3MA Midterm 1 Marketing Research PDF

lOMoARcPSD|32803390 3MA Mid 1 - midterm 1 Marketing Research (McMaster University) Scan to open on Studocu Studocu is not sponsored or endorsed by any college or university Downloaded by Jamie Li ([email protected]) ...

lOMoARcPSD|32803390 3MA Mid 1 - midterm 1 Marketing Research (McMaster University) Scan to open on Studocu Studocu is not sponsored or endorsed by any college or university Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 ROLE OF MARKETING RESEARCH WHAT IS MARKETING Marketing: the process of planning and executing the conception, pricing promotion, and distribution of ideas, goods, and services to create exchanges that satisfy individual and organization objectives ‘Right’ principles right goods or services to the right people at the right place at the right time at the right price, using the right promotion techniques THE MARKETING CONCEPT Consumer Orientation means that firms strive to: Identify the target market most likely to buy their products Create a market offering that meets the needs of the target market in the face of competition Goal Orientation Firms are consumer oriented only to the extent that allows them to achieve their goals System Orientation Creating a system that allows a firm to monitor the external environment and understand what consumers want. Consumer-oriented knowing what consumers want Goal-oriented environmental factors MARKETING RESEARCH The planning, collection, and analysis of data relevant to marketing decision-making and the communication of the results of this analysis to management Three functions Descriptive (what happened) Predictive (what will happen) Diagnostic (why & how) TWO TYPES OF RESEARCH Basic / Pure Research Investigates general theories and behaviours with the purpose of creating knowledge. Mostly conducted by academic researchers and results are shared as academic publications. E.g. Factors affecting consumers’ acceptance of electric vehicles Applied Marketing Research Initiated by a business solve specific business problems and will directly contribute to decision making E.g. consumers’ attitudes towards our new EV DO I NEED TO CONDUCT MARKETING RESEARCH? A business should NOT do marketing research if: Resources are lacking Research results would not be useful A decision has already been made Managers can not agree on the information needed Information already exists Costs outweigh the benefits Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 SECONDARY DATA SECONDARY VS. PRIMARY DATA There are two main sources of data: primary data and secondary data. Primary Data is new data gathered to help solve the problem under investigation (for the current study). Secondary Data is information that has already been gathered and might be relevant to the study at hand. Examples of Secondary Data: Previous research studies addressing the same problem Industry reports that help researchers understand the research environment Business’s own reports and studies from the company database WHY SECONDARY DATA? Saves time, money Clarify or redefine the problem during the exploratory research process May provide a solution to the problem May provide alternative methods to collect primary data May alert the marketing researcher to potential problems and/or difficulties for the research May provide necessary background information and build credibility for the research report May provide the sample frame. WHY NOT SECONDARY DATA? Lack of availability Lack of relevance Inaccuracy Insufficiency ERA OF BIG DATA Google handles a staggering 1.2 trillion searches every year The amount of data in the world was estimated to be 44 zettabytes at the beginning of 2020. 44000 million gigabyte INTERNAL DATABASES An internal database is a collection of related information developed from data from within the organization. A traditional starting point for a company’s internal database would be sales call reports that include the detailed transactions and interactions of salespeople with potential customers. With most businesses having some form of online presence, collecting data about potential customers have never been easier, even for smaller companies and startups. INTERNAL DATABASE MARKETING Database marketing – marketing that relies on the creation of a large computerized file of customers’ and potential customers’ profiles and purchase patterns to create a targeted marketing mix. Database marketing can get a customized, individual message to everyone simultaneously through direct mail. ONLINE AND WEB BEHAVIOR Behavioral targeting the use of online and offline data to understand a consumer’s habits, demographics, and social networks in order to increase the effectiveness of online advertising. PRIVACY The growth of databases is causing increased concern about privacy issues Identity theft when someone uses your personal information to assume your identity and apply for credit cards or government documents Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 Governmental actions Privacy Act Personal Information Protection and Electronic Documents Act MARKETING RESEARCH AGGREGATORS A company that acquires, catalogues, reformats, segments, and resells reports already published by large and small marketing research firms Role of aggregator firms is growing because their databases of research reports are getting bigger, more comprehensive, and more useful Major aggregators: AllNetResearch.com Profound.com Bitpipe.com USADATA.com MarketResearch.com INFORMATION MANAGEMENT Decision Support Systems (DSS) an interactive, personalized information management system designed to be initiated and controlled by individual decision makers Geographic Information Systems (GIS) computer-based systems that uses secondary and/or primary data to generate maps that visually display various types of data geographically THE MARKETING RESEARCH PROCESS STEP 1: IDENTIFYING THE RESEARCH PROBLEM STEP 1: RECOGNIZE THE PROBLEM OR OPPORTUNITY The external environment of business is constantly changing, so managers need to respond to these changes by with decision making Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 For example: Customers are not coming back ➔ improve products / services? Decreasing sales ➔ increase promotion? Carbon tax ➔ change pricing? STEP 2: FIND OUT WHY THE INFORMATION IS BEING SOUGHT Discuss what the information will be used for and what decisions might be made as a result of the research Finding out why the information is being sought helps researchers identify the correct research problem and research objectives. What is the real need? STEP 3:UNDERSTAND THE DECISION-MAKING ENVIRONMENT Situation analysis Understand the company operations, business portfolio and strategies Talk to brand managers Read company reports Visiting production facilities and retail stores Online search Exploratory research Secondary Data Secondary data analysis: studying data that have been previously collected by others and might be relevant to the current research Case analysis: reviewing information from situations that are similar to the one being studied. Intranet: some large organizations store reports about their old projects in their intranet. Primary data Pilot studies: short questionnaires with limited number of respondents Experience surveys: unstructured discussions with knowledgeable individuals Focus groups: in-depth discussions with 8-12 participants at the same time, led by a moderator STEP 4: USE SYMPTOMS TO CLARIFY THE PROBLEM Symptoms - phenomenon that occurs because of the existence of something else e.g. sales dropping, customer not return STEP 5: TRANSLATE MANAGEMENT PROBLEM INTO MARKETING RESEARCH PROBLEM Management decision problem a statement specifying the type of managerial action required to solve the problem Marketing research problem a statement summarizing the type of information needed by the decision-maker to help solve the management decision problem and show the manager how that information can be obtained efficiently and effectively Marketing research objective a goal statement defining the specific pieces of information needed to solve the marketing research problem STEP 6: DETERMINE WHETHER THE INFORMATION ALREADY EXISTS Using existing data can save managers time and money if such data can answer the research question. The researcher needs to decide whether this information needs to be updated through a new research study, or it is still valid and applies to the current situation. STEP 7: DETERMINE WHETHER THE QUESTION CAN BE ANSWERED Required information exists? Required information can be gathered? Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 Something completely new -> Is there a risk? STEP 8: STATE THE RESEARCH OBJECTIVES The research objectives are the end-result of the research process, therefore, it should contain the description of all the information that the manager needs, to answer the management decision problem Managers should agree and sign off on the research objectives before the actual research begins. Research objectives also serve as a standard that later will enable managers to evaluate the quality and value of the work RESEARCH OBJECTIVES AS HYPOTHESES Hypothesis – a conjectural statement about a relationship between two or more variables that can be tested with empirical data Contain clear implications for testing stated relationships Sets the stage for creating the research design QUALITATIVE RESEARCH QUALITATIVE VS. QUANTITATIVE DATA Qualitative Data is expressed in words that are open to interpretation and can not be quantified: I feel good today I don’t like this brand I hate SMS ads Quantitative Data is expressed in numbers that can be statistically analyzed: I give the taste of this coffee 7/10 Average score for the aroma of the coffee was 6.5/10 QUALITATIVE RESEARCH VS. QUANTITATIVE RESEARCH Qualitative research Findings that are not subject to quantification or quantitative analysis Could be used to examine the attitudes, feelings, or motivations of consumers Acquire qualitative data usually through open-ended unstructured questions Quantitative research Research that uses mathematical analysis Could be used to quantitatively measure the attitudes, feelings, or motivations of consumers Acquire quantitative data through specific close-ended questions using scales and multiple-choice questions WHY QUALITATIVE RESEARCH? Less expensive than quantitative research Understand the in-depth motivations and feelings of consumers Improve the efficiency of quantitative research LIMITATIONS OF QUALITATIVE RESEARCH Very large amount of data per respondent ➔ small sample size Not representative of the population of interest Qualitative data are usually difficult to interpret and analyze Results are open to interpretation ➔ not conclusive TYPE OF QUALITATIVE RESEARCH Focus Groups Group of 8 to 12 participants who are led by a moderator in an in-depth discussion on one particular topic or concept. The goal is to learn and understand what people have to say and why in order to find out their feeling about a product, concept, idea of organization Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 The quality of the data is related to the quality of group interaction Group dynamics STEP 1: FOCUS GROUP FACILITY Focus group facility – conference room with a large one-way mirror built into one wall and microphones placed in an unobtrusive location to record the discussion. STEP 1: RECRUIT PARTICIPANTS Researchers prepare sampling criteria and screening questions Sources of focus group participants Mall-intercept interviewing Random telephone screening Influentials – people ahead of the majority of consumers in terms of product knowledge and expertise Target market Marketing professionals – prosumers Avoid “professional” respondents FOCUS GROUP - SIZE Usually 8 – 12 participants More participants = less time available for each participants Size of a focus group also depends on the nature of the product STEP 2: CHOOSE THE MODERATOR Hired by the researcher to lead the focus group Key skills Background in psychology or sociology or marketing Good business skills Other attributes Interest in people, knowledge of the product, listening & observation skills, oral & written communication skills, attention to details, knowledge of the basics of research…… STEP 2: CREATE DISCUSSION GUIDE A written outline of topics to be covered during a focus group discussion Based on research objectives and information needs Three stages of discussion guide Rapport is established Intensive discussion is provoked Significant conclusion are summarized and tested STEP 3: CONDUCT THE GROUP - LENGTH Average length - 90 minutes Longer groups (> 2 hours) get more things done in a single session More involvement from participants More time to complete time-consuming tasks and interact more extensively Length of focus groups intertwined with the number of questions in the discussion guide STEP 4: FOCUS GROUP REPORT Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 After the final focus group in a series, the moderator can brief the client on their initial impression from the discussions. This is called instant debriefing Instant debriefing is not the official analysis or results of the data. It is just a first impressions brainstorming session between the client observers and the moderator. Limited memory, selective recall, among other factors can make instant debriefing inaccurate. Therefore, moderators and researchers should reserve the right to change their opinion after properly analyzing the tapes. ADVANTAGES OF FOCUS GROUPS Interaction among respondents can stimulate new ideas and thoughts Company executives can watch customers’ talking about their products first hand Can be executed more quickly that other techniques Findings from a group are easier to understand and more relatable to managers versus statistics and numbers. DISADVANTAGES OF FOCUS GROUPS Not representative. Not conclusive One atypical consumer can cause a lot of damage! Depends on moderator Possible unequal contribution from respondents. ONLINE FOCUS GROUPS Advantages: Lower costs Easier to combine participants who are geographically separate in one group Participants can be more comfortable in a less confrontational setting Access to hard to reach populations Easier communication between the moderator and client during the focus group Transcripts are digitally generated Disadvantages: Group dynamics are not as effective online Limitations on body language Watching a focus group “live” from behind a mirror is better than through a video External environment is not controlled Interacting with a stimuli (packaging, an ad concept, etc.) is different “live” than online INDIVIDUAL DEPTH INTERVIEWS (IDIS) Relatively unstructured one-on-one interviews that probe and elicit detailed answers to questions Often using non-directive techniques to uncover hidden motivations The interviewer will probe each answer to find out the underlying motivations or feelings. For example: “Can you tell me more?” “Would you elaborate on that?” IDIS – HERMENEUTICS & DELPHI METHOD Hermeneutic Research Doing one interview with each participant to interpret their behavior and motivations through conversation Delphi-Method Multiple rounds of individual data collection from knowledgeable people. Results are summarized and returned to the participants for further refinement E.g. new product development Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 APPLICATIONS OF IDIS Communication checks (e.g. attitudes towards ads) Sensory evaluation Exploratory research Advantages of IDIs No group pressure Share viewpoint without influence from others Longer time per respondent Heightened state of aware New directions of questioning can be improvised more easily Can be conducted anywhere Disadvantages of IDIs More expensive Client involvement Only 4-5 interviews per day No group dynamics PROJECTIVE TESTS A technique with the objective of tapping respondents’ deepest feelings by having them project those feelings into an unstructured situation. Why projective tests – consumers may not tell us everything that influences them because: They are unconscious or unaware of a particular influence They are aware of an influence but feel it is too personal or socially undesirable to admit They are not aware that a particular influence is worth mentioning WORD ASSOCIATION TESTS The interviewer says a word and the respondent must mention the first thing that comes to mind Application Brand names Advertising campaign themes Example: the first word that comes to your mind about Tim Horton's ANALOGIES Comparison of two items based on similarities Example: “There is a list of store brands: The Bay, Walmart, Sears, Canadian Tire, Costco…… Which of these stores is most similar to Toyota?” When respondents compare one brand to another, it can be easier for them to describe Toyota in adjectives and words they wouldn’t have thought of without comparing it to a store. So the objective behind Analogies is not to know which store is similar to Toyota, but rather helping respondents articulate their perceptions of the Toyota car. PERSONIFICATION Respondents draws a comparison between a product and a person The discussion then goes on to describe the “person’s” values, beliefs, demographics, hobbies, etc. Example: Describe Coca Cola as if it was a person SENTENCE AND STORY COMPLETION TESTS Respondents complete sentences or stories in their own words Example: The people who drink coca-cola are __________ Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 OTHER TYPES PROJECTIVE TESTS Cartoon tests Respondent fills in the dialogue of one of two characters in a cartoon Photo sorts Respondent sorts photos of different types of people Consumer drawing Respondents draw what they are feeling or how they perceive an object OTHER TYPES PROJECTIVE TESTS Storytelling Respondents are required to tell stories about their experiences with a company or product Third-person technique Interviewer learns about respondent feelings by asking them to answer for a third party, such as “your neighbor” or “most people” Why do a lot of people skip morning breakfast? Why would someone buy a Galaxy s10+ over an iPhone 11? RESEARCH DESIGN Research Design is the roadmap for conducting a research project. EXPLORATORY RESEARCH METHODS Qualitative analysis of secondary data Qualitative primary research Pilot surveys Case studies Expert surveys WHY EXPLORATORY RESEARCH? To formulate a problem or define a problem more precisely To develop hypotheses Hypothesis – a conjectural statement about a relationship between two or more variables that can be tested with empirical data. Contain clear implications for testing stated relationships To isolate key variables and relationships To establish priorities for further research DESCRIPTIVE RESEARCH A type of conclusive research that has the major objective of describing something. While exploratory research is used when the researcher does not know the exact problem, descriptive research is used when the research knows the exact causes of the problem or is able to precisely identify what data need to be collected. Example - coffee Exploratory Possible Coffee quality elements Qualitative difference between Starbucks & Tim Hortons Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 Descriptive. Importance of each of the Coffee quality elements Rating differences between Starbucks & Tim Hortons on these elements DESCRIPTIVE RESEARCH DESIGNS Cross-sectional design (sample survey) Involve the one-time collection of information form any given sample of population elements Longitudinal designs (panel) Involve a fixed sample of population elements that is measured repeatedly CAUSAL RESEARCH A type of conclusive research with the objective of examining the causal relationship between two variables. E.g. How does prices change (independent variable) effect sales (dependent variable) Correlation ≠ causal relationship Children that play a lot of video games are the most violent. Clearly, video games make children more violent! As ice cream sales increase, the rate of drowning deaths increases sharply. Therefore, ice cream consumption causes drowning! My friend and I got covid virus after having a Zoom class last week, so Zoom is spreading covid! Bill Gates, Steve Jobs, and Mark Zuckerberg dropped out of university and became billionaires. Clearly, I need to quit university to become rich! Causal relationship – A causes B A before B A & B correlated No alternative explanation In a causal design, the researcher manipulates the independent variable in a controlled environment and monitors the change in the dependent variable. The main research method used for causal research is experiments. Example: manipulate price (independent variable) to see the change on sales (dependent variable) in a lab setting (controlled environment) THE NATURE OF OBSERVATION RESEARCH Observation research is the systematic process of recording patterns of occurrences or behaviors without normally communicating with the people involved Observed by human or machine Conditions for using observation Behaviour is observable Behaviour is repetitive or predictable Behaviour is of relatively short duration WHY USING OBSERVATION RESEARCH? Watching what people do is more accurate compared to self reporting Some types of data can be gathered faster through observation Situation when participants cannot account for their own behaviors Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 WHY NOT? Only behaviour and physical personal characteristics can be examined Only public behaviour can be observed Behavior may not be projectable into the future Can be time-consuming and costly if the behavior is infrequent NATURAL VS. CONTRIVED SITUATIONS Natural situation participants are observed in a natural setting Those being observed usually have no idea they are under observation Key advantage: real-world inference Contrived situation participants are observed in a simulated environment Recruiting people to participate in an activity so that their behaviour can be studied Key advantage: controlled environment / fast OPEN VS. DISGUISED OBSERVATION Open observation the process of monitoring people who know they are being watched Disguised observation process of monitoring people who do not know they are being watched DIRECT VS. INDIRECT OBSERVATION Direct observation Concerned with current behavior Indirect observation Concerned with past behavior ➔ inferred from other indicators E.g. analyze household consumption patterns through sorting people’s garbage HUMAN VS. MACHINE OBSERVER Human observer Actual person is watching the situation Machine observer Observation can be done less expensively, more accurately, or more readily E.g. traffic-counting devices, movie cameras and audiovisual equipment, electronic scanners can all provide more accurate and timely data HUMAN OBSERVER ETHNOGRAPHIC RESEARCH Ethnographic research is the study of human behaviour in its natural context, involving observation of behaviour and physical setting MYSTERY SHOPPERS Mystery shoppers are observers who pose as consumers and shop at a company’s own stores or those of its competitors to collect data about: customer-employee interactions compare prices / displays 4 levels – differ in the depth and type of information Level 1: mystery telephone call Level 2: quick purchase – evaluate transaction and store image Level 3: initiate a conversation with a service or sales representative following a script Level 4: engage a long discussion on product or purchase process WHEN TO USE MYSTERY SHOPPERS Measure employee training Check compliance with standards and specifications Examine actual service Identify differences in customer experience Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 ONE-WAY MIRROR OBSERVATION One-way mirror observation is the practice of watching behaviours or activities from behind a one-way mirror participants should be informed that they are observed Used for: Focus groups Observe users of products Observe children MACHINE OBSERVER TRAFFIC COUNTERS Machines used to measure vehicular flow over a particular stretch of roadway Real-world application Retailers use the information to determine the location of certain types of stores (e.g. fast-food) Outdoor advertisers use the information to determine placement of a billboard or the number of exposures per day PHYSIOLOGICAL MEASUREMENT DEVICES Machines that measure human physical reactions to certain stimuli Heartbeats Facial expressions Brain wave Neuromarketing is the use of technology monitoring brain activity and other physical activations in making marketing activities more effective and tailored FUNCTIONAL MAGNETIC RESONANCE IMAGING (FMRI) fMRI is a neuro-imaging technique that measures brain activity and produces images of it In practice, marketers use fMRI to understand the motivations for consumers to make purchase decisions. OTHER PHYSIOLOGICAL MEASUREMENT DEVICES Eye Tracking wearing glasses that look like modern eyewear Electroencephalograph (EEG) measures electrical pulses on the scalp and generates a record of electrical activity in the brain Galvanic Skin Response (GSR) measures changes in the electric resistance of the skin associated with activation responses SCANNER-BASED RESEARCH Used to scan barcodes of purchased items ACNielsen Homescan Consumer Panel captures purchase information from a statistically reliable sample of 12,300 Canadian households on a continuing basis provides researchers with highly accurate data on repeated purchase behavior, geographic and demographic profiles of the buyers, and from where buyers buy their products. Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 OBSERVATION RESEARCH ON THE INTERNET Online activity can be tracked, and companies are building huge databases on online shopping behavior by leaving cookies on users’ computers through their websites Scraping Companies scrape their website and use advanced analytical tools to automatically analyze consumer generated content MEASUREMENT SCALES Scales are questions that assign values to the measured item depending on a certain rule. e.g. : weight scale, thermometer, grading system There are four different levels of measurement scales: Nominal, Ordinal, Interval, and Ratio, each with different characteristics and analytical powers. NOMINAL SCALES It is used in MCQ questions where the choices have no sequential relationship to one another. For example: Gender: (1) Male (2) Female Province: (1) ON (2) MB (3) QB ORDINAL SCALES Nominal scales are turned into ordinal scales when respondents are asked to order or rank the choices. Example: Please rank the following provinces from 1 to 5 with 1 being your least preferred and 5 being your most preferred: Ontario ___5_____ British Columbia ____4_____ Manitoba ____2______ Quebec _____3______ Nova Scotia _____1_____ The numeric values assigned only indicate rank or order. The range between the numbers do NOT reflect intensity of preference INTERVAL SCALES Scales with the same characteristics as ordinal scales (values reflect preference or order), but the interval between the different choices (or points on the scale) is equal or assumed to be equal: On a scale of 1 to 5, how likely are you to purchase this product again: 5 → Most likely 4 3 → Neutral 2 1 → Most unlikely In this scale it is assumed that the interval between 5 and 4 is the same as 4 and 3 and so forth. The scale has the properties of order and difference. Interval scales have an arbitrary zero. This means that a value of “zero” does not equal “nothing”. Think of temperature, which is a type of interval scale (the values of different temperatures can be ordered, and the range between consecutive temperatures on the thermometer are equal). A temperature of 0 degrees celsius does not mean that the temperature does not exist. Interval scales have more analytical power than nominal and ordinal scales. For example, you calculate the average temperature for the whole month. In addition, more powerful statistical analysis (e.g. correlations) can be performed on interval scales (more about that in Data Analysis). However, Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 we can not multiply or divide the values of an interval scale because the values do not reflect real quantities. On a scale of 1 to 5, how likely are you to purchase this product again: 5 → Most likely 4 3 → Neutral 2 1 → Most unlikely RATIO SCALES Ratio scales have all the characteristics of nominal, ordinal, and interval scales, plus having a meaningful zero point. E.g., age, income, weight That is why, the value in a ratio scale reflects the actual quantity of the variable being measured (age, income, etc.). Therefore, we can multiply and divide these values in a meaningful manner. We can say that someone earns double the income of another person. ATTITUDE MEASUREMENT SCALES MEASURING ATTITUDES Questions that measure the respondents’ attitudes towards a product, a situation, or a behavior Key challenge – not directly observable The scale can be applied to measure other information GRAPHIC RATING SCALES These are scales that include a graphic continuum, anchored by two extremes and respondents are asked to place their answer anywhere on the continuum. ITEMIZED RATING SCALES Measurement scales in which the respondent selects an answer from a limited number of ordered categories SEMANTIC DIFFERENTIAL SCALES Limitations: Adjectives have to be clear opposites The Halo effect (always choosing positive being affected by an overall positive experience, rather than giving each attribute a separate thought) Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 STAPEL SCALES An adaptation of the semantic differential scale, where the adjective is placed in the middle of a scale ranging from +5 to -5 LIKERT SCALE An itemized rating scale where respondents are asked to rate their level of agreement of each statement. PURCHASE INTENT SCALES Scales used to measure the respondent’s intention to buy or not buy a product. It can help managers estimate potential for a new product and its market share. Non-comparative scale: Judgement is made without reference to another object, concept, or person Comparative scale: One object, concept, or person is compared with another on a scale RANK-ORDER SCALES Rank order scales are comparative scales where respondents compare two or more items and rank them. The scale has several limitations: If the respondent’s actual no. 1 choice is not included, they will choose their second best as their number 1 which is misleading. If the scale includes choices that the respondent will never consider, the respondent will still have to rank them, which is also misleading. The scale only offers ordinal data. Q-Sorting: sort cards with different items (e.g. 60-120 cards, each with a different product feature) into different piles (e.g. Important, neutral, unimportant). The objective of this scale is to identify respondents with similar attitudes which can be a basis for segmentation Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 Paired Comparisons: Constant Sum Scales CONSIDERATIONS IN SELECTING A SCALE Nature of the construct being measured: E.g. When measuring customers’ overall satisfaction on behalf of a restaurant, the nature of the scale and its wording depends on how you define satisfaction. If overall satisfaction is the attitude of consumers towards their overall experience, then you can ask your questions about their overall experience to measure satisfaction. If overall satisfaction is defined as a factor of satisfaction from other elements (satisfaction from food, service, atmosphere, cleanliness, parking, etc.), then you will need to ask about the level of satisfaction in all of these elements. Type of scale: most commercial researchers prefer scales that can be administered over the phone or the internet. Balanced versus Unbalanced Scales: Balanced scales are used when the intensity of attitudes are expected to be evenly distributed between positive and negative (e.g. very good, good, neutral, bad, very bad) Unbalanced scales are used when attitudes are known to skew towards one extreme. For example, if it is known that the image is overly positive, the scale can be (Outstanding, very good, good, fair, poor) with more positive options than negative, to help marketers further analyze exactly how positive the image is. Number of categories: A 3-point scale will yield poor data while a 10-point scale can be confusing. 5 or 7 scales are most reliable Forced vs. Non-Forced Choice: Even or Odd? (include neutral point or not) Include a Don’t Know option or not? QUESTIONNAIRE DESIGN A QUESTIONNAIRE A questionnaire is a set of questions designed to generate the data necessary to accomplish the objectives of the research project. The questionnaire standardizes the wording and sequencing of questions and imposes uniformity on the data gathering process, which facilitates data analysis and comparison between respondents. A good questionnaire should: provide the data needed to achieve the research objectives consider the respondents in terms of the questionnaire length, wording, and jargons be easy to edit and code with clear skip patterns and branching. Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 STEP 1: DETERMINE SURVEY OBJECTIVES In this step, we determine what information we need to collect from the questionnaire Consumer perceptions towards two elements of the product mix of a coffee shop Their associations with the likelihood of repeated purchases and recommending the product 6 hypothesis: 1) On average, consumers have negative perceptions towards the coffee flavour. 2) The perception towards coffee flavour is associated with the likelihood of repeated purchases. 3) The perception towards coffee flavour is associated with the likelihood of recommending our product. 4) On average, consumers have negative perceptions towards customer service. 5) The quality of the customer service is association with the likelihood of repeated purchases 6) The quality of the customer service is associated with the likelihood of recommending our product Other things that are determined include the budget and the time available for the data to be collected. STEP 2: DETERMINE DATA COLLECTION METHOD The data collection method depends on the survey objectives among other factors discussed earlier. The data collection method affects what type of questions will be included in the questionnaire. STEP 3: DETERMINE THE QUESTION RESPONSE FORMAT There are many types of questions that can be included on a questionnaire. The research design affects the type of questions. For a qualitative survey, the following types of questions are usually included: Laddering questions: Help interviewers delve into the deeper reasons on how a participant relates to a brand or performs certain behavior (why is this important to you?) Open-ended questions: Questions to which the respondent replies in their own words. They provide rich information but are hard to record and analyze. Probing questions: Questions that encourage respondents to elaborate on an incomplete response. STEP 3: DETERMINE THE QUESTION RESPONSE FORMAT For a quantitative study, close ended questions are mostly used: Dichotomous questions: Questions that ask respondents to choose between two answers. E.g.: yes/no questions Prone to measurement error because they fail to capture a wide range of possible choices between the two extremes and fail to convey the intensity of emotion Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 The order of the choices can bias respondents into selecting the first choice. This is called position bias Multiple choice questions: Questions that ask the respondent to choose among several answers It needs a lot of research to make sure all possible or potential choices are known, and a second decision to select the choices to include in the questionnaire. A long list of choices can confuse respondents. Position bias can also be a problem Scaled-Response questions (measurement scales) STEP 4: DECIDE ON THE QUESTION WORDING This step might sound trivial but it can be challenging to decide on the wording of a question, which can be a source of a lot of measurement error. No guidelines for good wording. However, try to avoid: Leading questions (implying something) Double-barreled (asking about two things in the same question) No information Encourage lying No time limit Unclear, confusing, difficult to understand STEP 5: ESTABLISH THE QUESTIONNAIRE FLOW AND LAYOUT STEP 6: EVALUATE THE QUESTIONNAIRE After finishing the questionnaire draft, you have to evaluate the following attributes: Is the question necessary? (e.g. too many demographic and psychographic questions?) Is the questionnaire too long? (if it is too long, should we offer an incentive?) Will the question provide information needed to accomplish the research objectives? STEP 7: OBTAIN APPROVALS FROM ALL RELEVANT PARTIES STEP 8: PRETEST AND REVISE A pretest is a trial run of a questionnaire The researcher usually administered the questionnaire face to face to a small number of respondents, and would ask the respondents to let them know if there are any questions or words that are not clear or confusing. After the pretest, modifications are made if needed. STEP 9: PREPARE THE FINAL QUESTIONNAIRE COPY Precise instructions for skip patterns, numbering, and pre-coding must be set up at this stage. STEP 10: IMPLEMENT THE SURVEY In most mail and telephone research interviews, the implementation is not done by the researchers. The researchers hire field management companies that collect the data. In order to make sure the data is collected correctly, a supervisor's instructions, written directions to the field service firm on how to conduct the survey, is prepared. It includes information on the sample size, staff requirements, location, project supplies, among others. Field management companies only collect data. They do not interfere with the research design or the data analysis QUESTIONNAIRE DEVELOPMENT SOFTWARE Benefits: ❖ Easy to use ❖ Randomization of questions and choices (to avoid position bias) ❖ Forced completion for important questions (to avoid missing responses) ❖ Dynamic lists (using the respondent’s choices from an earlier questions as input to subsequent questions) Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 ❖ Control of sample size and quotas (e.g. 50% male 50% female) ❖ Response verification (e.g. telephone numbers) ❖ Skip logic ❖ Embedded audio, graphics, etc. ❖ Translation ❖ Downloadable organized responses (minimal processing needed) SURVEYS WHAT IS SURVEY RESEARCH? ❖ Survey research is the use of questionnaires to gather facts, opinions, and attitudes. It is the most popular way to gather primary data: It helps marketers understand why something is happening It helps in identifying the demographic, psychographic, etc. profiles of a certain segment It helps marketers understand how something happens (e.g. purchase decision process) Surveys can yield both quantitative and qualitative data. ❖ Quantitative data from a huge sample is common in descriptive research ERRORS IN SURVEY RESEARCH ❖ There is no study that is 100% free of errors. ❖ There are two major types of errors: random error (to be discussed later in data analysis) systematic error ❖ There are different sources of systematic error, most of which can be reduced by a solid research design and skilled interviewers. ❖ They can be divided into two types of errors: Sample design error Measurement error MEASUREMENT ERROR ❖ It is the error that results from a variation between the information being sought and what is actually obtained by the measurement process. Surrogate Information Error: results from discrepancy between information needed to solve a problem and that sought by the researcher → Wrong problem and objectives Interviewer Bias: when interviewers influence the answers of respondents consciously (e.g. cheating) or unconsciously (e.g. wearing the brand’s t-shirt when asking about attitudes towards the brand). Measurement Instrument Bias: error from problems in the questionnaire design (e.g. confusing questions, words with double meaning, etc.) MEASUREMENT ERROR ❖ It is the error that results from a variation between the information being sought and what is actually obtained by the measurement process. Processing Error: error when transferring answers from questionnaire to the computer. Non-Response Bias: occurs when there are systematic differences between respondents and non-respondents in a sample. Response Bias: occurs when respondents give false or inaccurate answers. TYPES OF SURVEYS ❖ Door to Door Interviews are in person at respondents’ homes or offices (executive interviews for B2B products) Advantages: immediate feedback, participants are at ease Disadvantages: expensive, time consuming, high refusal rates Downloaded by Jamie Li ([email protected])  lOMoARcPSD|32803390 ❖ Mall Intercept Shoppers are intercepted while shopping and asked to participate in a survey in the mall or in a different facility Advantages: less expensive and saves more time than door-to-door Disadvantages: impossible to get a representative sample, respondents can be distracted ❖ Telephone Survey over the landline or cellphone. Autodialers are used to facilitate the process. Computer Assisted Telephone Interviews (CATI) enables more accuracy and standardization. Advantages: Inexpensive, high quality sample Disadvantages: does not allow researchers to show respondents certain stimuli, limits quantity and type of information, patience wears out on the phone, calling cell phones is more expensive SELF-ADMINISTERED SURVEYS ❖ What are self-administered surveys? Questionnaire filled out by respondents with no interviewer present. Advantages: no interviewer bias Disadvantages: more suitable to collect quantitative data, respondents may not understand the questions, respondents can leave questions unanswered ❖ Mail Questionnaires sent to the sample through mail once (ad hoc mail surveys) or multiple times over a period of time (mail panels) Advantages: high quality sample, cheaper than earlier methods Disadvantages: low response rate, non-response bias, same disadvantages as self-administered ❖ Online Surveys Surveys administered over the internet Advantages: Fast, cheap, can be personalized, high response rates, easier to reach far respondents, easier to organize and manage especially for longitudinal studies Disadvantages: sample might not be representative for some countries, inaccurate sampling if unrestricted internet sample is used, can not reach the research population if you do not have their email addresses. WHICH SURVEY METHOD TO USE? ❖ This depends on a number of factors: Sampling precision: if high precision is required, choose methods that can help you get a representative sample, with high response rates. Budget: internet is the cheapest, face to face most expensive Requirements for respondents’ reactions: do respondents need to examine a certain stimuli? Quality of the data: quality of data from difficult questions can be compromised without an interviewer to explain or probe. Length of the questionnaire: long surveys can be more suitable for face-to-face as opposed to the phone or mall intercepts Questionnaire Structure: less structured questionnaires can require face to face Incidence rate: % of the research population from the entire population. If it is low, it is more difficult and expensive to find respondents. Study Time Frame: how quickly do you need the survey results Downloaded by Jamie Li ([email protected])