin and the market size are large.b A major decision has been made, and managers wish to be able to justify the decision.cA top manager is anxious and needs assurance about courses of action. d A manager is uncertain about a decision and wishes to decrease this uncertainty.

Ans: dRationale: Marketing research information helps reduce uncertainty in a decision that needs to be made; it should not be undertaken simply because the market is large, one needs justifications, or due to anxiety.

7 Which of the following is a problem? In the past 6 months:a sales revenue has increased half as quickly as projected.b sales revenue has decreased 20% overall.ccustomer satisfaction with the company has decreased 5%.d a new competitor is charging a lower price on quality-matched items.

Ans: dRationale: The other three are symptoms, not problems.

Exploratory research A.identifies the sampling frame and sampling procedure.B.aids the early stages of the decision-making process and investigates how to hold down cost and time expenditure.C.provides information that helps the manager evaluate and select a course of action.D.is essential to ensure that marketing programs are proceeding in accordance with preset plans.

B.aids the early stages of the decision-making process and investigates how to hold down cost and time expenditure.

All of the following would be considered advantages of open-ended questions exceptA.can serve as an excellent first question on a topic B.are relatively easy to code and analyze C.can establish rapport and gain the respondent's cooperation D.exert minimal influence on subsequent responses, because respondents are not biased by predetermined set of response

B.are relatively easy to code and analyze

Systematic error leads directly to which of the following in resulting measurements?A.random fluctuationsB.constant biasC.minor reductions of validityD.major reductions of reliability

B.constant bias

•When comparing laboratory experiments with field experiments, field experiments offer an advantage over laboratory experiments in terms of A.internal validity B.external validity C.costsD.speed

B.external validity

The depth interview is used primarily A.for conclusive research B.for exploratory research C.to test specific hypotheses D.to reduce costs when focus groups are not deemed necessary

B.for exploratory research

The outcome of the manager's final decision is limited by: A.the overall cost of the marketing research.B.the quality of courses of action considered.C.the nature of the performance measures leading to the research.D.all of the above.

B.the quality of courses of action considered.

The standard error of the mean isA.The degree of variation in the population B.The degree of variation in a sampleC.Estimated variation of the sample meanD.None of the above

C.Estimated variation of the sample mean

With simple random sampling, each element of the population A.has a known chance of being selected that is not necessarily equalB.has an equal chance of being selected that is not necessarily knownC.has a known and equal chance of being selectedD.has an unknown chance of being selected because the sample is designed through quotas

C.has a known and equal chance of being selected

•Qualitative research techniques can be identified by all of the following characteristics exceptA.small convenience or quota sample B.information sought relates to respondents' motivations, beliefs, feelings, and attitudesC.objective scientific approach to data gathering D.data collection format is open ended

C.objective scientific approach to data gathering

Conclusive research A.identifies the sampling frame and sampling procedure B.aids the early stages of the decision-making process and investigates how to hold down cost and time expenditure C.provides information that helps the manager evaluate and select a course of actionD.is essential to ensure that marketing programs are proceeding in accordance with preset plans

C.provides information that helps the manager evaluate and select a course of action

Which is a valid reason for undertaking a marketing research project?A.The contribution margin and the market size are large.B.A major decision has been made, and managers wish to be able to justify the decision.C.A top manager is anxious and needs assurance about courses of action.D.A manager is uncertain about a decision and wishes to decrease this uncertainty.

D.A manager is uncertain about a decision and wishes to decrease this uncertainty.

•Which of the following is an accurate statement regarding the scientific notion of causation?A.Concomitant variation, time order of occurrence, and elimination of other possible causal factors are required to prove causation.B.Rigorous examination of data under specific conditions is necessary to substantiate scientific statements of deterministic causation.C.An observed effect is never truly caused by one factor.D.Causal inferences demand that the cause and effect be observed to vary together, the effect be observed subsequent to the cause, and other possible causes be ruled out.

D.Causal inferences demand that the cause and effect be observed to vary together, the effect be observed subsequent to the cause, and other possible causes be ruled out.

With non-probability sampling, each element of the population A.has a known chance of being selected that is not necessarily equal B.has an equal chance of being selected that is not necessarily known C.has a known and equal chance of being selected D.has an unknown chance of being selected

D.has an unknown chance of being selected

The first step of sampling is to define the population. The second step is toA.physically select the sample B.mock up likely and extreme data values C.decide on a sample size D.identify the sampling frame

D.identify the sampling frame

The _______________ section of the questionnaire forms the major, and invariably the longest portion of the questionnaire.A.identification data B.request for cooperation C.instructions D.information soughtE.classification data

D.information sought

How might the following organizations effectively employ marketing research?a senatorial campaign

Determine demographic characteristics of segments of supporters.Test appeal of various messages to different segments.Target likely voters for Get-Out-The-Vote campaign based on consumer preferences.

Which is TRUE?A.Stratified samples are homogeneous within groupsB.Cluster samples are homogeneous between groupsC.Stratified samples are heterogeneous between groupsD.Cluster samples are heterogeneous within groupsE.All of the aboveF.None of the above

E.All of the above

What are some of the desirable characteristics of focus group moderators? How might a researcher go about ensuring that potential moderators are appropriate for a specific project, over and above general qualifications?

Moderators should be themselves knowledgeable about the subject, yet convey a sense of incomplete understanding in order to draw out clarification from the respondents. In addition to anticipating where the discussion is designed to go and may go, moderators should be able to respond creatively and flexibly to the current situation. Researchers interviewing potential moderators might, for example, engage the candidate in role play, ask the candidate to review issues raised by the moderator guide and explain how they would approach the subject, or check for the candidate's experience in group facilitation. They should also consider whether a particular gender is preferred for the study and whether the same moderator should lead multiple sessions.

Under what circumstances is it extremely difficult or even impossible for even the best design to control an extraneous variable?

One of the basic principles of experimental design is that all extraneous variables operate equally on all treatment and control groups. To the extent that they do not, even the best design will fail to control extraneous variables. It is virtually impossible to control all extraneous variables in a field experiment.

Discuss the issues involved in determining when and where an automobile tire manufacturer should conduct focus groups to get feedback from customers on its line of winter tires. Give a brief description of customers who would be good candidates for this type of focus group. What kind of questions would you use for screening?

OverviewThe firm must be clear about the purpose of this qualitative research. If the manufacturer will eventually, through further quantitative study, decide on which products to market or the location of its distribution centers, then the moderator's guide should appropriately reflect the options that will be under consideration.Physical LocationThis is probably not the best topic for an online group, as "tire owners" covers a much broader demographic base than computer owners or internet savvy people.To determine where the focus group should be held, it is important to again consider the purpose of the research. The answer will be different, for example, if the tire maker wants to find out about attitudes from people who do not use the tires very much than if it wants to find out about attitudes from people who live in cold and snowy environments and use winter tires all the time. We will suppose for the sake of this example, that the tire maker wants to consult people who use winter tires extensively. In this case, a group in the U.S. should probably be held in snowy and cold regions such as North Dakota or parts of Colorado.TimingThe group will probably yield best results if it is held just after the end of winter. At this point in the year, the participants will have used winter tires through the entire winter and will have a good idea of what they like and do not like regarding their tires.Screening ParticipantsGiven the premises that the tire maker wants to talk to people who use winter tires extensively, and that participants should be relatively homogenous, the focus group could, for example, consist of 30- to 50-year-old heads of household in the income range of $60,000-$80,000 per household who own a pick-up truck or van. Screening questions should probably deal mainly with vehicle or tire experience, preferably among those with a reasonably long purchase history in both categories. Questions screening out professional focus group participants are also permissible. This homogenous group should give the client a good idea of what extensive users think of the line of winter tires, because they have both the experience with the tires and the disposable income to be purchasers.

A researcher wishes to test consumer response to a new jingle for Fruit Roll-Ups. How can the effects of the extraneous variable of age best be controlled?A.Physically control the variable by testing only consumers in a narrow age bracket.B.Randomize age in the sample groups.C.Design the experiment to avoid the maturation effect on the test units.D.None of the above will work without a time-series experiment.

Physically control the variable by testing only consumers in a narrow age bracket.

What decisions precede the questionnaire design stage? What criteria govern the inclusion of questions in the questionnaire?

Prior decisions include:1. Research design2. Data sources3. Selection of target population4. Sampling plan5. Communication method6. Mode of data analysis (visualization of research findings)The criterion governing inclusion on the questionnaire is that each question should relate to a specific information need. Otherwise, the questionnaire will be lengthier than necessary, making it less efficient in terms of time and costs.

What problems are likely to occur in implementing a field study using simple random sampling?

Problems with simple random sampling may include:Getting a complete list of sampling elementsRandom selection resulting in choices that aren't feasible (e.g., huge geographic jumps)Lack of control of other variables that may bias results (e.g., non-response error)Random selection may not render sufficient members of a subgroup of interest

2 Which of the following is not a formal hypothesis?a Improving the quality of the product by using rubber instead of plastic will increase the percentage of repeat buyers by 20 percent.b A consumer panel will test how consumers respond to a 25 percent increase in advertising frequency.cA television campaign emphasizing the achievements of candidate X will result in an 8 percent boost in the polls.d Moving the root beer display nearer to the ice cream freezer will result in a sales increase of both products.

Rationale: A hypothesis is a firm statement that is testable as either true or false. Answer b does not postulate a hypothesis regarding consumer response.

What distinguishes marketing research from what is usually referred to as market research, as well as other sorts of social science research?a the application of rigorous methodologyb focus on managerial decision making, broader than markets and consumerscthe systematic collection of multiple, different types of datad the overall approach to data processing and statistical analysis

Rationale: All forms of research, if carried out well, rely on rigorous methodology, data processing, multiple data types, and statistical analysis. Marketing research is distinguished by the emphasis on supporting the decision-making process.

Why is the analysis preceding the decision to undertake research so crucial to the success of the project?

Research is costly and time-consuming. It is only justified when its benefits in terms of reducing uncertainty in a decision situation outweigh its costs. If analysis is skipped prior to research or done poorly, the research budget may be spent on research that is unnecessary or unlikely to aid decision-making.

What is the role of secondary data in the research process? What are the advantages and disadvantages of secondary data relative to primary data? How can they be used in concert to arrive at quicker, more accurate results than relying on either alone?

Secondary data can provide more easily available, lower cost information than survey research. They can aid in the formulation of the decision problem, suggest methods, serve as a source of comparative data, and reveal details that may facilitate the design of proposed research. Disadvantages are that the data often don't exactly fit the information needs of the project, that the data are often aggregated without the breakdown required by the project, that the accuracy of the data is difficult to evaluate from available sources, and that the data may be outdated.

MINICASE: Suppose you are researching attitudes about "smartphones" (those that have web-browsing and other advanced features like GPS, music and video playback, and an interactive screen) among your classmates. Specify the information needs for this project, and then develop a concise questionnaire to measure the target group's perceived needs, attitudes, and purchase intentions. Be sure to consider that knowledge, purchase readiness, and opinions themselves are all likely to be heterogeneous, even in this self-selected group (in fact, what is considered a smartphone may be heterogeneous as well). How might you use the resulting data set to determine which peripherals are most likely to be needed, valued, and purchased?

Students responses should analyze the research objective (to determine which smartphone peripherals are most likely to be needed, valued and purchased) and from that determine specific information needs. Questions should be designed to meet those information needs; each question should correspond to one of the information needs. Attention should be given to avoiding bias in the structure of the questions and to providing answer categories and scales appropriate to the question and the information needed.

Suppose that your role in a research project were to perform statistical analyses and help explain them for the research report. You realize later that, of the many dozens of analyses run, only those explicitly supporting a costly departmental expansion—and not those that either failed to support it or called it directly into question—were included in the final report put together by other members of the research team to be read by upper management. What might you do, and how would it affect how you carried out future analyses?

The choices include keeping silent in deference to fellow researchers, confronting this issue with those responsible, bringing it to the attention of a trusted manager, and in some cases even whistle-blowing to the outside world. Options for the future might include greater vigilance in noticing such a situation earlier, building connections with those involved in decision-making in order to educate them on the dangers of such a course of action as well as the value of impeccable marketing research, greater involvement in the writing of future reports, and leaving the company if its standards for behavior remain at an unacceptable level.

How does the respondent affect the content of the questions? Can respondents influence the forms of questions as well, or the range and wording of choices? In which ways?

The definition of the target population affects the design of the questions from the start, so that questions and answer categories cover the range of views within that target population. The design of structured questions is catered to participant input in exploratory research and testing in order to ensure that the explicit categories cover the vast majority of participant choices. If the range or wording of choices is found to bias responses, they must be re-scaled or reworded to eliminate or reduce that bias.

How can a design described as R O1 X1 O2be confounded? What design could control the confounding variables?

The experiment could be confounded by all sources of invalidity except selection bias. The extraneous variables (history, maturation, testing, instrumentation, statistical regression, and test unit mortality) could be controlled by adding a control group. The interactive testing effect could be controlled with either a second control group or doing a post-test-only control group design.

What factors determine the relative importance of managerial experience versus marketing research in a given situation?

The factors determining the relative importance of managerial experience versus marketing research in a specific decision situation include how relevant the manager's experience is to the situation, how familiar the situation is to the manager, the degree of uncertainty in the situation, and whether a decrease in the uncertainty of the decision is worth the cost of a marketing research project. Recurring, familiar decisions involve little uncertainty, so managers can generally rely on their experience and judgment. In unfamiliar decision situations, the level of uncertainty and the size of the risk are weighed against the costs of reducing the uncertainty through marketing research.

Kraft Foods does some mall-intercept interviewing to evaluate new food product offerings. These samples are typically done in only a few malls for any one test. Subjects who pass by particular locations in the mall are recruited on the basis of appearing to match certain demographic characteristics: gender, age, and so on.a What sampling procedure is being used? Evaluate the sampling procedure used in terms of its appropriateness to the stated task and objectives. What inferences to the broader population can be drawn? d Why has mall-intercept sampling become so widely used in practice? What biases may exist due to the sample selection method?

The procedure being used is quota sampling based on demographic characteristics. (Convenience sampling is also an acceptable answer.) The sampling frame is the group of people shopping at the malls where the study is conducted. Although the sampling procedure may result in a sample that mirrors the target population in desired demographic characteristics, it would not have enough validity statistically to form the basis of any major decisions. The sample consists of the first people matching the desired demographic characteristics who are approached by field research personnel and who agree to participate in the study. Inherent in this procedure are selection bias (the subjects are probably not randomly selected, because the researcher may perceive some potential subjects as being more approachable than others) and response bias (subjects who decline the offer to participate are likely to be different from those who do not). Also, because the test is conducted in "only a few malls," a truly random sample of the population cannot be obtained.If the purpose were to conclusively support or contradict a specific management decision regarding new product offerings, the sample would not be appropriate. If it is to explore hypotheses as part of an exploratory phase leading to a conclusive study, it may be appropriate, depending on the nature of the hypothesis and its fit with the group actually sampled. None. Again, if the decision being investigated is low enough impact, inferences may be made with the knowledge that large error could be involved. For example, a laundry detergent manufacturer might test which of two fragrance brands consumers preferred via mall intercept of adult women. The winning fragrance might be inferred to apply to American women in general, with the knowledge that a relatively small decision is at stake if mistaken. Although mall intercept research lacks statistical validity, it is widely used in practice, because:It is relatively low cost.It allows researchers to gather a great deal of data from a wide variety of consumer demographics, with many at the appropriate stage of purchase readiness.Researchers often know from experience how test results translate into "real world" consumer behavior, despite questions about external validity. For example, they can make adjustments to figures for stated purchase intent that provide accurate predictions of actual trial rates. Some biases include: The initial judgment of the interviewer biases selection.People willing to stop and take the survey differ from those who decline to participate.The type of people who shop in a mall are usually within a certain middle range of income levels.More women than men frequent some malls.People with children are more likely to shop on certain days of the week.Heavy shoppers will be overrepresented.The interviewer himself may influence the participants through word choice or approach.

How can a researcher overcome the problems associated with collecting data about events that are unimportant to respondents or that occur infrequently?

The researcher may either choose to select respondents most likely to remember (e.g., recent purchasers) or may stimulate memory with the aided recall.Further Probe: What problems may be associated with asking recent purchasers a survey question?Probe Response: A potential problem with seeking recent purchasers of a product or service is that these people have not seen the product or service through its entire life. In addition, persuasive techniques are often used by merchants to reassure a buyer, especially of large purchases such as cars. Such biasing may not yield a full spectrum of opinions.

Can you think of examples of unneeded research?

Unneeded research (called "pseudo-research" in the text) means research that does not provide information for a decision, including situations where a decision is pre-determined, where research is conducted that gathers data that is not relevant to a decision situation, and where research is conducted when decision criteria have not been defined in relation to the data to be collected.

General Motors does some evaluations of car designs and advertising themes using focus groups. These groups consist of from eight to ten potential customers who match the target segment profile for the car being evaluated. The focus group interviewees are selected by a research firm for General Motors through telephone solicitation of people who are qualified as matching the required target profile. This qualification of participants is done by a series of questions asked over the phone. The phone directories used to identify which individuals to call are from geographic areas where cars such as the one being evaluated have historically sold well. What sampling procedure is being used? Evaluate the sampling procedure used in terms of its appropriateness to the stated task and objectives. What inferences to the whole target segment population can be drawn from this sampling approach? What biases may exist due to the sample selection method?

a. This is a judgment sample with the phone book as a sampling frame. Convenience or quota sampling are also acceptable answers. creening procedure:Subjects will match the target segment on attributes questioned, but not necessarily on other variables. As long as researchers' assumptions about variables important to automobile purchase decision are correct, this is acceptable.Sampling frame:Many people use cell phones and internet-based calling instead of land lines that are listed in the phone book. The sampling frame also omits people with unlisted numbers and people without telephones. This is far from ideal, but the sample may still be fairly representative of the target segment.Geographical issues:Researching consumers in regions where GM cars have traditionally sold well does not allow the firm to obtain any information about selling in new regions.As long as researchers are aware they may be speaking with respondents with higher purchase likelihood than the general population (even with the required target profile), and selection biases are corrected for, this sampling plan would still be appropriate; this is largely because focus groups thrive on motivated, tightly-targeted participant groups. c)this procedure is acceptable for obtaining feedback on automobile design and advertising from a group of consumers defined by specific variables and residing in a particular geographic area. However, one should question how applicable this group's opinions are to other demographic or geographic segments. d.Some key biases include: people willing to participate in a focus group may differ from those who would chose not to people who rely solely on cell phones or internet calling, people with unlisted landlines, and people without phones are omittedphone directories inherently contain errors and are not accurate to the day the participants are recruitedthe area being surveyed is a historically good market for GM cars, so opinions of other geographic regions are not represented

Over which independent variables does a company exert control?a marketing mixb performance measurescsituational variablesd all independent variables

ans: aRationale: Performance measures are variables dependent on the behavioral response. Situational variables are not controllable. Independent variables include both the marketing mix and the uncontrollable situational variables.

Which of the following data are most likely to be considered information for someone anticipating voting in an upcoming election?a the number of people voting, according to recent pollsb the prior occupations of the candidates' staffcthe candidates' stated platformsd the countries that candidates have visited recently

ans: c Rationale: Information refers to data which reduce uncertainty in decision situations. Of the listed data sources, the platforms of the candidates are most directly relevant to the decision of whom to vote for.

Which of the following is not an example of statistical reasoning?a What are the chances of that happening?b Candidate B received 16 votes out of 19cWe're more likely to get what we want if we don't compromise up front.d Is there enough interest in this idea to be worth going ahead with it?

b Rationale: : Statement b merely states data, without using any kind of statistical reasoning regarding that data. Question a asks for an evaluation of the probability of a certain event happening. Statement c interprets the chances of a certain outcome based on one course of action. Question d asks for an evaluation of a course of action based on available data regarding the proportion of people interested in that course of action. See Marketing Research Focus 1.1, What Is Research?

Which best describes the history effect? A.events concurrent with the experiment that may affect the dependent variable B.effect of test units being selected for treatment on the basis of an extreme pre-treatment score C.effect of changes in the test units over timeD.effect of treatment groups differing on the dependent variable prior to the presentation of the treatmentsE.effect of taking a measure before treatmentF.changes in the observers or calibration of the measuring instrumentG.effect of test units withdrawing from the experiment while it is in progress

events concurrent with the experiment that may affect the dependent variable

Should the Census Bureau simply replace the process of taking a census with a well-defined series of samples of the population

robe Response: No matter how well sampling is conducted, it is never exhaustive, which is a census' defining characteristic. Sampling is meant to cut down on the time and effort required of researchers, who can then extrapolate, using formal statistical methods, to the entire population. But the error resulting from this process can be quite high, depending on the nature of the sample. Census results provide an exhaustive check and point of comparison for samples, so they should be continued. Censuses and sampling can therefore be seen as two tools that can serve as reality checks for one another.

Is it appropriate for the U.S. Census Bureau to agree to adjust the census count on the basis of sample results? What are the pros and cons?

t is appropriate for the Census Bureau to adjust figures that are known to be understated, such as the nation's homeless population. The sample must be fairly representative of the census population. A small sample of New York City, for example, is probably representative of that city as a whole but not of the country as a whole. As long as the sampling plan is carefully designed so that a representative sample of the censused population is obtained, the adjustment process will result in a more accurate and fair representation of the undercounted population.A complete census has higher potential for non-sampling errors, while an unbiased sample can minimize both sampling and non-sampling errors. So adjusted census counts may also be more accurate for other kinds of updates and interpolations.

Longitudinal designs allow for better distinguishing a treatment effect from underlying trends over timeA.TrueB.False

A.True

A nominal scale (not ordinal, or interval) would be appropriate for the representation of: A.ZIP codesB.simple preferencesC.ambient temperatureD.age ranges

A.ZIP codes Rationale: ZIP codes use numbers for naming or cataloguing only, with no ranking or higher-scale properties. Preferences and age ranges, which have an intrinsic order, should be represented by at least an ordinal scale; temperature, which measures underlying atomic motion, can support at least an interval scale and a ratio one if absolute zero is mapped to the zero point, as in the Kelvin scale.

Examples of inputs (independent variables) and outputs (dependent variables) studied in marketing research are, respectively:A.marketing mix; profits.B.performance measures; pricing.C.behavioral responses; situational factors.D.government regulation; economic climate.

A.marketing mix; profits.

What is the difference between a true experiment and a quasi experiment? A.A true experiment randomly assigns units B.A quasi experiment randomly assigns units C.They are not different, true and quasi experiments randomly assign units

A true experiment randomly assigns units

A _______________ may include an assessment of the competition and the internal resources of the company. A.demand analysis B.performance measure C.distribution assessment D.situational analysis

D. Situational analysis

How might the following organizations effectively employ marketing research?Dell Computer

Get feedback on new market segments.Evaluate preferred trade offs of memory (RAM) versus screen size at given price.Determine appropriate pricing levels.Test advertising copy.Test various sales techniques.Test product configuration preferences (hardware and software).

According to the text, what kind of educational or work background would you prefer when engaging the services of a focus group moderator?

Given that the position of a focus group moderator requires considerable facility in dealing with people and understanding group dynamics, background as a psychologist may be preferred, probably with supplementary skills in group facilitation through special training.

•Which best describes selection bias ? A.events concurrent with the experiment that may affect the dependent variable B.effect of test units being selected for treatment on the basis of an extreme pre-treatment score C.effect of changes in the test units over timeD.effect of treatment groups differing on the dependent variable prior to the presentation of the treatmentsE.effect of taking a measure before treatmentF.changes in the observers or calibration of the measuring instrumentG.effect of test units withdrawing from the experiment while it is in progress

.effect of treatment groups differing on the dependent variable prior to the presentation of the treatments

To construct an approximate 95% confidence interval, how many standard errors should be subtracted/added to the sample mean?A.1B.2C.3D.4E.None of the above

2

9 Suppose your research objective was to determine the potential market for a dog-themed work organizer. Specifically, you are to determine the probability that dog-owning office workers would purchase such an item. Which of the following pieces of secondary data would enable you to come up with such an estimate?a income profiles of dog-owning office workersb probability that a dog owner would buy such a productcprobability that an office worker would buy such a productd all of the above e none of the above

Ans: eRationale: Secondary data that is highly aggregated cannot be broken down so that more specific research needs can be met. The buying habits (i.e., probability of purchase) of dog owners and that of office workers may bear no relation to the buying habits of the conjoined group (dog-owning office workers). It would be wrong to merely multiply these probabilities, even as an estimate. Knowing income profiles (choice a) for the conjoined group would not suggest anything about purchase likelihood or the overall market size.

The sample mean isA.The average of all population unitsB.The average of all sample unitsC.The number of all sample unitsD.The degree of variation in the sampleE.None of the above

B.The average of all sample units

The sample standard deviation isA.The degree of variation in the population B.The degree of variation in a sampleC.Estimated variation of the sample meanD.None of the above

B.The degree of variation in a sample

Over which independent variables does a company exert control? A.marketing mix B.performance measures C.situational variables D.all independent variables

C. marketing mix

How might the following organizations effectively employ marketing research?a the census

Compare growth rates of various ethnic groups.Evaluate how accurately families without fixed residences are being represented in the census.Analyze how measured economic activity in various sectors relates to measured population migrations.

Which environment would be more favorable for a candy manufacturer looking to determine what design would be most appealing on their candy packages? Why?

Considering that the design of a candy package can easily be shown to a respondent in a laboratory, the laboratory would be more cost-efficient and faster. Given the infrequent purchase interval and unpredictability of candy purchases, a field experiment might be more difficult than it would be worth.

How might one choose between laboratory and field environments? What are the distinct advantages of each over the other in the course of an entire marketing research project, and at which stages would each be especially valuable?

Laboratory settings offer better control over confounding variables and thus have high internal validity; however, the results obtained in such a setting are often difficult to generalize, resulting in low external validity. Laboratory experiments usually involve fewer subjects and are generally simpler and more tightly defined graphically, resulting in lower cost and shorter time requirements than field experiments.Field experiments are conducted in actual market conditions, providing little control over confounding variables, but their results are more easily generalized. Thus field experiments are important to establish external validity.

The Wool Producers Board of New Zealand wants to stimulate the primary demand for wool in the world. What marketing research could the board do to facilitate the development of such a primary demand stimulation campaign? Describe a program of marketing research in detail. Explain how potential errors in marketing research will be controlled in this program.

Key areas that need to be addressed: global marketing research, error minimization, and developing a program using the marketing research process. Types of marketing research that could be performed in this case include trend analysis and demographic analysis in each region of the world under consideration through investigation of secondary data; exploratory research of both major distributors and important consumers groups for wool regarding consumer perceptions about and uses for wool, and possible messages for ad campaigns; and conclusive research regarding alternative courses of action. This type of research would lead researchers to the factors influencing the purchase and use of wool, including how wool might be substituted for other fabrics in wide current use. With these factors identified, marketers could then focus on using this knowledge to stimulate demand for wool.The error control plan should consider both sampling and non-sampling sources of potential error, particularly for any proposed primary research (for example, undersampling of consumers who are not currently purchasers of wool-based products).The following is just one example of various issues involved in creating a program for marketing research in this case. 1. Recognize and define the problem. In this case, the problem is finding a way to stimulate the primary demand for wool in the world, especially so in areas where utilization is below expectations. Further definition is needed to identify what areas of the world have the potential to use wool (e.g., areas with continual warm climates will have far less demand for wool than cooler regions).2. Determine the research objectives and information needs. To stimulate the demand for wool, research must determine who uses wool, why they use it, and how they use it. Information needed therefore includes demographics and industry trend analysis on a global scale.3. Data sources include interviews in various countries with industry experts, focus groups, and access to reference information regarding consumer buying patterns (both from wool manufacturers and from reports done by outside firms). Data collection procedures can be complex when performing marketing research on a global scale and when surveying both end-users and other businesses. Researchers will have to use the expertise of local firms and people to plan the collection of the data.4. Sample design will have to be planned carefully since there are many vagaries and variables in the sample population on a global scale. Segmented research may have to be performed.5. Collect, process, and analyze data.Potential errors in this marketing research program could result in sampling errors and non-sampling errors. To minimize errors, it will be important to carefully regulate the data collection process (especially since data from various sources around the world will be collected), provide standards for the various procedures used in analyzing and coding the collected data, and provide verification procedures to ensure that the data are accurate.6. Present results (oral presentation and/or written report)

How might the following organizations effectively employ marketing research?American Consumer Satisfaction Index

Measure consumer attitudes toward specific companies and products.Assess trends in consumer satisfaction with a brand over the past decade.Compare various companies in an industry on specific satisfaction measures.

When designing multiple-response questions, response alternatives should be _______________, which means the set of responses should include all of the possible alternatives. •collectively exhaustive •mutually exclusive •position unbiased •label inclusive

collectively exhaustive

Under what conditions would dichotomous questions be inappropriate? What sorts of questions would be indicated instead?

dichotomous question is inappropriate when a large percentage of the respondents sincerely hold a neutral view on the question or hold an opinion that does not fit the dichotomous framing. If there are more than two grades of response present in the respondent group or indecision predominates, dichotomous questions may yield results which contain substantial measurement error. In such situations, a question with a 5-point scale or a wider range of answer categories would be more appropriate.

Which best describes the maturation effect? A.events concurrent with the experiment that may affect the dependent variable B.effect of test units being selected for treatment on the basis of an extreme pre-treatment score C.effect of changes in the test units over timeD.effect of treatment groups differing on the dependent variable prior to the presentation of the treatmentsE.effect of taking a measure before treatmentF.changes in the observers or calibration of the measuring instrumentG.effect of test units withdrawing from the experiment while it is in progress

effect of changes in the test units over time

How does one choose between these two environments?

esponse: A laboratory environment is preferable when internal validity is the primary concern, whereas field environments are preferable when external validity is the main issue. Internal validity is more appropriate when a causal inference needs to be confirmed or contradicted, and the experimenter needs to rule out the effects of extraneous variables. Lab data also allows manipulation of variables not available in the real world, like the existence of prototype products, services yet to be introduced or previously unachievable combinations of attributes (e.g., "thin yet durable"). External validity is more appropriate when a relationship supported by other means (e.g., pre-tests, expert opinion, previous laboratory experiments) needs to be extended or generalized to real world applications. External data are preferable when a single result needs to be measured based on a complex set of input variables, like when testing a new promotional plan or conducting a full-scale test market for a new product.Ultimately, the choice falls to the judgment of the experimenter who must consider trade-offs between the factors of internal validity, external validity, cost, and time to arrive at a design that best satisfies the requirements of the given situation. Often external validity is tested after establishing internal validity. For example, pre-test of a magazine advertisement in a magazine might be done in a laboratory where the physical environment and distractions could be controlled by the researcher; the pre-tested advertisements could then be run on a small scale in a field experiment to gauge consumers' response to those ads