MBA Marketing Research Project Guidelines

Published on May 2016 | Categories: Documents | Downloads: 34 | Comments: 0 | Views: 295
of x
Download PDF   Embed   Report

research project

Comments

Content

1

MBA Marketing Research Project Guidelines This document communicates the guidelines and expectations for the marketing research project. Additional details may be provided depending on the nature of your particular project. A Consumer Behavior Focus The vast majority of marketing research dollars are spent in an effort to describe, explain, and/or predict consumer behavior. The reason is simple – consumers are a complex phenomenon – far more complex than any business model, financial market, tax code, supply chain, or organizational chart. The truth is that most companies and organizations are terrified that they don t really kno! !ho their customers or potential customers are or !hat these customers "really# !ant. There is an underlying fear that competitors understand the customer better and, in doing so, !ill serve them better. $eeping up !ith consumers is hard. Their perceptions, emotions, beliefs, and preferences are constantly evolving as they navigate myriad cultural, social, spiritual, physical, and cognitive environments. %o degree of logic, expertise, or experience can provide a magical portal into their hearts and minds. &f !e hope to create and nurture relationships !ith consumers, !e must stay close to them, interact !ith them, listen to them, and learn ho! to respond to their needs appropriately. 'nly research can provide organizations !ith the re(uisite information. Accordingly, !e shall focus our projects on elucidating aspects of consumer behavior. )ome projects may have a more applied focus, such as understanding ho! consumers respond to online consumer revie!s or ho! gift cards impact spending behaviors. 'ther projects may focus more on improving the tools !e use to try to understand consumers *i.e., "research on research#+. ,esearch firms and departments routinely conduct both kinds of research, as managerial projects pay the bills and "research on research# creates and enhances one s ability to produce valueadded insights in managerial projects. .ost, if not all of your projects !ill execute an experimental or (uasi-experimental methodology and !ill employ some form of (ualitative and/or archival research. */e !ill do (uestionnaire development as a class+. The types of data analysis used may vary significantly from project to project depending on the research (uestions at hand. I. Overview o Project Activities 0elo! is an overvie! of the activities that comprise the project. This is not an outline of the project report but rather an organized list of things that !e !ill be doing to produce the report. &t is intended to give you a sense of the scope and nature of the project. A. Specify the research problem and objectives. • All good research improves our kno!ledge by filling a gap, enhancing accuracy, or establishing boundary conditions. Thus, our first task is to identify the specific !ays in !hich our current kno!ledge is deficient. &n other !ords, !hy !ould our client/manager !ant to allocate precious and scarce resources to our project1 • 2aving identified a kno!ledge deficiency, !e need to propose a solution for addressing

2

it. This is a focusing task. 'ne of the most common problems in research is a lack of focus, !hich leads to trying to do too much. /hen !e find ourselves spending a lot of time !ondering !hat the right methods and measures are, it is a sure sign that !e lack focus. • ,emember that the purpose of research is not to make decisions but rather to inspire decision makers. &f a decision maker !ants us to produce research and make decisions, !e are no longer researchers but instead management consultants and should expect to be compensated at a higher rate. B. Design the key study or studies. • All research ultimately seeks to describe, explain, and/or predict something. This "something# is typically an outcome or effect that is of managerial interest. 3or example, a manager !ho !ants to make a decision about the type of background music to play in a retail space may hypothesize that "higher tempo music makes people !alk faster, !hich leads to less time for shopping, !hich leads to less time for deliberative decision making, !hich leads to lo!er sales of high involvement products.# To aid the manager, !e may seek to design a study in !hich !e measure shopping time, product type, and sales. • 4ust as !e seek to observe or measure outcomes of interest, !e are also interested in determining the impact or influence of marketing decision variables and other "causal# factors. 3or example, in the preceding example !e might systematically vary the tempo, volume, and/or genre of the music to better understand consumer responses. &n some cases !e can not actually vary a factor *e.g., gender+ but !e may still think about it as a potential causal factor of interest. • There are many !ays to carry out the same research idea. Thus, !e must make choices and specify a particular study procedure. That is, !e must figure out the logistics of running the study. &t should be very clear in our minds ho! the study !ill transpire, moment by moment. As multiple persons !ill probably be gathering data, it is important to ensure uniformity of the protocol. /e also need to understand or anticipate the influence of various "contextual# factors that are not of focal interest but nevertheless may influence the outcomes of interest *e.g., lighting, (uestionnaire !ording or modality, researcher-respondent interaction, etc.+ C. Design the collection method (i.e., study materials . • Typically, the data collection methods !ill contain both the stimuli *i.e., !ords, pictures, sounds, or objects that !e create and/or muster in order to vary causal factors of interest+ and the measured outcomes of interest. /e must think very carefully about all of the data that is re(uired to test our ideas, including various participant factors *e.g., demographics, psychographics, individual differences+. • 4ust as important, !e must think about the analyses !e are planning to run. &f !e don t measure the responses in the right !ay, !e may not be able to analyze them or !e may be forced to use crude techni(ues that only partially ans!er our (uestions. D. Sample the relevant population. • As best as !e can, !e !ill administer our (uestionnaires to people that are appropriate for the study at hand. &n most cases, a minimum of 56-788 respondents is necessary to have a chance to extract insights. )ome study designs may necessitate a larger sample.

3

&f using online blogs, chat groups, etc., !e !ant to be able to document them !ell and make sure they are relevant. • /e !ill discuss the much misunderstood concept of sample "representativeness.# • 9ou may collect data online or offline. !. Analy"e the data and #rite the report. II. !ummar" o #elivera$les A. ,eport *56: of product grade+ • 2ard copy, typed, double-spaced, 7; font. *see section &&& for suggested outline and level of detail+. • )oft copy of the hard copy in !ord processing format *i.e., no pdfs or other image files+ • <rammar and spelling are important/count. 0. =o!er=oint presentation *;6: of project grade+ • /e !ill be creating a "deck# of slides that could be used to present the research to management or a client. • 9ou !ill be presenting the findings in class. • >isual appeal and spelling are important/count. ?. )upporting .aterials *must be turned in to receive project grade+ • )=)) file containing study data if doing a survey. =rintouts of blogs, chat groups, focus groups depth intervie!s, and electronic copies *!ord docs+ and listing of !eb addresses. @xperience-based tips on getting the report doneA • /rite/draft the report as you goB Con t !ait until you are done !ith all of the data collection and analysis. The details !ill not be as fresh and you !ill have many other things on your plate. Teams !ho have follo!ed this advice in the past have not only done very !ell, they have completed the project earlier and !ith smiles on their faces. 2appiness is a choice. • @veryone should be involved in the data analysis in some form. .anaging and interpreting the data constitutes the major, value-added activity in the project. The !orld is full of good ideas, interesting data, and poor analysis. • Civision of labor and specialization can sometimes be a good thing. 4ust make sure everyone is comfortable !ith the !ay it is divided. There !ill be a peer evaluation. III. Generic Outline or the %ritten Re&ort A generic, but detailed outline for the report is provided in this section. A couple of other comments are provided belo!. • Artful, effective deviations in structure and content are !elcome. • The expected length is 15-20 pages *double-spaced, excluding appendices+. &t is possible to !rite an effective report that is shorter. &n contrast, reports that are longer than this are usually lacking in cogency and clarity. • &t is not good practice to include reams of output in the appendices. &nstead, this sends up a red flag that the report contains uncritical or unfocused analysis. All output should be clearly

4

labeled and referenced in the !ritten portion of the report. 'ther!ise, it does not belong in the report. A. $ntroduction (%&' pages • /hat is the general area of research1 • /hy is the research necessary1 • /hat !ill be accomplished in this research1 • 2int at the results *tease the reader – make them !ant to read more+ B. Background and (heory (%&' pages • ,evie! relevant prior research. /hat do !e already kno! about this topic1 • /e are primarily interested in peer-revie!ed literature, not unsubstantiated *i.e., nonempirical+ opinions or commercially motivated observations in magazine articles or blogs. o The best place to start is 0usiness )ource =remier. 9ou can search for many types of articles and, in most cases, access pdfs if you are on the 0D net!ork o The next place to look is <oogle )cholar. )ometimes you can dig up an article that isn t in 0usiness )ource =remier. o Another good place to look is /eb of )cience. • 0uilding on !hat !e learn, !e can develop our o!n theory about !hy things happen. 0ased on our theory, !e can state some expectations or hypothesis in terms of the study factors and responses. %ote that !e are making a distinction bet!een our explanation of something and the data !e !ill be using to test the explanation. The former is abstract and logical !hile the latter is concrete and factual. C. )ethods (*&+ pages 7. =articipants. o 3rom !hom did you collect the data1 o 2o! !ere they recruited to participate1 o =rovide some descriptive statistics. ;. =rocedure and .aterials o Cescribe ho! the study !as conducted. The focus here is on logistics. <ive enough detail so that the reader gets a sense of "!hat happened# o Cescribe any materials that are pertinent to the study *e.g., pictures or stories used as stimuli+. =rovide justifications for material choices !here appropriate. E. Cesign o This is !here you state the formal design in terms of factors and response variables. o &ndicate exactly ho! the factors are manipulated and ho! you assigned participants to receive the manipulations *e.g., random assignment+ ,. )easures o Cescribe the response measures *scales and observations+ and any other measures that !ere taken. o =rovide evidence of the reliability and validity of the measures.

5

D. -esults (,&+ pages 7. .anipulation checks. /ere the factors varied as intended1 /hat evidence do !e have that this is the case1 &t is important to have some evidence so that !e don t end up !ondering if non-significant effects !ere due to failed manipulations rather than poor measures, an underpo!ered sample, or an incorrect theory. ;. Tests of theory/hypotheses o This is !here !e !ant to report analyses of the data from the main study design o &t may help to organize the analysis according to the response measures o ,eport all necessary statistics. &n many cases a table may be desirable as a summary o /here possible, complement the analysis !ith a graphical depiction of the data o Co not editorialize. That is, do not comment on !hether the results are "good# or "bad# or !hat the potential implications are. The results section should be objective and factual. Fet the data tell its o!n story in this section. o 0e sure to indicate !here the data are consistent or inconsistent !ith the research hypotheses/expectations !. Discussion (%&' pages • )ummarize the key results • ,eiterate for the reader !hy the research is necessary and important .. $mplications (%&' pages • <iven kno!ledge of our results, !hat are the implications for other researchers or managers !ho may be affected by our findings1 /. 0imitations and .uture -esearch (%&' pages • All research has limitations because there are only so many factors that can be examined at one time. • 3ocus on limitations that are uni(ue to this study as opposed to philosophical issues that could be applied to virtually any in(uiry. • 0e sure to maintain a distinction bet!een potential and actual limitations. • Fimitations usually present an opportunity for future research to improve our kno!ledge. Ciscuss any particularly promising avenues for future research to advance your in(uiry. • /hat !ould be your advice for researchers thinking about running a study similar to yours1 3or example, !ere there any special challenges in terms of measurement or logistics or stimulus construction1 • <iven our likely time and budget constraints, the samples may be limited in terms of size *sampling error+ and representation *non-sampling error+. To !hat extent are !e concerned about these factors in terms of our o!n data1 o 2int 7A sample size impacts the po!er *sensitivity+ of statistical techni(ues to detect effects o 2int ;A sample representativeness does not mean that your respondents must physically or demographically resemble the population of interest. These variables are often completely irrelevant. /hat matters are the dependent variablesGdo your respondents resemble the population in terms of the processes underlying the specific responses gathered1 • /hat notable observations, if any, can !e make about respondents reactions or

6

behaviors during data collection *i.e., respondent-based non-sampling error+1 2. Appendices • ?opy of stimuli and/or (uestionnaire used to generate and collect data • ?oding sheet *i.e., the "key# for getting the variables and data from the (uestionnaire to the )=)) data file+ • 'ther selected output that is too large to place in the body of the report itself I'. PowerPoint !lide #eck 3ollo! the basic outline of the !ritten report, but be very selective in terms of content and detail. There are t!o important reasons for this. • 3irst, design the presentation to be about ;8 minutes long. This corresponds to the typical manager s or client s attention span for research results A + • )econd, in order to be interesting and effective, the presentation should focus more on !hat you found and less on ho! you found it *this is good advice for any type of research presentation+. ,emember, the intended audience may be fairly strong in terms of analytical thinking, but not necessarily !ell-versed in the specific techni(ues being presented. =resentation Assessment *dimensions !eighted e(ually+ A. Presentation qualityA ho! !ell the presentation communicates. A good presentation is one !here the recipient is provided !ith good transitions bet!een ideas and doesn t have to !ork hard to understand the main points. Dse diagrams and pictures !herever possible in place of numbers and in place of long chains of logic. 0. Content qualityA the in-depth thinking underlying the analysis and recommendationsH ability to include only the most pertinent issues in the presentation. ?. Use of the time limitA Tailor the level of detail to the time limit. )imply adhering to the time limit is not sufficient. /e could, in theory, adhere to any time limit by rushing through an over-prepared presentation or by dragging through an under-prepared presentation.

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close