IMC 302 Pt. 2
DISGUISE
(Key Issues for Collecting Information by Communication: DISGUISE) The amount of knowledge people have about a study in which they are participating. An undisguised questionnaire makes the purpose of the research obvious, either in the introduction, the instructions, or the questions themselves. Disguise is especially useful when... ...knowing the purpose or sponsor is likely to bias respondents' answers. E.g., Imagine asking people to "name the first three brands of cars that come to mind" or to "indicate your top two choices among the following brands of cars," when they can clearly see that Ford sponsors the survey. The researcher has "primed" a particular response, and the results will be of little value
Primary Methods of Administration
(Key Issues for Collecting Information by Communication: Primary Methods of Administration) Personal Interviews - highly flexible, can iterate based on conversation; trust developed Telephone Interviews - highly flexible, can iterate based on conversation; trust somewhat developed (Problems Caller ID, screening calls Unpublished mobile numbers Short attention span of consumers Benefits Because of the human contact, a degree of trust can be developed, particularly on longer surveys allow probing and follow-up on respondent answers when necessary) Mail Surveys (least common) - can't iterate; lower trust; cheap(A survey administered by mail to designated respondents with an accompanying cover letter. The respondents return the questionnaire by mail to the research organization.) Online Surveys (most common) - can iterate a little (branching questions); lower trust; cheap
Low structure: OPEN-ENDED QUESTIONS
(Key Issues for Collecting Information by Communication: STRUCTURE) A question in which respondents are free to reply in their own words rather than being limited to choosing from among a set of alternatives. (There is a standardized question that everyone receives, but people get to answer in any way that they choose) ex) Overall, how do you feel about Target, the department store chain? Please type your answer in the box below.
High structure: FIXED-ALTERNATIVE QUESTIONS
(Key Issues for Collecting Information by Communication: STRUCTURE) Questions in which the responses are limited to stated alternatives. Considering all aspects, what is your overall evaluation of Target, the department store chain? - Extremely unfavorable - Unfavorable - Neither favorable nor unfavorable - Favorable - Extremely favorable
Developing the data collection form Step 7: Determine Question Sequence
*Use simple and interesting opening questions *Use the funnel approach - Start with broad questions (global measure) and progressively narrow the scope (composite measures) - It's best not to make sudden changes in topics or to jump around from topic to topic *Design branching questions with care - Branching questions direct respondents to different places in a questionnaire, based on their responses to the question at hand - Main issue is programming in Qualtrics * Place difficult or sensitive questions late in the questionnaire - Once respondents have become involved in the study, they are less likely to react negatively or refuse to answer when delicate questions are posed TARGET INFORMATION: The information that addresses the subject of the study (e.g., intentions or attitudes of respondents toward a new product) CLASSIFICATION INFORMATION: Information used to classify respondents, typically for demographic breakdowns (e.g., gender)
Communication vs. Observation
- Communication: Versatility, Speed, and Cost. - Observation: Objectivity and Accuracy.
Developing the data collection form Step 8: Determine Appearance of Questionnaire
- No clutter! (Keep it professional, watch for typos, Use Qualtrics templates (plain white background, NU background)) - Optimize for mobile - Keep it as short as possible -> avoids fatigue - Number questions within sections For example, 1-1 (brand consumption), 2-1 (brand attitude 1), 2-2 (brand attitude 2), 3-1, 3-2, etc. - Include an organization name (NU) - Go easy on instructions, unless they are absolutely necessary
Three Purposes of Descriptive Research
1) DEMOGRAPHIC PROFILES: Describe the characteristics of certain groups - e.g., A research group gathered information from individuals who had eaten at a particular barbecue restaurant chain in a Midwestern US city to help managers develop a profile of the "average user" with respect to income, sex, age, and so on 2) RECORD BEHAVIOR: Estimate the proportion of people who behave in a certain way - e.g., Estimate the proportion of people within a specified radius of a proposed shopping complex who currently shop or intend to shop at the center 3) INVESTIGATE WHY: Ask why something is happening --> PRESCRIPTIVE ANALYSIS: Make specific predictions/forecast - E.g., predict the level of sales for a certain brand for each of the next five years so that we could plan for the hiring and training of new sales representatives
Six types of error
1) Sampling Error SAMPLING ERROR - The difference between results obtained from a sample and results that would have been obtained had information been gathered from or about every member of the population. - Decreased by increasing sample size - Can be estimated (assuming probability sample) - Usually less troublesome than other kinds of error 2) Noncoverage Error 3) Nonresponse Error NONRESPONSE ERROR - Error from failing to obtain information from some elements of the population that were selected and designated for the sample. - This is a potential problem that only occurs when those who respond are systematically different in some important way from those who don't respond 4) Response Error RESPONSE ERROR - Error that occurs when an individual provides an inaccurate response, consciously or subconsciously, to a survey item. 5) Recording Error 6) Office Error
The online survey
1) keep as short as possible 2) optimize the survey for use on mobile phones 3) begin with engaging question 4) keep questions simple 5) use visuals/graphics if they help, but must still be easy to navigate 6) remind respondents about incentives, and explain how to obtain them 7) for long surveys, let repondents see progression through it 8) pretest the online survey
Steps in developing a sampling plan
1. Define the target population 2. Identify the sampling frame 3. Select a sampling method procedure 4. Determine the sample size 5. Select the sample elements 6. Collect the data from the designated elements
Three Types of Primary Data Research
1. Exploratory Research 2. Descriptive Research 3. Causal Research Sometimes managers will get results from descriptive or causal projects that lead to more questions. Then these questions, in turn, lead to more exploratory research. *** Can we establish causality? We can never really prove that one thing causes another - 3rd variable? However, In order to say that one thing "caused" another, three conditions should be met: (1) There must be consistent variation between the cause and the effect, (2) the time order of the cause and the effect must be correct, and (3) other explanations must be eliminated
Developing the data collection form
1. Specify what information will be sought 2. Determine Method of Administration 3. Determine Content of individual questions 4. Determine form of response to each question 5. Determine wording of each questions 6. Prepare Dummy tables 7. Determine question sequence 8. Determine appearance of questionnaire 9. Develope recruiting message or script 10. Reexamine steps 1-9, pretest questionnaire, and revise if necessary
Key Issues for Collecting Information by Communication
3 criteria to distinguish different types of communication research Degree of Structure Degree of Disguise Method of Administration
Learning Objective 1: Explain the difference between a parameter and a statistic.
A PARAMETER is a characteristic of the population; if it were possible to take measures from all population members without error, we could arrive at the true value of a parameter. A STATISTIC is a characteristic or measure of a sample; statistics are used to estimate population parameters.
Constant-sum Method
A comparative-ratings scale in which an individual divides some given sum among two or more attributes on a basis such as importance or favorability. - If a respondent gave 25 points to each of the attributes, they would be judged to be equally important. If he assigned 50 points to courteous service, 25 points to convenient location, 15 points to convenient hours, and 10 points to low-interest-rate loans, it's clear that courteous service is most important, low-interest-rate loans are least important, with the other attributes falling in between
Two Methods of Data Collection: OBSERVATION
A method of data collection in which the situation of interest is watched and the relevant facts, actions, or behaviors are recorded. e.g. secret shopper; supermarket scanners may be used to determine how many boxes of a particular brand of cereal are sold in a given region in a typical week
FILTER QUESTION
A question used to determine if a respondent is likely to possess the knowledge being sought; also used to determine if an individual qualifies as a member of the defined population. - "Do you do the grocery shopping for your family?" - "Have you eaten at Mickey's restaurant, located at Sixth Street and Manvel Avenue, within the past six months?" - "Did you vote in the last presidential election?"
ITEMIZED-RATINGS SCALES
A scale on which individuals must indicate their ratings of an attribute or object by selecting the response category that best describes their position on the attribute or object. Interval EXAMPLES: Likert Summated-Ratings Scales Semantic Differential Scales
COMPARATIVE-RATINGS SCALES
A scale requiring subjects to make their ratings as a series of relative judgments or comparisons rather than as independent assessments.
SUMMATED-RATINGS SCALE
A self-report technique for attitude measurement in which respondents indicate their degree of agreement or disagreement with each of a number of statements.
RECALL LOSS
A type of error caused by a respondent's forgetting that an event happened at all. the best time frame seems to be between two weeks and one month
TELESCOPING ERROR
A type of error resulting from the fact that most people remember an event as having occurred more recently than it did. Suppose that you needed to determine how much Diet Coke had been purchased among a sample of consumers over the prior week. The estimates provided by your respondents will almost always be too high as they recall purchases made prior to the focal seven-day period
A measure can be reliable but not valid.
A valid measure must be reliable.
Handling Sensitive Questions
A) Don't ask unless absolutely necessary! B)Guarantee anonymity C) Place sensitive questions near end - This will allow the researcher and the respondent a little time to develop trust and rapport, especially with personal interviews and telephone interviews. - There's another practical advantage too: If the respondent decides to stop answering questions at that point, at least he or she has already completed most of the questionnaire. D) Include a counter biasing statement Counterbiasing Statement Recent studies show that one of every four households has trouble meeting its monthly financial obligations. Is your household currently experiencing financial difficulties?
High Structure: Advantages and Disadvantages
ADVANTAGES 1) ease of administration, 2) ease of coding and analysis, 3) measure reliability (Avoids possible interpretation bias Provides an identical frame of reference- every day vs. 3 hrs a week More reliable because it includes a fixed set of options) DISADVANTAGES Response Bias: - Force Choice - Omitted category of response - Precision of response (it doesn't capture all the rich data a person has to offer) (Doesn't capture "fanatic" attitudes on the scale, only extremely favorable which may mean different things to different people fixed alternatives may force an answer to a question on which the respondent has no opinion Doesn't capture exactly what they want to say)
RELIABILITY
Ability of a measure to obtain similar scores for the same object, trait, or construct across time, across different evaluators, or across the items forming the measure. A reliable measure isn't heavily influenced by transitory factors that cause random errors
Learning Objective 2: List some demographic and socioeconomic characteristics that interest marketers.
Age, education, occupation, marital status, sex, ethnicity, income, social class.
Learning Objective 3: Explain the concept of validity as it relates to survey items.
Any scale or other measurement instrument that actually measures what it was intended to measure is said to have validity. As systematic and/or random error increases, the validity of a measure decreases.
Communication Research ( not on study guide)
Asking people questions helps understand what they do and why they do it, which can help predict future behavior Different types of communication data - Filling out a carefully designed questionnaire - Applying for a loan; - Completing an application for a grocery store discount card
Learning Objective 4: Discuss the concept of total sampling elements (TSE).
Because it is rare that all of the people who have been selected to participate in a study will do so, it is usually necessary to draw a larger pool of sample elements from the sampling frame than is actually needed in the study. The larger set drawn from the sampling frame is referred to as total sampling elements.
Primary data
Can be descriptive or causal
Two Methods of Data Collection: COMMUNICATION
Communication : A method of data collection involving questioning respondents to secure the desired information using a data collection instrument called a questionnaire.
Key Issues for Collecting Information by Observation: DISGUISE
DISGUISE The amount of knowledge people have about a study in which they are participating. Disguised vs. Undisguised Observation - With undisguised observation, subjects are aware that they are being observed. - In the soup example, observers could stand next to the soup shelves in the grocery store in plain sight of customers, pencil and clipboard in hand, announcing their purpose to each customer - Results in more "smart" shopping behavior because participants know they are observed - With disguised observation, subjects are not aware that they are being observed. - E.g., have observers become part of the shopping scene
Developing the data collection form Step 6: Prepare Dummy Tables
DUMMY TABLE A table (or figure) used to show how the results of an analysis will be presented. Preparing a complete set of dummy tables forces you to think carefully about each piece of information to be collected It is better that the exact variables and categories to be investigated, as well as the necessary statistical tests, are specified before you begin to collect the data -> doesn't always happen, I can help you figure this out
Learning Objective 1: Cite three major purposes of descriptive research.
Descriptive research is used when the purpose is to (1) describe the characteristics of certain groups, (2) determine the proportion of people who behave in a certain way, and (3) make specific predictions.
RANDOM ERROR
Error in measurement due to temporary aspects of the person or measurement situation that affects the measurement in irregular ways. Random error is present when we repeat a measurement on an individual and don't get the same scores as the first time we did the measurement, even though the characteristic being measured hasn't changed. Mood, state of health, fatigue, temperature of the room - When you only have random error, if you measure the same thing multiple times, your measurements will tend to cluster or vary around the true value. Some values will be higher than the true score, while others will be lower. When you average out these measurements, you'll get very close to the true score. - For this reason, random error isn't considered a big problem when you're collecting data from a large sample—the errors in different directions will cancel each other out when you calculate descriptive statistics. But it could affect the precision of your dataset when you have a small sample. - Systematic errors are much more problematic than random errors because they can skew your data to lead you to false conclusions. If you have systematic error, your measurements will be biased away from the true values. Ultimately, you might make a false positive or a false negative conclusion (a Type I or II error) about the relationship between the variables you're studying
SYSTEMATIC ERROR
Error in measurement that is also known as constant error because it affects the measurement in a constant way. Can be consistent across people (measurement tool issue) or within people (personality bias) MEASUREMENT TOOL The questions on your survey are designed in a way that encourages people to answer yes PERSONALITY: maybe you are more willing to express negative feelings than most other people; some people seem to be systematically negative in all their responses. Sometimes personality traits or other stable characteristics of individuals add systematic error to the measurement process. BEHAVIOR: those who perform behaviors frequently tend to underreport the level of behavior, and those who perform those behaviors less frequently tend to over report the level of behavior.
HYPOTHESIS
HYPOTHESIS A statement that specifies how two or more measurable variables are related. (H1): Women are more likely than men to make impulse purchases of our brand. (H2): Decreasing price by 10% will increase unit sales by 30%. (H3): Adoption of our new product will be greater in Northern states than in Southern States.
Key Issues for Collecting Information by Observation: METHOD OF ADMINISTRATION
How is the observation data collected? HUMAN VS. MECHANICAL OBSERVATION - With human observation, individuals are trained to systematically observe a phenomenon and to record on the observational form the specific events that take place. - Researchers commonly use written field notes to record their impressions at the time they are observed in the field and later write up summary thoughts back at the office - With electrical or mechanical observation, an electrical or mechanical device observes a phenomenon and records the events that take place. Examples of Mechanical Observation - Video cameras (How long they spend at each display) - Eye camera (an eye camera records where a subject's eye is focused: Where did the individual look first? How long did the person linger on any particular place?) - Bar code scanners (Tie purchases to households) - People meters(Who is watching what) Other : response latency = the amount of time it takes an individual to respond to a stimulus: the amount of time it takes to answer a question is one measure of a respondent's strength of feeling (or the reverse, degree of uncertainty) about her answer Galvanometer: When a person experiences emotions, slight changes occur in the skin's electrical resistance that aren't under the individual's control. The galvanometer records these changes, which are used to infer the subject's interest or attitude toward the stimulus.
CONVENIENCE SAMPLE: sample design
IN Developing a sampling plan STEP 3: Select a Sampling Procedure 1) CONVENIENCE SAMPLE A nonprobability sample in which population elements are included in the sample because they were readily available. - Sometimes referred to as "accidental" sampling; population elements are sampled simply because they are in the right place at the right time. - Easy to conduct, but no way to know if sample is representative of the population (i.e., cannot statistically assess sampling error). - Commonly used with exploratory research, when the goal is to generate insights or to develop hypothese 2) JUDGMENT SAMPLE - A nonprobability sample in which the sample elements are handpicked because they are expected to serve the research purpose. - E.g., choosing only upper class consumers to study luxury purchasing behavior - The researcher may believe that the sample elements are representative of the larger population or that they can offer the information needed, but he's still picking them himself - A judgment sample becomes dangerous when it is used in descriptive or causal studies and its weaknesses are ignored -> missing representative elements 3) QUOTA SAMPLE A nonprobability sample chosen so that the proportion of sample elements with certain characteristics is about the same as the proportion of the elements with the characteristics in the target population. - A "quota" representing these characteristics is established (e.g., 25 males between the ages of 20 and 29; 25 females between the ages of 20 and 29; 35 males between the ages of 30 and 39; etc.) so that when the sample is complete it will mirror the population on the key characteristics - the specific sample elements to be used in a quota sample are left to the discretion of the researcher. That's what makes a quota sample a nonprobability sampling plan
Internal Validity vs. External Validity
INTERNAL VALIDITY - The degree to which an outcome can be attributed to an experimental variable and not to other factors. Lab experiments tend to have higher levels of internal validity. EXTERNAL VALIDITY - The degree to which the results of an experiment can be generalized, or extended, to other situations. Field experiments tend to have higher levels of external validity. - Having greater external validity means giving up a degree of internal validity and vice versa. - For example, to gain greater internal validity, we must hold more elements in the situation constant across conditions, and this makes the experiment less realistic, thereby lowering external validity. - Similarly, if we want more external validity, the more realistic environment that is required serves to decrease internal validity. - Both are important; conduct both lab and field experiments
Learning Objective 2: Know the difference between a nonprobability and probability sample
In NONprobability, you pick the people In probability, there is a random nonzero chance of each member of the population being included
Developing the data collection form Step 9: Develop Recruiting Message or Script
Include a cover letter/ consent form: The usual things to include: Who you are Why you are contacting them The request for help Promise of anonymity or confidentiality How long it will take Any incentives
Other Considerations in Designing Scales
Including a "don't know" or "not applicable" response category - Pro: good idea if a fairly sizable percentage of respondents are likely not to have encountered or thought about the object or issue being addressed in the study. - Con: some respondents will choose it because it's the easiest way to get through the survey Number of items in a scale Global measures = 1 item, overall impression Composite measures = designed to provide a comprehensive assessment of an object or phenomenon with items to assess all relevant aspects or dimensions Assess both: global first, composite second Number of scale positions Generally 5-11 More allows for further refinement but too many can mean each point is hard to interpret Odd or even number? Odd allows for central position (neutral) - do you want a neutral position included? What does neutral mean?
Developing the data collection form Step 3: Determine Content of Individual Questions
Is the question necessary? - Do you know what you are going to do with it? What analysis will you do? This may be hard to know before we do the analyses lectures but I can help you figure this out when you design your questionnaire Are several questions needed instead of one? - "Why do you use Crest toothpaste" may be answered in terms of why you use it now or how you started using it - "How did you first happen to use Crest?" & "What is your primary reason for using it?" KEY ISSUES: - Do respondents have the necessary information? They will answer anyway, so... (1) the respondent needs to know something about the issue addressed by the question and To address, ask a filter question - Will respondents give the information? If no, may be a function of the amount of work involved in producing an answer or - If no, may be the sensitivity of the issue How did you vote in 2016 election? - ***FILTER QUESTION, TELESCOPING ERROR, RECALL LOSS, handling sensitive questions, counterbiasing statement,
Learning Objective 1: Clarify the difference between laboratory experiments and field experiments.
Laboratory experiments differ from field experiments primarily in terms of environment. - The researcher creates a setting for a laboratory experiment; a field experiment is conducted in a natural setting. Both types, however, involve control and manipulation of one or more presumed causal factors.
Key Issues for Collecting Information by Observation: MEASUREMENT
Measurement is defined as "rules for assigning numbers to objects in such a way as to represent quantities of attributes" NOMINAL SCALE Measurement in which numbers are assigned to objects or classes of objects solely for the purpose of identification.(Football jerseys, lockers, social security numbers: a basketball player wearing uniform number 15 isn't necessarily taller or a better shooter than a player wearing the number 14) * simply identify the individual or object that has been assigned the number if we assign the number 1 to represent female respondents to a survey and the number 2 to represent male respondents, we've used a nominal scale that allows us to identify the sex of a particular respondent and to determine the relative proportions of each sex in our sample all you can do is count the number of people that fall into the various categories Mode matters ORDINAL SCALE Measurement in which numbers are assigned to data on the basis of some order (for example, more than, greater than) of the objects. - With an ordinal scale, the numbers have more meaning in that they represent this relative standing - we know that one option has more of some attribute than some other option, but we don't know how much more: can't say how much someone likes coke more than mountain dew - Median matters INTERVAL SCALE Measurement in which the assigned numbers legitimately allow the comparison of the size of the differences among and between members. - The difference between 1 and 2 is equal to the difference between 2 and 3 - We can't compare the absolute magnitude of numbers - what is a 3 except in relation to 2 and 4? - there is no obvious point at which attitude is equal to zero - By definition, interval scales represent "relative" values or differences. - It's still all relative - Mean matters RATIO SCALE Measurement that has a natural, or absolute, zero and therefore allows the comparison of absolute magnitudes of the numbers. - Height, weight - Because there is an absolute zero, comparison of the absolute magnitude of the numbers is legitimate - using the ratio scale will allow us to compute a mean age across a sample of respondents; compute correlations between age and ot
Learning Objective 1: Define the term measurement as it is used in marketing research.
Measurement is the process of assigning numbers to represent properties of an object's attributes.
Learning Objective 2: Cite the main reason researchers may choose to disguise the presence of an observer in a study.
Most often, an observer's presence is disguised in order to control the tendency of people to behave differently when they know their actions are being watched.
Types of primary data used in descriptive studies
Motivation Demographic/Socioeconomic Personality/Lifestyle Attitudes Awareness/Knowledge Intentions Motivation all lead towards BEHAVIOR
Simple sample size calculator
N = z2 x s x (1-s) = 384 e N = required sample size S = standard deviation of population (assume .5 if you don't know; look at previous literature to figure out) E = margin of error (assume 5% or .05) Z = z-score associated with 95% confidence interval (=1.96)
Establishing the Validity and Reliability of Measures
Nothing in marketing research can be measured without error
Why Use Observation Research?
Observation is often the best method for generating valid data about individuals' behavior because of heightened objectivity and accuracy - E.g., most people wouldn't want to acknowledge that they spend more on cat food than on baby food, but consumers can be observed doing so It is often better to measure consumers' actual behavior and then work backward to determine what led to that behavior than it is to measure their attitudes and intentions and then hope that their future behavior matches their intentions
Developing the data collection form Step 4: Determine Form of Response to Each Question
Open-Ended Questions (low structure) versus CLOSE ENDED Questions (high structure) - the response categories must be exhaustive; all reasonable responses must be included. - In addition, response categories must be mutually-exclusive, except in special cases where more than one answer is acceptable (e.g., check all that apply) - e.g., Age: 20-29, 30-39 rather than Age: 20-30, 30-40 Factual - Seek direct answers from the respondent; each question has a correct answer E.g., Can you name 3 sponsors of the Monday night football games? 3 columns; frequency count Exploratory - Uncover motivations and rich descriptions of feelings and attitudes OTHER CONSIDERATIONS: RESPONSE ORDER BIAS An error that occurs when the response to a question is influenced by the order in which the items or alternatives are presented. In general, response options presented earlier in a list of options are often more likely to be selected than options appearing later in the list. Solution: randomize presentation of items/alternatives
developing a sampling plan STEP 1: Define the Target Population
POPULATION All cases (usually people but can be brands) that meet designated specifications for membership in the group (Researchers must be very clear and precise in defining the population.) EX) Households in the city limits of Sacramento, California, with one or more children under the age of 18 living at home. CENSUS Collect them all: A type of sampling plan in which data are collected from or about each member of a population. VS SAMPLE Selection of a subset of elements (e.g., people) from a larger group of objects (e.g., people). - Goal is to make inferences about the overall population PARAMETER A characteristic or measure of a population. e.g., average age, proportion of population with a college degree, attitude toward a new service offering When we work with a sample drawn from a population, we are attempting to describe the population parameters based on the measures we take from the sample members. ***STATISTIC A characteristic or measure of a sample. ***We calculate statistics from sample data in order to estimate population parameters
Developing a sampling plan STEP 4: Determine the Sample Size
PRECISION The degree of error in an estimate of a population parameter. CONFIDENCE The degree to which one can feel confident that an estimate approximates the true value. Precision and confidence are inversely related; as one increases, the other decreases, all else equal.
Developing the data collection form Step 10: Reexamine Steps 1-9, Pretest
PRETEST Use of a questionnaire (or observation form) on a trial basis in a small pilot study to determine how well the questionnaire (or observation form) works. The pretest is the most inexpensive insurance you can buy to ensure the success of the project Purpose of pretest: see the distribution of responses to questions, read comments on the survey, and other useful info to know if people actually understood your survey and could complete it
Learning Objective 3: Explain the role of pretesting in the questionnaire development process.
Questionnaire pretesting is the final step in the survey development process. It is the last chance that the researcher has to ensure that the data collection form is working properly prior to data collection; pretesting must not be overlooked.
Tips for increasing response rates on online surveys
Recruiting messages on email: 1) use professional "from" name 2) keep subject line simple, but interesting 3) avoid spam-like language 4) personalize by using the person's name 5) include short, effective message about who you are, request for help, length/time, incentives, confidentiality, purpose of study 6) if offering incentives, make them meaningful 7) consider timing of email 8) Send reminder emails - but no more than 2 9) Pretest the recruiting message
DESCRIPTIVE RESEARCH
Research in which the major emphasis is on describing characteristics of a group (e.g., brand attitude, preferences, demographic variables) or the extent to which variables are related (e.g., is brand attitude different for men versus women?). Descriptive research should be conducted only when researchers know what the key issues are and what questions need to be asked
Key Issues for Collecting Information by Communication: STRUCTURE
STRUCTURE Degree of standardization used with the data collection instrument. With high structure, everyone asked to answer the question will get the identical question and respond within a set of possible answers
Key Issues for Collecting Information by Observation: STRUCTURE
STRUCTURE The degree of standardization used with the data collection instrument. You must decide precisely which behaviors will be observed and which specific categories and units will be used to record the observations. To make these decisions, you'll need to have specific hypotheses in mind Structured Observation (Method of observation in which the phenomena to be observed (typically behaviors) can be defined precisely along with the categories used to record the phenomena.) ex) first soup to be picked up in observation: Campbell's. Time in front of shells: 12 seconds Unstructured Observation "purchaser first stopped in front of campbells brand. He glanced,... Structured vs unstructured observation - On the plus side, structuring the observation reduces the potential for bias in recording and increases the reliability of observations. - Another advantage is that the coding and analysis of the data is far simpler (and less costly) when there are a few well-defined behaviors or observable demographic variables to be considered. - However, by structuring the observation, you have necessarily given up additional information that might have been coded, some of which might be important for understanding purchasing behavior
Developing a sampling plan STEP 3: Select a Sampling Procedure
Sample designs: 1) NONPROBABILITY SAMPLE - A sample that relies on personal judgment in the element selection process. - With nonprobability samples, sampling error cannot be estimated and we cannot calculate the margin of sampling error. Can't extrapolate to population - Convenience - Judgment (e.g., snowball) - Quota - see nonprob sample slide for more info 2) PROBABILITY SAMPLE*** - A sample in which each target population element has a known, nonzero chance of being included in the sample. - With probability samples there is a random component to which elements are selected not subject to whims of researcher WHY USE PROBABILITY SAMPLING? ...because the analyst can statistically assess the level of sampling error and make projections to the population. - (just don't forget that sampling error is only one kind of error... and it usually isn't the biggest problem)
EXPERIMENT
Scientific investigation in which an investigator manipulates one or more independent variables and observes the degree to which the dependent variables change. The basic point of an experiment is to change the levels of one or more X variables and examine the resulting impact on Y while at the same time controlling (holding constant) other variables that might impact Y. LABARATORY VS FIELD EXPERIMENT LAB EXPERIMENT: Research investigation in which investigators create a situation with exact conditions in order to control some variables and manipulate others. FIELD EXPERIMENT: Research study in a realistic situation in which one or more independent variables are manipulated by the experimenter under as carefully controlled conditions as the situation will permit Example of lab experiment - Researchers hypothesized that loud (vs. quiet) music (X) led to more dancing (Y) - They recruited subjects and randomly assigned them to listen to the same song at high volume (condition A) or low volume (condition B) in a private room - They then measured body movements (Y) using a predetermined methodology - In support of their hypothesis, body movements were higher in condition A than in condition B Example of Field Experiment - The same researchers went to several randomly selected clubs with similar characteristics (price of drinks, crowd size, type of music) and measured loudness of music (X) and amount of body movements (Y) - In support of their hypothesis, they observed more body movements in clubs with louder music
Survey design
See questionnaire preparation checklist in Survey Design lecture slides (some copied below) for summary - these are hidden slides, they just summarize everything mentioned in the lecture nicely
Learning Objective 3: Give an important reason that marketers are interested in people's motives.
So they can meet those needs! Marketers need to understand individuals' known—and sometimes unknown—needs and what motivates them if they are to develop and deliver products and services that meet those needs.
total sampling elements (TSE)
TOTAL SAMPLING ELEMENTS (TSE) The number of population elements that must be drawn from the population and included in the initial sample pool in order to end up with the desired sample size.
Learning Objective 1: Define telescoping error and recall loss and explain how they affect a respondent's ability to answer questions accurately.
Telescoping error refers to people's tendency to remember an event as having occurred more recently or more distantly than it did. Recall loss means they forget it happened at all.
Learning Objective 2: Explain what is meant by disguise in a questionnaire context.
The amount of disguise in a questionnaire is the amount of knowledge hidden from the respondent as to the purpose and/or sponsor of the study. An undisguised questionnaire makes the purpose of the research obvious by the questions posed; A disguised questionnaire attempts to hide the purpose of the study.
Learning Objective 1: Explain the concept of structure as it relates to questionnaires.
The degree of structure in a questionnaire is the degree of standardization imposed on it. In a high structure questionnaire, the questions to be asked and the responses permitted by the subjects are completely predetermined. In a low structure questionnaire, the response categories are not provided
Sampling error
The difference between results obtained from a sample and results that would have been obtained had information been gathered from or about every member of the population *** - Decreased by increasing sample size - Can be estimated (assuming probability sample) - Usually less troublesome than other kinds of error .
VALIDITY
The extent to which differences in scores on a measuring instrument reflect true differences among individuals, groups, or situations in the characteristic that it seeks to measure are true differences in the same individual, group, or situation from one occasion to another, rather than systematic or random errors.
Learning Objective 2: List the four scales (levels) of measurement.
The four types of scales on which an attribute can be measured are nominal, ordinal, interval, and ratio scales.
Learning Objective 2: Explain which of the two types of experiments has greater internal validity and which has greater external validity.
The laboratory experiment typically has greater internal validity because it allows greater control of the variables. Field experiments are generally considered more externally valid, meaning that their results are better able to be generalized to other situations.
Illustration of Difference between random and systematic error
The old rifle is unreliable; despite the fact that the sights are set on the center of the target, the shots go off in random directions. The first new rifle is relatively reliable—it hits about the same spot on the target each time—but its sights are set incorrectly in the center diagram. The error is systematic and not random, but the rifle still misses the mark. The right-hand diagram shows a new rifle with its sights set correctly. Only in the right-hand diagram could a user of any of the rifles be expected to hit the center of the target with regularity. -This represents a measure that is both reliable and valid.
Learning Objective 3: Cite three factors that influence the necessary sample size.
The three factors that influence sample size are: 1) the desired degree of precision 2) the desired degree of confidence 3) the degree of variability in the population on the parameter in question.
Learning Objective 1: Describe the two basic means of obtaining primary data.
The two basic means of obtaining primary data are communication and observation. Communication involves questioning respondents to secure the desired information using a data collection instrument called a questionnaire. Observation involves scrutinizing the situation of interest and recording the relevant facts, actions, or behaviors.
CAUSAL RESEARCH
Type of research in which the major emphasis is on determining cause-and-effect relationships. XXX Predictor variable Independent variable YYYY Outcome variable Dependent variable When a company needs precise answers about the effects of various proposed marketing actions on important outcomes, managers use causal research. - Using test markets to determine which version of a product to offer, which package design to use, which advertising campaign is most effective, which price to charge, and so on are examples of causal research. - Strategy-related decision problems (1) There must be consistent variation between the cause and the effect, (2) The time order of the cause and the effect must be correct, and (3) Other explanations must be eliminated
Developing the data collection form Step 5: Determine Wording of Each Question
Use simple words - Language used should be driven by the ability level of the population; err on the side of simplicity - E.g., arbitrarily vs. randomly avoid ambiguous words and questions, double-barreled questions, generalizations and estimates, Problem words: All, Always, And, Feel, Government Similarly, avoid the possibility of ASSUMED CONSEQUENCES = When a question is not framed to clearly state the consequences and thus generates different responses from individuals who assume different consequences. Assumed consequences differ between people: When people were asked whether they would: "favor or oppose taking military action in Iraq to end Saddam Hussein's rule," 68% said they favored military action Assumed consequences defined in the question: However, when asked whether they would "favor or oppose taking military action in Iraq to end Saddam Hussein's rule even if it meant that U.S. forces might suffer thousands of casualties," responses were dramatically different; only 43% said they favored military action BETTER SOLUTIONS TO ASSUMED CONSEQUENCES PROBLEM 1)Follow up the question with an open ended "why" question - E.g., "Do you favor a 5% increase in taxes?"(Yes/No) followed by "Why did you choose the answer you chose to the previous question?" -> code responses for themes 2)Follow up the question with a fixed alternative question that allows people to select from a preset range of answers
Learning Objective 2: List some of the techniques researchers use to secure respondents' cooperation in answering sensitive questions.
When asking sensitive questions, researchers may find it helpful to: (a) guarantee respondent anonymity or confidentiality; (b) make use of a counter-biasing statement; (c) phrase the question in terms of others and how they might feel or act; (d) put sensitive questions near the end; (e) use categories or ranges rather than specific numbers;
Key Issues for Collecting Information by Observation: SETTING
Where the study takes place NATURAL SETTING - Subjects are observed in the environment where the behavior normally takes place. - Shopping in a store - Using or consuming a product at home CONTRIVED SETTING - Subjects are observed in an environment that has been specially designed for recording their behavior. - "fake" store - computer simulation - E.g., researchers might want to examine the effectiveness of point-of-purchase display materials and could include such materials in the stores where observations were taken. Then measure the amount of time consumers spend reading the display materials or the time they spend looking at the brand being promoted.
Two Methods of Data Collection
communication and observation
Key Issues for Collecting Information by Observation
degree of structure, disguise, setting, method of administration
Sampling plan
the process of selecting the people or objects (i.e., companies, products, and so forth) to be surveyed, interviewed, or observed