MAR Final

UGC

'User generated content'.

Decision Support Systems

(DDS) Is designed to support the needs and styles of individual decision makers. In theory, it represents something close to the ultimate in data management. Managers use DDS to conduct sales analysis, forecast sales, evaluate advertising, analyze product lines, and keep tabs on market trends and competitor action.

Screening

- Begin with a question that gets the respondent's interest. - Ask general questions first. - Ask questions that require 'work' in the middle. - Insert 'prompters' (short encouraging statements to rebuild respondent interest) at strategic points. - Position sensitive, threatening, and demographic questions at the end. - Put instructions in capital letters. - Use a proper intro and closing.

Classic Applications of In-depth Interviews

- Communication checks - Sensory evaluations - Exploratory research - New product development, prototype stage - Packaging or usage research

Limitations of Experimental Research

- Cost of experiments - Security issues - Implementation problems

Application of Data Mining

- Customer acquisition - Customer retention - Customer abandonment - Market basket analysis

Define the Objective

- Develop share and volume estimates. - Determine characteristics of people who are purchasing the product. - Determine frequency and purpose of the product. - Determine where purchases are made. - Measure effect of sales of the new product on sales of similar existing products in the line.

Key Benefits of a Web Community

- Engages customer in a space where they are most comfortable. - Uncover 'exciters' and 'eureka moments' resulting in customer driven innovations. - Establish brand advocates who are emotionally invested in a company's success. - Offer real-time results, enabling clients to explore ideas. - Create a forum in which natural dialogue allows customers to initiate topics important to them. - Maps psyche of consumer segments. - Brainstorm new ideas. - Co-create and test new products. - Observe natural consumer behavior. - Rally the company around a customer-centered perspective.

Mystery Shopping Objectives

- Establishing an organization to monitor compliance with product/ service delivery standards and specification. - Enabling marketers to examine the gap between promises made through advertising/ sales promotions and actual service delivery. - Helping monitor the impact of training and performance improvement initiatives on compliance with or conformance to product/ service delivery specification. - ID differences in customer experience across different times of day, locations, product/ service types, and other potential sources of variation in product/ service quality.

Advantages of Observation Research

- Firsthand info is not subject to many of the biasing factors associated with surveying research. - Research avoids problems associated with willingness and ability of respondents to answer survey questions. - Some forms of data are gathered more quickly and accurately.

Key Attributes for Conducting a Focus Group

- Genuine interest in people, their behavior, emotions, lifestyles, passions, and opinions. - Acceptance and appreciation for the differences in people, especially those whose lives vary greatly from your own. - Good listening skills: Ability to hear what is being said and to identify what is not being said. - Good observation skills: Ability to see in detail what is happening or not happening and to interpret body language. - Interest in a wide range of topics and the ability to immerse yourself in the topic and learn the necessary knowledge and language quickly. - Good oral and written communication skills. - Objectivity: Ability to set your own personal feelings aside and remain open to ideas of others. - Sound knowledge of basic principles, foundations, and applications of research, marketing, and advertising. - Flexibility - Good attention to detail and organizational ability.

Advantages of In-depth Interviews over Focus Groups

- Group pressure is eliminated. - Personal one-on-one situation gives respondent feeling of being focus of attention. - Respondent attains a heightened state of awareness. - Longer time devoted to individual encourages revelation of new info. - Respondents can be probed at length to reveal feelings and motivations that underlie statements. - Without restrictions of cultivating a group process, new directions of questioning can be improvised more easily. - Allows interviewer to be more sensitive to nonverbal feedback. - Singular point of view can be obtained from a respondent without influence from others. - Interview can be conducted anywhere.

Typical Programmatic Questions

- Has its target market changed? How? - Does the market exhibit any new segmentation opportunities? - Do some segments appear to be more likely candidates than others for the firm's marketing efforts? - What new product or service opportunities lie in the various segments?

Disadvantages of Focus Groups

- Immediacy and apparent understanding of findings can cause managers to be misled instead of informed. - Related to focus group process: Ex. Recruited wrong type of people. - Greatest potential for distortion is the interview itself. As a participant in the the social interaction, moderators must take care not to behave in ways that could prejudice the responses. - Respondents can also be a problem when they don't like to speak out in group settings.

Characteristics of a True DDS

- Interactive: Manager gives simple instructions and sees results generated on the spot. - Flexible: DDS can sort, regroup, total, average, and manipulate data in a variety of ways. - Discovery Oriented: DDS helps managers probe for trends, isolate problems, and ask new questions. - Easy to Learn and Use: Managers need not be particularly knowledgeable about computers.

Limitations of Secondary Data

- Lack of availability - Lack of relevance - Inaccuracy

Choose a Data Collection Method

- Mail surveys: Suffer from low response bias. - Telephone surveys: Less significant problem with non-responses, but suffer from call screening technologies used by potential respondents and the fact that some people only have mobile phones. - Internet surveys: Problems with professional respondents and the fact that panel/e-mail lists used often do not provide appropriate representation of the population of interest.

Advantages of Secondary Data

- May help to clarify/ redefine the problem during exploratory research process. - May provide an actual solution to the problem. - May provide primary data research method alternative. - May alert the marketing researcher to potential problems and or difficulties. - May provide necessary background info and build credibility for research report. - May provide sample frame.

Disadvantages of In-depth Interviews Relative to Focus Groups

- More expensive. - Do not generally get same degree of client involvement. - Do not cover as much ground as focus groups. - Focus groups give moderator an ability to leverage dynamics of the group to obtain reactions that might not be generated in a one-on-one interview

Approaches to Observation Research

- Natural vs contrived situations - Open vs disguised observation - Human vs machine observers - Direct vs indirect observation

Four Basic Levels of Measurement (ex. 9.3)

- Nominal - Ordinal - Interval - Ratio

Disadvantages of Observation Research

- Only behavior and personal physical characteristics can be examined. - Research cannot learn about motives, attitudes, intentions, or feelings. - Only public behavior is observed. Private behavior (ex. Getting dressed for work.) is beyond the scope of observation research. - The present observed brand may not be projectable into the future. - Time consuming and costly if observed behavior occurs infrequently.

Population and Sample Distributions

- Population distribution: Is a frequency distribution of all the elements of the population. It has a mean and standard deviation both usually represented by its Greek symbol. - Sample distribution: Is a frequency distribution of all the elements of an individual/ single sample. Mean is usually represented by X and standard deviation by S.

Telephone Interviews

- Predictive dialing - Cell phone interviewing - Call center interviews

Select a Sampling Method

- Probability samples: Selected in such a way that every element of the population has a known, non-zero likelihood of selection. 'Simple Random Sampling' is the best known and most widely used probability sampling method. Researchers must adhere to precise selection procedures to avoid arbitrary or biased selection of sample elements. (Difference between sample value and population value is referred to as 'sampling error') - Non Probability samples: Those in which specific elements from the population have been selected in a nonrandom manner. ('Nonrandomness' results when population elements are selected on the basis of convenience. 'Purposeful nonrandomness' occurs when a sampling plan systematically excludes or over represents certain subsets of the population.)

Problem Definition Process (Ex. 2.1)

- Recognize the problem or opportunity - Find out why the information is being sought - Understand the decision-making environment (the industry, company, product, and target market) - Use the symptoms to help clarify the problem - Translate the management problem into a marketing research problem - Determine whether the info already exist - Determine whether the question can really be answered - State the research objectives

Types of Test Markets

- Scanner or electronic test market: Markets where scanner panel research firms have panels of consumers who carry scannable cards for use in buying particular products, especially those sold at grocery stores. - Controlled test markets: Managed by research suppliers who ensure that the product is distributed through the agreed upon types and number of distributors. - STMs (Simulated Test Markets): simulations of the types of test markets noted above. Can normally be conducted more quickly than the other 2 approaches, at a lower cost, and produce highly predictive results of what will actually happen.

Select Test Markets

- There should be a minimum of two test markets, in addition to a control market for an existing national brand. - The markets selected should be geographically dispersed. - Markets should be demographically representative of the United States. - Test should run for at least 6 months. - Must have a variety of media outlets. - There should be a dominant newspaper in the market.

Select a Basic Approach

- Traditional or standard test market - Scanner or electronic test market - Controlled test market - STM

Quantitative Research (Ex. 4.1)

- Types of questions: Limited probing - Sample size: Large - Amount of info from each respondent: Varies - Requirements for administration: Interviewer with fewer special skills or no interviewer - Type of analysis: Statistical, summation - Hardware: Questionnaires, computers, printouts, mobile devices - Degree of replication: High - Researcher training: Statistical, decision models, decision support systems, computer programming, marketing, marketing research - Type of research: Descriptive of casual

Gaining Survey Completion

- Use conversational language. - Be honest and upfront about the time required to complete the study. - Provide more opportunities for participants to provide open ended answers. - Ensure all possible answer choices are give. Avoid using the term 'other'. - Keep the survey to less than 20 minutes in length and provide participants with progress info as they advance through the survey. - Consider using graphics when possible/appropriate. - Explore new ways to facilitate interactions between respondents and researchers. - Offer participants the opportunity to be contacted again to receive updates on projects of products being tested.

Guidelines for Determining Accuracy of Secondary Data

- Who gathered the data? - What was the purpose of the study? - What info was collected? - When what the info collected? - How was the info collected? - Is the info consistent with other info?

Sections of a Formal Research Request

1. Action: Decision maker should describe action to be taken on basis of the research. 2. Origin: Decision maker should state the events that let to a need for a decision. 3. Information: Decision maker should list the questions that they will need to have answered to take action. 4. Use: This section should explain how each piece of info will be used to help make the actual decision. 5. Target Groups and Subgroups: Describing from what group info must be gathered from. 6. Logistics: Approximations of the amount of money and time available must be included. 7. Comments: Any other relevant comments must be stated.

Steps in a Test Market Study

1. Define the objective. 2. Select a basic approach. 3. Develop detailed test procedures. 4. Select test markets 5. Execute the plan 6. Analyze the test results

Developing a Sampling Plan

1. Define the population of interest. 2. Choose a data collection method. 3. ID a sampling frame. 4. Select a sampling method. 5. Determine sample size. 6. Develop operational procedures for selecting sample elements. 7. Execute the operational sampling plan.

The Questionnaire Design Process

1. Determine survey objectives, resources, and constraints. 2. Determine the data collection method. 3. Determine the question response format. 4. Decide on the question wording. 5. Establish questionnaire flow and layout. 6. Evaluate the questionnaire. 7. Obtain approval of all relevant parties. 8. Pretest and revise. 9. Prepare final copy. 10. Implement the survey.

Market Research Process (Ex. 2.2)

1. Identification of the problem and statement of the research objectives. 2. Creation of the research design. 3. Choice of method of research. 4. Selection of the sampling procedure. 5. Collection of the data 6. Analysis of the data. 7. Writing and presentation of the report. 8. Follow-up.

Steps in Conducting a Focus Group (Ex. 4.2)

1. Prepare for the group: Select a focus group facility and recruit the participants. 2. Select a moderator and create a discussion guide. 3. Conduct the group. 4. Prepare the focus group report.

3 Basic Research Methods

1. Survey 2. Observation 3. Experiment

Characteristics of a Normal Distribution

1. The normal distribution is bell shaped and has only one mode. The mode is a measure of central tendency and is the particular value that occurs most frequently. (bimodal/2 mode distribution would have two peaks/humps) 2. Is symmetric about its mean.Three measures of central tendency are all equal (mean, median, mode). 3. A particular normal distribution is uniquely defined by its mean and standard deviation. 4. Total area under a normal curve is equal to one, meaning that it takes in all observations. 5. Area of a region under the normal distribution curve between any two values of a variable equals the probability of observing a value in that range when an observation is randomly selected from the distribution. 6. The area between the mean and a given number of standard deviations from the mean is the same for all normal distributions. The area between the mean and +/- one standard deviation takes in 68.26 percent of the area under the curve, or 68.26 percent of observations. This 'proportional property of the normal distribution' provides the basis for statistical inferences.

How can the researcher estimate the population standard deviation before selecting the sample?

1. Use results from prior survey. 2. Conduct a pilot survey. 3. Use secondary data. 4. Use judgment.

Overview of the Data Analysis Procedure

1. Validation and editing (quality control) 2. Coding 3.Data entry 4. Logical cleaning of data 5. Tabulation and statistical analysis

4 Major Factors to Consider Whether to Conduct a Test Market

1. Weigh the cost and risk failure against the probability of success and associated profits. 2. Consider the likelihood and speed with which competitor can copy your product and introduce it on a national level. 3. Consider the investment required to produce the product for the test market versus the investment required to produce the product in the quantities necessary for national roll out. 4. Consider how much damage an unsuccessful new product launch would inflict on the firm's reputation.

7 Questions are Basis of Customer Care Research (CCR)

1. What started you on the road to making this purchase? 2. Why did you make this purchase now? 3. What was the hardest part of this process? Was there any point where you got stuck? 4. When and how did you decide the price was acceptable? 5. Is there someone else with whom I should talk to get more of the story behind this purchase? 6. If you've purchased the product before, how does the story of your last purchase differ from this one? 7. At what point did you decide you trusted this organization and this person to work with your best interests?

Decision to Conduct Testing Marketing

2 Benefits: - Test market provides a vehicle by which the firm can obtain a good estimate of a product's sales potential under realistic marketing conditions. - The test should ID weaknesses of the product and the proposed marketing strategy.

Facial Action Coding Service (FACS)

43 muscles responsible for all human facial expression. Used to get the 'truth'. To measure initial gut reactions to a commercial/ ad, electrodes are attached to the side of a subject's mouth, above the eyebrow, and on 2 fingers.

Web Community

A carefully selected group of consumers who agree to participate in an ongoing dialogue with a particular corporation. All community interaction takes place on a custom designed website. During life of the community (6 months or more), members respond to questions posed by the corporation on a regular basis. Discussions usually take form of a dialogue.

Neural Network

A computer program that mimics the processes of a human brain and thus is capable of learning from examples to find patterns in data. Ex. American Express uses neural network to examine hundreds of millions of entries in it's database that tell how cardholders transact business. Result is a set of purchase propensity scores for each cardholder. AmEx uses info to match cardholders with merchant offers enclosed in their monthly statements.

The Portable People Meter (PPM)

A device that looks like a pager and is about as complex as a cellphone. Worn on the waist in put in pocket. Detects radio, television, smart phone viewing, and even cinema advertising.

Marketing Research Proposal

A document that presents the research objectives, design, time line, and cost of a project. I. Title Page II. Statement of research objectives III. Study Design IV. Areas of Questioning V. Data Analysis VI. Personnel Involved VII. Specifications and Assumptions VIII. Services IX. Cost X. Timing

Goal Orientation

A firm must be consumer-oriented only to the extent that it also accomplishes corporate goals. The goals of profit-making firms usually center on financial criteria, such as 15% return on investment.

Blogs

A frequent, chronological publication of personal thoughts and Web links. Companies use blogs to communicate directly with customers and other businesses. Marketing researchers have used blogs to monitor brands, track trends, profile customers, and I.D. unmet needs. Used as exploratory research tool for many formal survey projects.

Identifying a Sampling Frame

A list of the members or elements of the population from which units to be sampled are to be selected. May simply mean specifying a procedure for generating such a list.

Electroencephalograph (EEG)

A machine that measures electric pulses on the scalp and generate a record of electrical activity in the brain. Most versatile and sensitive procedure to detect arousal, often involves expensive equipment, a laboratory setting, and complex data analysis requiring software programs.

Construct Validity

A measure has a construct validity if it behaves according to the theory behind the prediction. Exist when measure correlates highly with other measures (convergent) or has a low correlation with the unrelated constructs of brand loyalty and a high level of aggressiveness. (discriminant) 2 statistical measures of construct validity: - Convergent validity: Reflects the degree of correlation among different measures that purport to measure the same construct. - Discriminant validity: Reveals lack of correlation among constructs that are supposed to be different.

Research Design

A plan for addressing the research objectives or hypotheses. In essence, the research develops a structure or framework to answer a specific research problem/opportunity.

Newsgroups

A primary means of communicating with other professionals and special interest groups on the internet. Functions much like a bulletin board for a particular topic or interest. Readers stop by that newsgroup to read messages left by other people, post responses, and send rebuttals. More than 250,000 newsgroups in existence with more being created daily.

Consumer Drawing

A projective technique in which respondents draw what they are feeling or how they perceive an object.

Experience Surveys

A second form of exploratory research. Involves talking with knowledgeable individuals, both inside and outside of the organization, who may provide insights into the problem. Rarely include a formal questionnaire, usually are discussion topics.

Before and After with Control Group Design

A true experimental design that involves random assignment of subjects or test units to experimental and control groups and pre- and post- measurements of both groups. Symbolically: Experimental Group: (R) O1 X O2 Control Group: (R) O3 O4

After-Only With Control Group Design

A true experimental design that involves random assignment of subjects or test units to experimental and control groups, but no pre-measurement of the dependent variable. Symbolically: Experimental Group: (R) X O1 Control Group: (R) O2

Observation Research and Virtual Shopping

A virtual store duplicates the distraction and clutter of an actual market. Consumers shop in an environment with a realistic level of complexity and variety. Researchers can set up and alter tests very quickly. Once images of a product are scanned onto the computer, the researcher can make changes to the assortment of brands, product packaging, pricing, promotions, and shelf space within minutes. Data collection is fast and error free. Production costs are low because displays are created electronically. Stimulation has a high degree of flexibility.

Validity

Addresses the issue of whether what the researcher was trying to measure was actually measured.

Internet Sampling

Advantages of internet interviewing: - Target respondents can complete the survey at their convenience. - Data collection is relatively inexpensive. - The interview can be administered under software control. - The survey can be completed quickly.

Online Survey Research

Advantages: - Rapid deployment and real time reporting. - Reduced costs. - Ready personalization. - High response rates. - Ability to contact the hard to reach. - Simplified and enhanced panel management. - External internet panels simplify life for research suppliers. Disadvantages: - Unrestricted internet sample. - Sample frame needed may not be available on the internet. - Lack of 'call back' procedures to clarify open-ended responses, potential for questionnaire programming errors, and lack of bandwidth.

Probability Samples

Advantages: - Researcher can be sure of obtaining info from a representative cross section of the population of interest. - Sampling error can be computed. - Survey results can be projected to the total population. Disadvantages: - More expensive than non probability samples of the same size.

Online Focus Groups

Advantages: Much lower cost (about half as much), faster turnaround time, respondents, ability to be geographically separate and be located anywhere, and intangibles such as increased openness on the part of the respondents when they don't have the moderator staring at them in the face, access to the hard-to-reach target population, and more efficient moderators. Disadvantages: Difficult to observe group dynamics, usually not possible to see nonverbal inputs, reduces client involvement, exposure to external stimuli is limited.

Online Individual Depth Interviewing

After respondents are recruited via a secured chat platform, each participant is given a private blog where they can create their online journal for the project. Over a period of days, respondents are given a series of questions to ponder on their blogs. The result is an immediate perfect transcription. Appropriate for when group dynamics are not important. Advantages: effective method for interviewing high level people like executives, richer content and deeper insight per respondents, and ability to view transcripts at anytime and make adjustments in the process by communicating with moderators.

Applied Research

All marketing research conducted for these purposes: To better understand the market, to find out why a strategy failed, or to reduce the uncertainty in management decision making. Most research undertaken by businesses is applied research because it must be cost-effective, and of demonstrable value to decision maker.

One Way Mirror Observation

Allows clients to observe the group discussion as it unfolds.

Galvanic Skin Response (GSR)

Also known as 'electro-dermal response', is a change in the electric resistance of the skin associated with activation responses. A small electric current of constant intensity is sent into the skin through electrodes attached to the palmar side of the fingers. The changes in voltage observed between electrodes indicate the level of stimulation. Because equipment is portable and not expensive, this is the most popular way to assess emotional reaction to stimulus. Primarily used to measure response to advertising but is also sometimes used in packaging research.

Basic Research

Also known as Pure Research, attempts to expand the frontiers of knowledge. It is not aimed at a specific problem. It is conducted to validate and existing theory or to learn more about a concept or phenomenon. In long run, helps us understand more about the world we live in. Findings usually cannot be implemented by managers in the short run. High-tech firms (Microsoft, HP, IBM, etc.) use basic research to focus on finding ways to solve real-world problems.

Multi Stage Area Sampling

Also referred to as multistage area probability sampling, which involves three or more steps. Samples of this type are used for national surveys or surveys that cover large regional areas. Here, the researcher randomly selects geographic areas in progressively smaller units.

Standard Error of the Mean

Also referred to as standard deviation of a distribution of sample means, computed as follows: S(x bar) = standard deviation / (square root of 'n') n = sample size

Temporal Sequence

An appropriate causal order of events. The effect follows closely the hypothesized cause. It is one criterion that must be met for causality.

Judgment Samples

Applied to any sample in which the selection criteria are based on the researcher's judgments about what constitutes a representative sample.

Word Association Test

Are among most practical and effective projective toolds for marketing researchers. An interviewer reads a word to a respondent and asks them to mention the first thing that comes to mind.

Pre-Experimental Designs

Are research designs that do not include basic elements required in a true experimental design. Studies using pre-experimental designs generally are difficult to interpret because such designs offer little or not control over the influence of extraneous factors. Are often not much better than descriptive studies when it comes to making causal inferences. Frequently used in commercial testing marketing because of simplicity and inexpensiveness.

Pilot Studies

Are surveys using a limited number of respondents and often employing less rigorous sampling techniques than are employed in large, quantitative studies.

Mystery Shoppers

Are used to gather observational data about a store (ex. Are the shelves neatly stocked?) and to collect data about customer-employee interactions. Classified as an observational research marketing method, even though communication is involved. Estimated that 70 percent of America's national retailers use this technique.

Paired Comparisons

Asks a respondent to pick one of 2 objects from a set, based on some stated criteria. The respondent therefore makes a series of paired judgments between objects.

Internal Consistency Reliability

Assesses the ability to produce similar results when different samples are used to measure a phenomenon during the same time period. Theory rests on the concept of equivalence.

Likert Scale

Avoids the problem of developing pairs of opposite/ dichotomous adjectives. The scale consist of a series of statements expressing either a favorable or an unfavorable attitude toward the concept under study. The respondent is asked to indicate their level or agreement or disagreement to each statement by assigning in a numerical score. Steps in building the scale: 1. ID concept or activity to be scaled. 2. Assemble large number of statements (75-100) concerning the public's sentiments toward the concept. 3. Each test item is classified as favorable or unfavorable. 4. In pretest, respondent should indicate whether they agree or not with every item (A. Strongly agree B. Agree C. Undecided D. Disagree E. Strongly disagree) 5. Each response is given a numerical weight. 6. Total attitude score is represented by the algebraic summation of weights associated with the items checked. 7. After seeing results of pretest, researcher selects only items that appear to discriminate well between high and low total scorers. 8. 20 to 25 items are selected to be best in high and low scorers. 9. Steps 3 to 5 are repeated in the main study.

Balanced vs Non Balanced Scale

Balanced scale: Has the same number of positive and negative categories. Used when researchers expect a wide range of opinions. Non balanced scale: Weighted toward one end or the other. Used if past research/ a preliminary study has determined that most opinions are positive or negative.

Research Request

Before conducting a research project, a company might require approval of a formal research request. Moderate and large-size retailers, manufacturers, and nonprofit organizations often use the Research Request as a basis for determining which projects will be funded.

Semantic Differential Scale

Begins with determination of a concept to be rated, such as the image of a company, brand, or store. The researcher selects opposite pairs of words or phrases that could be used to describe the concept. Respondents the rate the concept on a scale, usually 1-7. The mean is computed and plotted on as an image or profile. Quick and efficient at examining strengths and weaknesses of a product. Disadvantage: the halo effect.

Decision to Conduct Marketing Research (When Not To)

Best not to conduct marketing research when: - Resources are lacking - Research results would not be useful - The opportunity has passed - The decision already has been made - Managers cannot agree on what they need to know to make a decision - Decision-making info already exist - The costs of conducting research outweigh the benefits

State Laws

California's Notice of Security Breach Law: If a company/ agency that has collected the personal info of a California resident discovers that non-encrypted info has been taken by an unauthorized person, the company/ agency must inform the resident.

Sentence and Story Completion Tests

Can be used in conjunction with word association tests. Respondent is furnished with an incomplete story or group of sentences and asked to complete it.

Statistical Control

Can be used to account for extraneous causal factors if these factors can be IDed and measured throughout the course of the experiment. Procedures such as analysis of covariance can adjust for the effects of a confounded variable on the dependent variable by statistically adjusting the value of the dependent variable for each treatment condition.

Importance of Marketing Research

Can be viewed as playing 3 functional roles: descriptive, diagnostic, and predictive.

Response Bias

Can result from deliberate falsification or unconscious misrepresentation.

Randomization

Carried out by randomly assigning subjects to treatment conditions so that extraneous causal factors related to subject characteristics can be reasonably assumed to be represented equally in each treatment condition, thus cancelling out extraneous factors.

Scaled Response Questions

Closed ended questions where the response choices are designed to capture intensity of feeling.

Regression Analysis Measures

Coefficient determination or R^2 is a key output of regression analysis. Can assume values from 0 to 1 and provide a measure of the percentage of the variation in the dependent variable that is explained in the independent variable.

Field Experiments

Conducted outside the lab in an actual market environment, Test markets are frequently used as a type of field experiment. Field experiments often solve the problem of realism of the environment but often open up the problem of control of spurious factors that might influence the dependent variables, such as actions of competitors, weather, economy, societal trends, and political climate. Have problems with internal validity.

Descriptive Studies

Conducted to answer who, what, when, where, and how questions. Implicit in descriptive research is the fact that management already knows/ understands the underlying relationships among variables in the problem. Can shed light on associations/ relationships, it helps researchers select variables for causal study. Ex. Descriptive study for Starbucks may include demographics and lifestyle characteristics of typical, light, and heavy patronizers of their coffee, baked goods, and sandwiches.

Programmatic Research

Conducted to develop marketing options through market segmentation, market opportunity analysis, or consumer attitude and product usage studies. Arises from management's need to obtain a market overview periodically because current info is needed to develop viable marketing options.

Considerations in Selecting a Scale

Considerations include: Construct being measured, type of scale, balanced vs non balanced scale, number of scale categories, and forced vs non forced choice.

Focus Groups

Consists 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. A distinction is made between group dynamics and group interviewing.

Photo Sorts

Consumers express their feelings about brands by manipulating a specially developed photo deck depicting different types of people, from business executives to college students. Respondents connect the individuals in the photos to the brands they think each would use.

Commercial Online Panels

Created for the use of multiple projects by many different companies. The companies providing access to online panels have invested in the pre-recruitment of people who opted to participate in online market research surveys. Some online panels are for use by a specific industry and may have a few thousand panel members. When people join most online panels, they answer an extensive profiling questionnaire that records demographic, lifestyle, and psychographic info, typically with hundreds of dimensions.

Respondent Cooperation

Critical to success of research process. 2 Incentive models: Sweepstakes model and the Pay all model - Sweepstakes: Offers survey participants a chance to be entered into a drawing for a prize. - Pay all: Pay each respondent a small incentive for their time and participation. Study found that the pay all model is more effective than the sweepstakes model.

Tabulation and Statistical Analysis (Cross Tabulation)

Cross tabulation: Represents a simple to understand, yet powerful, analytic tool. Many marketing research studies go no further than cross tabulations in terms of analysis. The idea is to look at the responses to one question in relation to the responses to one or more other questions. When considering the set up of cross tabulation tables and the determination of percentages within them: - Selection of an appropriate base. - Three different percentages may be calculated for each cell in a cross tabulation table: column, row, and total percentages. - Common way of set up is to use columns to represent factors such as demographics and lifestyle characteristics, which are expected to be predictors of the state of mind, behavior, intentions as rows of the table.

Test Market Studies

Designed to provide info on the following issues: - Estimates of market share and volume. - Effects that the new product will have on sales of a similar product already marketed by the company. Referred to as the 'cannibalization rate'. - Characteristics of the consumer who buy the product. - Behavior of competitors.

Gender and Age Recognition System

Digital signs placed in high traffic public places, such as malls or airports, that can recognize gender and approximate age of consumers. Also counts the number of people that pass the digital sign. Uses an algorithm that draws on a database on thousands of faces as reference. It looks at distinguishing points of the face.

Cost of Test Marketing

Direct Costs: - Production of commercials - Payments to an advertising agency for services - Media time - Syndicated research info - Customized research info and associated data analysis - Point of purchase materials - Coupons and sampling Indirect Costs: - Cost of management time spent on the test market - Diversion of sales activity from existing products - Possible negative impact of a test market failure - Possible negative trade reactions to products - Cost of letting competitors know what the firm is doing

Stratified Sampling

Distinguished by the following procedural steps: - The original, or parent, population is divided into two or more mutually exclusive and exhaustive subsets (Ex. Male or female) - Simple random samples of elements from the two or more subsets are chosen independently of each other. Steps in implementing a properly stratified sample: 1. ID salient (important) demographics or classification factors. Factors that are correlated with the behavior of interest. (Ex. gender: male or female) 2. Determine what proportions of the population fall into the various subgroups under each stratum. (Ex. If gender is a salient factor, determine what proportion of population is female or male) See next couple of slides for proportional allocation and etc. 3. Select separate simple random samples from each stratum.

Experiments

Distinguished by the researcher's changing one or more independent variables (price, package, design, shelf space, advertising theme, etc.) and observing those effects on the dependent variable (usually sales). Objective is to measure causality.

Evaluative Research

Done to assess program performance. Includes tracking advertising recall, doing organizational image studies, and examining customer attitudes on a firms's quality of service.

Analogies

Draws a comparison between 2 items in terms of similarity.

Random Error or Random Sampling Error

Error that results from chance variation.

Evaluating Web Pages

Evaluate the quality of the webpage: 1. What can the URL tell you? Is it somebody's personal page? What type of domain does it come from? 2. Scan the perimeter of the page looking for: "About Us", "Philosophy", "Biography", etc. Look for the date "Last Updated." Questions to ask: Who wrote the page? What are the implications? 3. Look for indicators of quality information: "Links", "Additional Sites", "Related Links", etc. Texts includes footnote numbers or links referring to documentation. Look at the publisher page.

Criterion Related Validity

Examines the ability of a measuring instrument to predict a variable that is designated a criterion. 2 subcategories of criterion related validity: - Predictive validity: Is the extent to which a future level of a criterion variable can be predicted by a current measurement on a scale. - Concurrent validity: Is concerned with the relationship between the predictor variable and the criterion variable, both of which are assessed at the same point in time.

Experimental Setting

Experiments can be conducted in a laboratory or field setting. Most experiments in the physical sciences are conducted in a lab setting while many marketing experiments are conducted in a field setting.

Government Action to Protect Consumers from Identity theft

Federal Laws: - Gramm-Leach-Bliley Act: (Financial Services Modernization Act) aimed at financial companies. Requires those companies to tell their customers how they use their personal info and to have policies that prevent fraudulent access to it. - Health Insurance Portability and Accountability Act: aimed at the healthcare industry. Limits disclosure of individual's medical info and imposes penalties on organizations the violate privacy rules. - Fair Credit Reporting Act: (FCRA) enforced by Federal Trade Commission, promotes accuracy in consumer reports, and is meant to ensure the privacy of the info in them. - Children's Online Privacy Protection Act: (COPPA) aims to give parents control over what info is collected from their children online and how such info can be used.

Validation

First, the researcher must determine, to the extent possible, that each of the questionnaires to be processed presents a valid interview. 'Validation' is defined as the process of ascertaining that interviews were conducted as specified. No assessment is made regarding the validity of the measurement. Goal of 'validation' is solely to detect interviewer fraud or failure to follow key instructions.

Analyze the Test Results

Focuses on 4 areas: - Purchase data: Most important data. Levels of initial purchases throughout the course of the experiment provide indication of how well the advertising program worked. - Awareness data: How effective were the media plans in creating awareness? Do consumers know the product's key features? etc. - Competitive response: Actions of competitors may provide indication of what they will do if the product moves into national distribution. - Source of sales: Important to determine where sales are coming from. Which brands did people who purchased the test product previously purchase? Acts as a true indicator of real competitors.

Hermeneutic Research

Focuses on interpretation as a basis of understanding the consumer through conversation. The researcher answers participant's questions and, as in traditional methods, researchers only question the respondent. There are no predetermined questions, but questions arise spontaneously as the conversation unfolds.

Iceberg Principle

Focusing on the symptoms and not the true problem.

Central Limit Theorem

For any population, regardless of it distribution, the distribution of sample means or sample proportions approaches a normal distribution as sample sizes increases.

Correlation Analysis

For metric data: Pearson's Product --- Moment Correlation Correlation is the degree to which changes in one variable (dependent) are associated with changes in another. When the relationship is between two variables, it is called simple/ bivariate correlation analysis. With metric data, Pearson's product/ moment correlation may be used.

Secondary Data Analysis

Form of exploratory research. Data that has been gathered for some purpose other than the one at hand.

Controlling Extraneous Variables

Four basic approaches to controlling extraneous variables: - Randomization - Physical control - Design control - Statistical control

Measure of Dispersion

Frequently used measures of dispersion include standard deviation, variance, and range. Indicates spread of the data. Formula for computing standard deviation for a sample of observations: S = square root of ( (Summation of n with i = 1(Xi - Xbar)^2) / (n - 1) ) Where: S = sample standard deviation Xi = value of the i th observation Xbar = sample mean n = sample size Variance is calculated by using the formula for standard deviation with the square root sign removed. Range is equal to the maximum value minus the minimum value.

Cluster Analysis

General approach involves measuring the similarities among people or objects in regard to their values on the variables used for clustering. Similarities among people or objects being clustered normally determined on basis of some type of distance.

Traditional (Synchronous Group)

Goal is to get 8 to 10 participants online at the same time, for a moderator to send questions to participants, and the participants provide their comments.

Examples of Extraneous Variables

History, maturation, instrument variation, selection bias, mortality, testing effects, and regression to the mean.

Multiple Time-Series Designs

If a control group can be added to an interrupted time series design, then researchers can be more certain that their interpretation of the treatment effect.

Rule of Thumb

If the researcher determines that the sample size requested is not adequate to support the objectives of the proposed research, then she or he has a professional responsibility to present arguments for a larger sample size to the client and let the client make the final decision.

Spurious Association

Important issue in studying causality is recognizing the possibility of Spurious Association, in which other variables are actually causing changes in dependent variables.

Ordinal

In addition to identification, provides info about the relative amount of some characteristics possessed by an event, object, etc. Operations: Determination of greater or lesser Typical Descriptive and Usage: Ranking/ ratings Statistics: Median (mean and variance metric)

Human vs Machine Observers

In some situations, machine observers replace humans because they can do the job less expensively, more accurately, and readily. (ex. Traffic counting devices, electronic scanners in retail stores)

Determining Sample Size

In the case of non probability, researchers tend to rely on such factors as available budget, rules of thumb, and number of subgroups to be analyzed in their determination of sample size. However, with probability samples, researchers use formulas to calculate the sample size required, given target levels of 'acceptable error' (the difference between sample result and population value) and 'levels of confidence' (the likelihood that the confidence interval --- sample result plus or minus the acceptable error --- will take in the true population value).

Door to Door Interviews

In which consumers are interviewed in person in their own homes. Use of this method has declined steeply since 1970's. It is still used in government research (US census).

Focus Groups

In-depth discussions, usually consisting of 8 to 12 participants, which are lead by a moderator and are generally limited to one concept, idea, or theme.

Secondary Data Originating from the Company

Includes annual reports, reports to stockholders, sales data, customer profiles, purchase patterns, product testing results, and house periodicals composed by company personnel for communication to employees, customers, or others.

Descriptive Function

Includes gathering and presenting statements of fact. Q's: What is the historic sales trend in the industry? What are the consumers' attitudes and beliefs toward a product?

Outside Sources of Secondary Data

Includes innumerable government (federal, state, and local) departments/ agencies that compile and publish summaries of business data, as well as trade and industry associations, business periodicals, and other news media that regularly publish studies on the economy, specific industries, and individual companies.

Panel Management

Includes recruitment methods, respondent participation, panel management practices, and types of incentives offered.

Ratio

Incorporates all properties of nominal, ordinal, and interval scales plus an absolute zero. Operations: Determination of equality of ratios Typical Descriptive and Usage: Preferred measure when precision instruments are available Statistics: Geometric mean/ harmonic mean

Population Size and Sample Size

Independence assumption: A normal presumption that assumes sample elements are drawn independently of one another. This assumption is justified when the sample is small relative to the population. Is not appropriate when the samples is a relatively large proportion of the population (5 percent or more). Refer to the formula for standard error of the mean: S(x bar) = standard deviation / (square root of 'n') n = sample size For a sample that is 5 percent or more of the population, the independence assumption is dropped, producing the following formula: Standard deviation / square root of 'n'( square root of (N - n) / (N - 1)) The factor (N - n) / (N - 1) is referred to as the finite population correction factor (FPC). In these situations in which the sample is larger (5 percent or more) in relation to the population, the researcher can appropriately reduce required sample size using the FPC. This calculation is made using the following formula: n' = nN / N + n - 1 Where: n' = revised sample size n = original sample size N = population size Ex. If the population has 2,000 elements and the original sample size is 400, then: n' = 400(2,000) / 2000 + 400 - 1 n' = 800,000 / 2399 n' = 333 With FPC adjustment, a sample of only 333 is needed, rather than the original 400.

Executive Interviews

Industrial equivalent of door-to-door interviews. Involves interviewing business people at their offices concerning industrial products or services. This method is expensive, long waits are common, and cancellations are frequent.

Group Dynamics

Interaction associated with Group Dynamics is essential to the success of focus group research. This interaction is the the reason for conducting research with a group rather than with individuals.

Interrupted Time-Series Designs

Involve repeated measurements of an effect both before and after a treatment is introduced that interrupts previous data patterns. Has a greater interpretability than the previous designs because the many measurements allow more understanding of extraneous variables. 2 fundamental weaknesses: The experimenter's inability to control history and the possibility of interactive effects of testing apprehension resulting from repeated measurements taken on test units.

Survey Research

Involves an interviewer (except in mail and internet) who interacts with respondents to obtain facts, opinions, and attitudes. A questionnaire is used to ensure an orderly and structured approach to data gathering.

Editing

Involves checking for interviewer and respondent mistakes. The editing process for paper surveys involves manual checking for problems such as: - Whether the interviewer failed to ask certain questions or record answers for certain questions. - Whether skip patterns were followed. - Whether the interviewer paraphrased respondent's answers to open ended questions. The person doing the editing must make judgment calls regarding substandard responses to open ended questions. Editing process is extremely tedious and time consuming.

One-Shot Case Study Design

Involves exposing test units to the treatment variable for some period of time and then taking a measurement of the dependent variable. 2 basic weaknesses of this design: No pretest observations are made of the test units that will receive treatment and no control group of test units that did not receive the treatment is observed. Therefore, lack validity.

Sampling Distribution of the Mean

Is a conceptual and theoretical probability distribution of the means of all possible samples of a given size drawn from the given population.

Symphony IRI Consumer Network

Is a continuous household purchase panel that receives purchasing data from the National Consumer Panel (NCP), a joint venture between Symphony IRI and the Nielson Companies. Households are recruited by the NCP and given incentives to record all of their UPC based purchases with a hand held in home scanning device. Data available from the panel include: - Average amount of product purchased by each household during a specific period. - Total projected category or product volume sold in the specific period. - Projected number of households that purchased the category or product within the specific time. - Average number of shopping trips when the product was purchased. - Average volume sold for the product per purchase occasion.

Sampling Frame

Is a list of population elements or members from which units to be sampled are selected.

Stapel Scales

Is a modification of the semantic differential. A single adjective is placed in the center of the scale, which is typically 10 points ranging from +5 to -5. The technique is designed to measure both the direction and the intensity of attitudes simultaneously.

Sampling Distributions of the Proportion

Is a relative frequency distribution of the sample proportions of a large number of random samples of a given size drawn from a particular population. Has the following characteristics: - Approximates a normal distribution. - Mean proportion for all possible samples is equal to the population proportion. - Standard error of a sampling distribution of the proportion can be computed with the formula: Sp = square root of ( (P(1 - P) / n) ) Sp = Standard error of sampling distribution of proportion P = estimate of population proportion n = sample size

Qualitative Research (Ex. 4.1)

Is a term used loosely to refer to research whose findings are not subject to heavy quantification or quantitative analysis. - Types of questions: Probing - Sample size: Small - Amount of info from each respondent: Substantial - Requirements for administration: Interviewer with special skills - Type of analysis: Subjective. interpretive - Hardware: Sound recorders, projection devices, video recorders, pictures, discussion guides - Degree of replication: Low - Researcher training: Psychology, sociology, social psychology, consumer behavior, marketing, marketing research - Type of research: Exploratory

Regression Analysis

Is an extension of bivariate regression. Instead of fitting a straight line to observations in a 2D space, multiple regression analysis fits a plane to observations in a multi-dimensional space. The output obtained and the interpretation are essentially the same as for bivariate regression. General equation for regression is as follows: Y = a + b1 X1 + b2 X2 + b3 X3 + . . . bn Xn Where: Y = dependent or criterion variable a = estimate constant b1 - bn = Coefficients associated with the predictor variables so that a change of one unit in X will cause change of b1 units in Y; values for coefficients are estimated from regression analysis. X1 - Xn = predictor (independent) variable that influence the dependent variable.

Systems Orientation

Is an organized whole -- or a group of diverse units that form an integrated whole -- functioning or operating in unison.

Equivalent Form of Reliability

Is determined by measuring the correlation of the scores on the two instruments (questionnaires). Theory is same as test retest reliability approach. Primary difference between the two approaches is that the retest uses the same instrument and the equivalent form uses a highly similar, but different measuring instrument. Problems: It is very difficult/ impossible to create 2 totally equivalent forms. If equivalence can be achieved, it may not be worth the time, trouble, or expense.

Observation Research

Is examining patterns of behavior as opposed to asking consumers why they do what they do.

Test Re-Test Reliability

Is obtained by repeating the measurement with the same instrument, approximating the original conditions as closely as possible. Theory is that if random variations are present, they will be revealed by differences in the scores between the 2 tests.

Content Validity

Is representative of the content of measurement instrument.

Design Control

Is the control of extraneous factors by means of specific types of experimental designs developed for this purpose.

Reliability

Is the degree to which measures are free from random error and therefore provide consistent data. The less error there is, the more reliable the data is.

One-Group Pre-test - Post-Test Design

Is the design employed most frequently for testing changes in established products or marketing strategies. The fact that the product was on the market before the changes provides the basis for the pretest measurement. History and maturation are both threats to the internal validity of this design.

Chance Variation

Is the difference between the sample value and the true value of the population mean. This error cannot be eliminate, but reduced by increasing the sample size.

Refusal Rate

Is the percentage of persons contacted who refuse to participate in a survey. Today, overall refusal rate is approximately 60%.

Behavioral Targeting

Is 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.

Data Mining

Is the use of statistical and other advanced software to discover non-obvious patterns hidden in a database. The objective is to identify patterns that marketers can use in creating new strategies and tactics to increase a firm's profitability.

Sample Size for Qualitative Research

Issues usually arise when making decisions about the number of traditional focus groups, individual depth interviews, or online bulletin board focus groups to conduct. Experts suggest talking to 20-30 people in a qualitative setting, the general pattern of responses begins to stabilize.

4 Basic Levels of Mystery Shopping

Level 1: The mystery shopper conducts a mystery telephone call when they phone the client location and evaluate the level of service received by following a scripted conversation. Level 2: The mystery shopper visits an establishment and makes a quick purchase; little or no customer-employee interaction is required. (ex. Gas, a hamburger, a lottery ticket.) Level 3: The mystery shopper visits an establishment and ,using a script or scenario, initiates a conversation with a service or sales representative. Usually does not involve an actual purchase. (ex. Discussing different cell phone packages.) Level 4: The mystery shopper performs a visit that requires excellent communication skills and knowledge of the product. (ex. Discussing a home loan, process of purchasing a new car, visiting apartment complexes.)

Graphic Representations of Data

Line charts: Perhaps the simplest for of graphs. Useful for presenting a given measurement taken at several points over time. Pie charts: Frequently used. Appropriate for displaying marketing research results in a wide range of situations. Bar charts: Most flexible. Anything that can be shown with a line or pie chart can also be shown on a bar chart. Types of bar charts: - Bar chart - Multiple row, 3D bar chart: Provides most visually appealing way to present cross tabulation info.

People Reader

Looks like a lamp. When respondents sit in front of it, they are not aware that it is simultaneously recording both their reading material and the activity of their eyes.

Test Market

Loose term referring to any research that involves testing a new product or change in an existing marketing strategy in a single market, group of markets, or region of the country through use of experimental or quasi-experimental designs.

Evaluate Reliability and Validity of the Measurement

M = A + E M is measurement A is accuracy E is error Two scores on a measurement scale can differ for a number of reasons. Only the first of the following 8 reasons does not involve error. A researcher must determine whether any of the remaining 7 sources of measurement differences are producing random or systematic error: 1. A true difference in characteristics being measured. 2. Differences due to stable characteristics. 3. Differences due to short term personal factors. 4. Differences caused by situational factors. 5. Differences resulting from variations in administering the survey. 6. Differences due to the sampling of items included in the questionnaire. 7. Differences due to lack of clarity in the measurement instrument. 8. Differences due to mechanical or instrument factors.

Develop Detailed Test Procedure

Manufacturing and distribution decisions must be made to ensure that adequate product is available. The basic positioning approach must be selected, the actual commercials must be developed, a pricing strategy must be chosen, a media plan must be developed, and various promotional activities must be specified.

Focus Group Length

Many managers today prefer shorter focus groups (around an hour). Yet the average is still about 90 minutes.

Translate the Management Problem into a Marketing Research Problem

Marketing Research Problem: Specifies what info is needed to solve the problem and how that info can be obtain efficiently and effectively. Marketing Research Objective: Is the goal statement. Defines the specific info needed to solve the marketing research problem. Management Decision Problem: Action oriented and tend to be much broader in scope than marketing research problems.

Nature of Applied Research

Marketing research studies can be classified into 3 broad categories: programmatic, selective, and evaluative.

Measure of Central Tendency

Mean: Is properly computed only from interval or ratio (metric) data. Computed by adding the values for all observations for a particular variable and dividing the result by the number of observations. Median: Computed for all types of data except nominal data. Calculated by finding the value below which 50 percent of the observations fall. Mode: Computed for all types of data (nominal, ordinal, interval, or ratio). Determined by finding the value that occurs most frequently.

Consumer Orientation

Means firms strive to identify the people or firms most likely to buy their product (target market) and to produce a good or offer a service that will meet the needs of target customers in the face of competition.

Unrestricted Internet Sample

Means that anyone who wishes to complete the questionnaire can do so. It is fully self selecting and probably representative of no one except web surfers. The problem gets worse if the same user can access the questionnaire over and over.

Stability

Means that very few differences in scores are found between the first and second administrations of the test. The measure instrument is said to be stable. Problems with this test: Difficult to locate respondent for the 2nd test, first measurement may alter person's response on second measurement, environmental or personal factors may change, causing second measurement to change.

Related Samples

Measurement of the variable of interest in one sample may influence measurement of the variable in another sample.

Traffic Counters

Most common and popular machines in observation research. Measures vehicular flow over a particular stretch of roadway. Outdoor advertisers rely on traffic counters to determine the number of exposures per day to a specific billboard. Retailers use the info to ascertain where to locate a particular type of store.

Self-Administered Questionnaires

Most frequently employed in high traffic locations such as shopping malls or in captive audience situations such as classrooms or airplanes. Disadvantage: No one is present to explain things to the respondent and clarify responses to open-ended questions.

Direct vs Indirect

Much of observation carried out for marketing research is direct observation of current behavior. (ex. Observing what items women in Oregon and Texas carried in their purses) Pictures can also be used to see what people have done is certain situations.

Natural vs Contrived Situations

Natural: Those that have no idea they are being observed. The observer plays no role in the behavior of interest. (ex. Counting how many people use the drive-in window at a particular bank during certain hours.) Stimulated Environment: Participants have at least some idea that they are being observed. (ex. Recruiting people to shop in a stimulated supermarket so their behavior can be closely observed. Testing for effectiveness of product displays.) Contrived: Enables researchers to better control extraneous variables that might have an impact on a person's behavior. Also tends to speed up the data-gathering process. More observations can be collected in same amount of time, result would either be a larger sample or faster collections of targeted amount of data. The latter would also lower project cost. Disadvantage is that it is artificial, thus the observed behavior would be different than what would have happened in a natural setting.

Focus Group Moderator

Needs 2 sets of skills: Must be able to conduct group properly and must have good business skills in order to interact with the client.

Number of Scale Categories

Number of categories can't be too small (ex. good, fair, poor) or too big because it goes beyond a person's ability to measure their attitude.

Projective Test

Objective is to delve below surface responses to obtain true feelings, meanings, and motivations.

Machine Observation

Observation by machines includes things like traffic counters, physiological measurement devices, opinion and behavior management devices, and scanners.

Contamination

Occurs when buyers from outside the test area come into the area to purchase the product being tested, thereby distorting the results of the experiment.

Selection Error

Occurs when sampling procedures are incomplete or improper or when appropriate selection procedures are not properly followed.

Surrogate Info Error

Occurs when there is a discrepancy between the info actually required to solve a problem and the info being sought by the researcher. It relates to general problems in research design with defining the actual problem.

Measurement Error

Occurs when there is variation between the info being sought and the info actually being obtained by the measurement process.

Non-response Bias/ Errors

Occurs when: - Person cannot be reached at a particular time. - A potential respondent is reached but cannot or will not participate at that time. - A person is reached but refuses to participate in the survey.

Graphic Rating Scale

Offer respondents a graphic continuum, typically anchored by 2 extremes. Respondents are instructed to mark their responses on the graphic continuum. Scores are attained by dividing the line into as many categories as desired and assigning a score based on the category to which the mark has been placed.

Mobile Internet Research

Offers: (Advantages) - An alternative way to conduct research with hard to reach groups. - Increased respondent cooperation from all demographic groups. - Immediate feedback on research questions concerning marketing campaigns, ad testing, etc. - Cost savings. Fast replies, shorter project completion time. - Use as a mobile recruiting tool to direct respondents to online surveys. - Reaches people on the go. Disadvantages: - Questionnaires must be short. - Question types are limited, as is the length of questions. Higher incentives must be offered to remove the financial burden of receiving and completing mobile surveys from the respondent.

Systematic Sampling

Often used as a substitute for simple random sampling. Produces samples almost identical to those generated via simple random sampling. Simpler, less time consuming, and less expensive than random sampling. To obtain a systematic sample, the researcher first numbers the entire population. Then the researcher determines the skip interval and selects names based on this interval. To compute the skip interval: Skip interval = Population size/ sample size A random starting point should be used in systematic sampling. From that point, employ the skip interval until the desired sample size had been reached.

The Delphi Method

Often used in new product development when firms are looking for creative ideas to incorporate into products or services. Relies on people who are experts in some area (product development researchers, marketing managers, professional people, magazine editors, executives, etc.). Involves a number of rounds of data collection. First round in unstructured, second is a questionnaire. Key characteristics of Delphi Method are anonymity, iteration, feedback, and aggregation of group responses.

Limitations of Qualitative Research

One drawback relates to the fact that marketing successes and failures many times are based on small differences in attitudes or opinions about marketing mix, and qualitative research does not distinguish those small differences as well as large-scale quantitative research does. A second limitation is that they are not necessarily representative of the population of interest to the researcher.

Tabulation and Statistical Analysis (One Way Frequency Table)

One way frequency table: Most basic tabulation which shows the number of respondents who gave each possible answer to each question. Issue of what base to use for the percentage of each table is dealt with in the following ways: 1. Total respondents 2. Number of people asked the particular question 3. Number of people answering the question

Open vs Disguised

Open Observation: When a person knows they are being observed. They may behave differently. Presence of an observer can offer potential bias. Disguised Observation: Process of monitoring people who do not know they are being observed. Common form is observing from behind a one-way mirror.

Panel Recruitment

Open online panel recruitment: Intercepting people as they surf the internet through ads. Allows any person who has access to the internet to 'self-select' and enroll in a market research panel. Provides the benefit of building a panel quickly. Drawback is lack of control over who is recruited. Closed online panel recruitment: By invitation only, invites only pre-validated individuals, or individuals who share known characteristics, to enroll in a market research panel. Most often, accomplished by inviting customers from large, highly trusted leading brands who collectively have a large, diverse base of customers in a given population.

Recognize the Problem or Opportunity

Opportunity Identification: Using marketing research to find and evaluate new opportunities. Q: Should we change the existing marketing mix? If so, how? A: Marketing research can be used to evaluate products, services, promotion, distribution, and pricing alternatives.

Number of Subgroups Analyzed

Other things being equal, the larger the number of subgroups to be analyzed, the larger the required total sample size. It has been suggested that a sample should provide, at a minimum, 100 or more respondents in each major subgroup and 20 to 50 respondents in each of the less important subgroups.

Physical Control

Physical control of extraneous factors may involve somehow holding constant the value or level of the extraneous variable throughout the experiment. Another approach is matching respondents in regard to important personal characteristics before assigning them to different treatment conditions. Goal is to make sure that there are no important differences between characteristics of respondents in the test and control groups.

Mall-Intercept Interviews

Popular method for conducting personal interviews. Shoppers are intercepted in public areas of shopping malls and either interviewed on the spot or asked to come to a permanent interviewing facility at the mall.

Conjoint Analysis

Popular procedure used by marketers to help determine what features a new product or service should have and how it should be priced.

Interval

Possesses all properties of nominal and ordinal scale plus equal intervals between consecutive points. Operations: Determination of equality of intervals Typical Descriptive and Usage: Preferred measure of complex concepts/ constructs Statistics: Mean/ variance

Social Media and User-Generated Content

Previously, marketers created a message and then one or a series of traditional media, TV, print, radio, and billboards to deliver that message to the target market. Now, more people than ever are participating in blogs, forums, online communities, product/service reviews, and social media sites. Marketers are tapping into these huge streams of data to determine what users think about their products and services as well as their competition.

Processing Error

Primarily due to mistakes that occur when info from survey documents is entered into the computer.

Convenience Samples

Primarily used for reasons of convenience just as its name implies.

Online Qualitative Research

Primary form of online qualitative research is focus groups. These come in 2 forms in the online world: Traditional, sometimes referred to as synchronous groups, and online bulletin boards, sometimes referred to as asynchronous groups.

Selecting the Sampling Report

Probability Sample: A sample for which every element in the population has a known nonzero probability of being selected. Allows the researcher to estimate how much sampling error is present in a given study. Non-probability Sample: Are those in which the chances of selection for the various elements in the population are unknown. Researchers cannot statistically calculate the reliability of a non-probability sample.

Simple Random Sampling

Probability of selection = sample size/ population size Ex: If population size is 10,000 and sample size is 400, the probability of selection is 4 percent. 0.04 = 400/ 10,000 If a sampling frame (listing of all elements of the population) is available, the researcher can select a 'simple random sample' as follows: - Assign a number to each element or the population. A population of 10,000 elements would be numbered from 1 to 10,000. - Using a table of random numbers, begin at some arbitrary point and move up, down, or across until 400 (sample size) five-digit numbers between 1 to 10,000 have been chosen. The numbers selected from the table ID specific population elements to be included in the sample.

Determining Sample Size for Probability Samples

Process involves financial, statistical, and managerial issues. As a general rule, the larger the sample is, the smaller the sampling error. However, larger samples cost more money and resources available for a project are always limited. Cost of increasing sample size tend to rise on a linear basis, sampling error decreases at a rate equal to the square root of the relative increase in sample size.

Disproportional (Optimal) Allocation

Produced most efficient samples and provides the most precise or reliable estimates for a given sample size. This approach requires a double weighing scheme. Under this scheme, the number of sample elements to be taken from a given stratum is proportional to the relative size of the stratum and the standard deviation of the distribution of the characteristic under consideration for all elements in the stratum.

Laboratory Experiments

Provides a number of important advantages like the ability to control extraneous causal factors, such as temperature, light, humidity, and so on - and focus on the effect of a change in A and B. In the lab, researchers can effectively deal with the third element of proving causation and focus on the first two elements. This additional control strengthens the researcher's ability to infer that an observed change in the dependent variable was cause by a change in the experimental, or treatment, variable. As a result, lab experiments are viewed as having a greater internal validity. On the other hand, the controlled and sterile environment of a lab may not be a good analog for a market place. Therefore, lab experiments are seen as having greater problems with external validity.

Factor Analysis

Purpose is data simplification. Objective is to summarize the info contained in a large number of metric measures (ex. rating scales0 with smaller summary measures, called 'factors' (defined as a linear combination of variables. It is a weighted variable summary score of a set of related variables, similar to the composite derived by averaging the measures).

Case Analysis

Purpose is to review info from a few other situations that are similar to the research problem. Researchers must always take care to determine the relevancy of any case study to the present problem.

Bulletin Board (Asynchronous Group)

Questions are posted for participants each day and they have all day to provide their responses. They usually run for 2 to 3 days and sometimes even longer. The moderator answers questions from participants throughout the day.

Third Person Technique

Rather than directly asking respondents what they think, researchers couch the questions in terms of "Your neighbor" or "Most people" or some other third party. Often used to avoid questions that may be embarrassing or evoke hostility if posed directly to a respondent.

Internal Validity

Refers to extent to which competing explanations for the experimental results observed can be ruled out. If the researcher can show that the experimental or treatment variable actually produced the differences observed in the dependent variable, then the experiment can be said to be internally valid. This kind of validity require evidence demonstrating that variation in the dependent variable was caused by exposure to the treatment variable and not other possible cause factors.

Instrument Variation

Refers to the changes in measurement instruments that might explain differences in the measurements taken. It is a serious problem in marketing experiments where many different people are used as interviewers or observers to measure the dependent variable. Causes variation.

Maturation

Refers to the changes in subjects during the course of the experiment that are a function of time. It includes getting older, hungrier, more tired, and the like.

External Validity

Refers to the extent to which the causal relationships measured in an experiment can be generalized to outside persons, settings, and times. The issue here is how representative the subjects and settings used int the experiment are of other populations and settings to which the researcher would like to project the results.

History

Refers to the intervention, between the beginning and end of the experiment, of any variable or event - other than those manipulated by the researcher - that might affect the value of the dependent variable.

Mortality

Refers to the loss of test units during the course of an experiment. It is a problem because there is no easy way to know whether the lost unit would have responded to the treatment variable in the same way as those units that remained throughout the entire experiment.

Regression to the Mean

Refers to the observed tendency of subjects with extreme behavior to move toward the average for that behavior during the course of an experiment. Test units may exhibit extreme behavior because of chance, or they may have been specifically chosen.

Coding

Refers to the process of grouping and assigning numeric codes to the various responses to a particular question. Close ended survey questions are pre-coded (numeric codes have been assigned to the various responses on the questionnaire). Open ended questions create a coding dilemma. The coding process: 1. List responses 2. Consolidate responses 3. Set codes 4. Enter codes - Read responses to individual open ended questions on the questionnaire. - Match individual responses with the consolidated list of response categories, and determine the appropriate numeric code for each response. - Write the numeric code in the appropriate place on the questionnaire or electronic database.

Data Entry

Refers to the process off converting info to a form that can be read by a computer.

Storytelling

Requires consumers to tell stories about their experiences. It is a search for subtle insights into consumer behavior.

Closed Ended Questions

Requires the respondent to make a selection from a list of responses. 2 types: dichotomous questions (with 2 item response options) and multiple choice questions.

Symptom

Researchers must distinguish between a real problem and a symptom. A symptom is a phenomenon that occurs because of the existence of something else. Ex. Managers often talk about poor sales, declining profits, increased customer complaints, or defecting customers. Each of these is a Symptom of a deeper problem.

Controlling the Panel

Researchers should see that panel participants have a positive experience with every research project. They should be given enough survey opportunities to stay effectively engaged, but not too much to be burdened with invitations. Provide respondent privacy, safeguard personal info, and protect panel members from bogus research. Panel providers are always recruiting new members to keep up with the growth in demand for online samples and to replace any drop outs. Needs to ensure panel freshness by updating member profiles.

Testing Effects

Result from the fact that the process of experimentation may produce its own effect on the responses observed. - Main testing effects: the possible effects of earlier observations on later observations. - Interactive testing effects: is the effect of a prior measurement on a subject's response to a later measurement.

Population Specification Error

Results from an incorrect definition of the population or universe from which the sample is to be selected.

Systematic Error or Bias

Results from mistakes or problems in the research design or from some flaws in the execution of the sample design. Sometimes called non-sampling errors because they can systematically influence survey answers and categorized as sample design error and measurement error.

Interviewer Error or Interviewer Bias

Results from the interviewer's influencing a respondent - consciously or unconsciously - to give untrue or inaccurate answers. Influences include: Dress, age, gender, body language, tone of voice, facial expression, etc.

Frame Error

Results from using an incomplete or inaccurate sampling frame. The problem is that the sample drawn from a list that is subject to frame error may not be a true cross section of the target population.

Point and Interval Estimates

Results of a sample can be used to generate two kinds of estimates of a population mean: point and interval estimates. The sample mean is the best 'point estimate' of the population mean. An interval estimate is a particular interval or range of values within which the true population value is estimated to fall.

Standard Normal Distribution

Same features as normal distribution. However, the mean is always equal to zero and the standard deviation is always equal to one. Transformation formula, based on proportional property of normal distribution, is used to transform any value X from any normal distribution to its equivalent value Z from a normal distribution: Z = (value of the variable - mean of the variable) / standard deviation of the variable

Budget Available

Sample size for a project is often determined, at least indirectly, by the budget available. Thus, it is frequently the last project factor determined.

Independent Samples

Samples in which measurement of the variable of interest in one sample has no effect on measure of the variable in the other sample.

Cluster Sampling

Sampling units are selected in groups. 2 basic steps in cluster sampling: 1. The population of interest is divided into mutually exclusive and exhaustive subsets. 2. A random sample of the subsets is selected. If the sample consists of all the elements in the selected subsets, it is called a one-stage cluster sample. However, if the sample of elements is chosen in some probabilistic manner from the selected subsets, the sample is a two stage cluster sample. Most popular type of cluster sample is the area sample in which the clusters are units of geography. Assumes that the elements in a cluster are as heterogeneous as those in the total population.

In-depth Interviews (IDI)

Second most popular form of qualitative research. Relatively unstructured one-on-one interviews. Interviewer is thoroughly trained in the skills of probing and eliciting detailed answers to each question. Sometimes psychologists are used as interviewers. They can employ non-directive clinical techniques to uncover hidden motives.

Nature of Secondary Data

Secondary data consist of info that has already been gathered and might be relevant to the problem at hand. Can be cost effective and an efficient means of obtaining info.

Focus Group Facility

Setting is often a conference room, with a large one-way mirror built into one wall. Microphones are placed in non-obtrusive locations (ceiling) to record discussion. Behind mirror is a viewing room and recording/ videotaping equipment.

Develop Operational Procedures for Selecting Sample Elements

Should be developed and specified, whether a probability or non probability sample is being used. However, procedures are much more critical to the successful execution of a probability sample, in which case, they should be detailed, clear, and unambiguous and should eliminate any interviewer discretion.

Personification

Similar to analogies, involves drawing a comparison between a product and a person.

Itemized Rating Scale

Similar to graphic rating scales except that respondents must select from a limited number of ordered categories rather than placing a mark on a continuous scale.

Purchase Intent Scale

Single most often used scale in marketing research. Used to measure a respondent's intent to buy or not buy a product.

Using the Internet for Secondary Data

Sites of Interest to Marketing Researchers - Websites - Periodicals - Newspaper Databases - Book Databases

Understand the Decision-Making Environment

Situation Analysis: The better the marketing researcher understands the decision-making environment, the more likely it is that the problem will be defined correctly. Exploratory Research: If informed discussions with managers/suppliers and onsite visits are not enough, Exploratory Research may be conducted to obtain a greater understanding of a concept or to help crystallize the definition of a problem. Also used to identify important variables to be studied. It is the preliminary research, not the definitive research used to determine course of action.

Ad Hoc Mail Surveys

Sometimes called One Shot Mail Surveys, the researcher selects a sample of names and addresses from an appropriate source and mails them questionnaires.

Measurement Instrument Bias

Sometimes called questionnaire bias, results from problems with the measurement instrument or questionnaire.

Internet Search Strategies

Step 1: Analyze your topic to decide where to begin. Step 2: Test run a word or phrase in the search engine. Consider synonyms or equivalent terms. Step 3: Learn as you go and vary your approach with what you learn. Don't assume that you know what you want to find. Look at search results and see what else you might use in addition to what you thought of. Step 4: Don't get bogged down in any strategy that doesn't work. Consider using a subject dictionary. Step 5: If you haven't found what you want, go back to earlier steps better informed.

Hypothesis Testing

Steps in hypothesis testing: 1. Stating the hypothesis in 2 basic forms: the null hypothesis (Ho) and the alternative hypothesis (Ha). The null is being tested against the alternative. Both must be stated in a way that both cannot be true. 2. Choosing the appropriate test statistic. 3. Developing a decision rule. Using a significance level (Ex: .10, .05, .01) is the probability that is considered too low to justify acceptance of the null hypothesis. 4. Calculating the value of the test statistic. In this step, the researcher does the following: - Use appropriate formula. - Compares value just calculated to the critical value of the statistic, based on decision rule chosen. - Based on comparison, determines to either reject or fail to reject the null hypothesis. 5. Stating the conclusion.

Strengths and Weaknesses of Selected Data Collection Methods in Terms of Quality of Data Produced (Ex. 5.6)

Strength = (S) and Weakness = (W) - Mall-Intercept Interview: (S) Interviewer can show, explain, and probe. (W) Many distractions in mall environment, respondents may be in a hurry or not in a proper frame of mind. Chance for interviewer bias. Non-probability sampling problems arise. - Call Center Telephone Interview: (S) Supervisor can monitor the interviewing process. Excellent samples can be obtained. Interviewer can explain and probe. (W) Respondent may be distracted by things in their location. Many refuse to participate. - Self-Administered Questionnaires: (S) Interviewer and Bias are eliminated. Can be completed when convenient. Info can be looked up and worked on at respondent's own pace. (W) No interviewer to show or explain. Who actually completes the questionnaire cannot be controlled. - Mail Survey: (S) and (W) Same as self administered questionnaires.

Primary Data

Surveys, observation, and experiments collected to solve the particular problem under investigation.

Sample Design Error

System error that results from a problem in the sample design or sampling procedure.

Execute the Plan

Test usually runs for 6 to 12 months, sometimes shorter or longer depending on the product cycle. Must run long enough for an adequate number of repeat purchase cycles to be observed in order to provide a measure of the 'staying power'.

Traditional Statistical Methods

The 3 pieces of info required to make the necessary calculations for a sample result: - An estimate of the population standard deviation. - The acceptable level of sampling error. - The desired level of confidence that the sample result will fall within a certain range (result +/- sampling error) of true population values.

Define the Population of Interest

The basic issue in developing a sampling plan to specify the characteristics of those individuals or things from whom or about whom info is needed to meet the research objectives. Is often specified in terms of geographic area, demographic characteristics, product or service usage, and or awareness measures. Screening questions used to determine if a particular individual is of interest. Also includes reasons to exclude certain individuals.

2 Basic Sources of Secondary Data

The company itself (internal databases). Other organizations or persons (external databases)

Concomitant Variation

The degree to which a presumed cause and a presumed effect occur or vary together. It is also a criterion that must be met for causality. Ex. Direct mail promotion and coffee sales.

Battle over Privacy

The downside to behavioral targeting: invasion of consumer privacy. Researchers say that their data contain no personal identifiable info.

True Experimental Designs

The experimenter randomly assigns treatments to randomly selected test units.

Predictive Function

The final role of research. Specification of how to use descriptive and diagnostic research to predict the results of a planned marketing decision. Q's: How can the firm best take advantage of opportunities as they arise in the ever-changing marketplace?

Execute the Operational Sampling Plan

The final step in the sampling process. Requires adequate checking to ensure that specified procedures are followed.

Determining Sample Size (Problems Involving Means)

The formula for calculating the required sample size for problems that involve the estimation of a mean: n = (Z^2 P^2) / E^2 Z = level of confidence expressed in standard errors P = population standard deviation E = acceptable amount of sampling error

Proportional Allocation

The number of elements selected from a stratum is directly proportional to the size of the stratum in relation to the population. Formula: n/N n = size of the stratum N = size of the population

Confidence Level

The probability of that the interval will include the true value of the population mean. That interval is called the 'confidence interval'.

Causal Studies

The researcher investigates whether the value of one variable causes or determines the value of another variable, in an attempt to establish linkage between them. Experiments are often used to measure causality.

Quasi-Experiments

The researcher lacks complete control over the scheduling of treatments or must assign respondents to treatments in a non-random fashion.

Diagnostic Fuction

The second role of research wherein data and or action are explained. Q's: How can product/ service offerings be altered to better serve customers and potential customers?

Selection Bias

The threat to validity posed by selection bias is encountered in situations where the experimental or test group is systematically different from the population to which the researcher would like to project the experimental results or from the control group.

Snowball Samples

These sampling procedures are used to select additional respondents on the basis of referrals from initial respondents. Is used to sample from low incidence or rare populations. Advantage: dramatic reduction in search costs. Disadvantage: Low quality sample and likely to be biased.

Extraneous Variables

Things that stand in the way of the researcher when trying to reach a conclusion. The researcher needs to design the experiment so as to eliminate as many extraneous variables as possible.

Determination of Survey Method

Things to consider: Sampling precision, budget, requirements for respondent reactions, quality of data, length of questionnaire, incident rate, structure of questionnaire, and time available to complete survey.

Open Ended Questions

Those to which the respondent replies in his/her own words.

Marketing Concept

To accomplish goals, firms adopt the Marketing Concept. Requires: 1. A consumer orientation 2. A goal orientation 3. A systems orientation

Television Audience Measurement and the Portable People Meter

Traditional 'people meters' are electronic TV set up boxes (STB) that continually record which channel is tuned in. Today, trend is straying away from traditional people meters and more towards direct measurement by STB. Researchers able to tap into second by second viewing habits of millions of households. People meter panel consist of only 18,000 people.

Types of Errors in Hypothesis Testing

Type I Error: Involves rejecting the null when it is, in fact, true. The researcher may have reached this incorrect conclusion because of sampling errors. Probability of committing a type I error is referred to as the alpha level. 1 - alpha is the probability of making a correct decision by not rejecting the null when it is true. Type II Error: Involves failing to reject the null when it is actually false. Referred to as a beta error. 1 - beta reflects the probability of making a correct decision in rejecting the null when it is false.

Split Half Technique

Typically calls for scale items to be randomly assigned to one half or the other. Problem with this method is that the estimates of the coefficient of reliability is totally dependent on how the items were split. To overcome this problem, researchers use the Cronbach alpha technique, which involves computing mean reliability coefficient estimates for all possible ways of splitting the set of items in half.

Cartoon Tests

Typically consists of 2 characters with balloons for dialogue. One is filled and the other is blank. Respondents are asked to fill in the blank balloon. Extremely versatile and highly projective.

Quota Samples

Typically selected in such a way that demographic characteristics or interest to the researcher are represented in the sample in target proportions. Thus many people confuse quota samples with stratified samples.

Eye Tracking

Used to help improve website designs. Tracks where eyes go to first. Lessons learned from eye tracking research: - Ads at the top and left portions of a homepage receive the most eye fixations. Ads in the right column are treated as an afterthought area. Ads at the bottom of the page are typically only seen by a small percentage of people. - Close proximity to popular editorial content helps ads get noticed. - All types of ads tested, text ads are viewed more intently. - Size matters. Bigger ads have a better chance of being seen. - Larger images hold the eye longer than small images. - Clean, clear faces in images attract more eye fixations on homepages. - People are more likely to recall facts, names, and places correctly when they were presented with the info in a text format. - Shorter paragraphs perform better than longer ones. Receives twice as much eye fixation.

Selective Research

Used to test decision alternatives. Ex. Testing concepts for new products, advertising copy testing, and test marketing.

Constant Sum Scales

Used when researchers want to avoid long lists of paired items. Requires respondent to divide a given number of points, typically 100, among 2 or more attributes based on their importance to him/her.

Nominal

Uses numerals to ID objects, individuals, events, or groups. Operations: Determination of equality or inequality. Typical Descriptive and Usage: Classification Statistics: Frequency counts, percentage/ modes

Participants

Usually recruited from a variety of sources. Two traditional procedures include mall-interception interviewing and random telephone screening.

Experimental Validity

Validity is defined as the degree to which an experiment actually measures what the researcher was trying to measure. Two specific kinds: Internal and external.

Face Validity

Weakest form of validity. It is concerned with the degree to which measurement seems to measure what it is supposed to measure. It is a judgement called by the researcher, made as the questions are designed. Thus, each question is scrutinized. Can refer to the subjective agreement of researchers, experts, people familiar with the market that a scale logically appears to be accurately reflecting what is supposed to be measured.

Accepting Ho versus Failing to Reject Ho

When a hypothesis is tested, Ho is assumed to be true until it is demonstrated to be likely false. Either the evidence supports Ha (reject Ho) or supports Ho (fail to reject Ho).

Physiological Measurement Devices

When an individual is aroused or feels inner tension or alertness, his/ her condition is referred to as 'activation'. Activation is stimulated via a subcortial unit called the reticular activation system (RAS) located in the human brain stem. The sight of a product or advertisement can activate RAS. Researchers have used a number of devices to measure the level of a person's activation.