Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
111 Cards in this Set
- Front
- Back
Advantages and disadvantages of different communication methods for collect primary data
|
sampling control, information control, and administrative control
|
|
aided recall
|
"do you remember seeing ads for personal computers?"
|
|
Attitudes/ Opinions
|
Attitudes are your overall evaluations of something
opinions are what you actually verbally express about something marketers also measure many "attitude-like" variables including value, quality, and satisfaction |
|
barcode scanner
|
observes actual purchase behavior by the normal recording of transactions in retail establishments. allows researchers to look at current and past purchase to predict future purchases
ex kroger plus card |
|
Behavior
|
what people have done or are doing, typically with respect to purchase and consumption of goods and services
its hard to measure future behavior with a questionnaire because the action hasnt happened yet. So, it is better to use a questionnaire when measuring past behavior |
|
branching question
|
A technique used to direct respondents to different places in a questionnaire, based on their response to the question at hand.
|
|
Collecting info through observation
|
structured/ unstructured
disguised/ undisguised natural/ contrived settings |
|
Comparative
|
a scale that requires subjects to make their ratings as a series of relative judgments or comparisons rather than independent assessments
|
|
comparative > constant sum method
|
an individual divides some given sum among two or more attributes on a basis such as importance or favorability
this is a RATIO level of measurement bc you can conduct absolute comparisons between two different attributes allows you to see which attributes are more important to consumers in comparison to other attributes shows what areas to improve on |
|
Constant Sum Scale Question
|
Consumers rank the statements based on importance from a preset scale they are given. ex you have a total of 100 points, allocate the points based on what you find most important when choosing a coffee shop
|
|
Construct Validity
|
- Assessment of how well the instrument captures the construct, concept, or trait it is supposed to be measuring.
|
|
Content Validity
|
The adequacy with which the important aspect of the characteristic are captured by the measure
|
|
Contrived Settings
|
Subjects are observed in an environment that has been specifically designed for recording their behavior
ex a virtually reality on a computer. virtra specializes in giving virtual reality contrived settings LAB EXPERIMENTS |
|
Demographics
|
divides a population into groups based on:
age education income gender occupation marital status social class |
|
Dichotomous Question
|
- A fixed-alternative question in which respondents are asked to indicate which of the two alternative responses most closely corresponds to their position on the subject.
- Example: Do you think laws requiring passengers in motor vehicles to wear seat belts are needed? Yes No |
|
Disguised
|
subjects are not aware that they are being observed
ex. mystery shoppers this is preferred but there are may be ethical concerns because people's privacy may be invaded. a way to resolve this concern would be to tell the people afterward and make sure it is okay with them; have a debriefing |
|
Disguised Structured
|
Form:
standard questions/ responses characteristics: simple administration simple analysis difficult interpretation least used methos |
|
Disguised Unstructured
|
Form:
Standard Stimuli, Non standard responses characteristics: difficult analysis subjective interpretation suited to exploratory research |
|
Electrical/ Mechanical observation
|
an electrical or mechanical device observes a phenomenon and records the events that take place
ex barcodes |
|
establishing the validity of measures
|
predictive validity
content validity construct validity reliability |
|
eye camera
|
observes visual attention by tracking what elements of a marketing stimulus the eye focuses on, and how long it fixates on each element. the speed of movement makes it hard to track
|
|
funnel approach
|
An approach to question sequencing that gets its name from its shape, starting with broad questions and progressively narrowing down the scope.
|
|
galvanometer
|
observes emotional arousal by measuring changes in electrical resistance of skin (due to perspiration) is response to some marketing stimulus
|
|
Graphic
|
individuals indicate their ratings of something by checking a point on a line that runs from one extreme to the other
this are no discrete answer choices if the respondents are not that knowledgeable about the topic the response objects can be frustrating bc they may not feel strongly enough about the statements. however, if the subjects are very knowledgeable you can get more specific data about their feelings this is an INTERVAL level of measurement |
|
General Itemized Scale
|
can be used to measure attractiveness, likelihood, pleasure, desirability, consistency, and agreement
they are usually very easy to design it is a type of INTERVAL scale |
|
General Itemized Scale Question
|
using the scale provided, please rate the level of service quality provided by Central perk in each area
|
|
Graphic Scale Question
|
please evaluate each of the following attributes of cd players according to how important the attribute is to you personally by placing an x at the position on the horizontal line that most accurately reflects your feelings
|
|
guidelines for question sequencing
|
• Use simple, interesting opening questions
• Use the funnel approach, asking broad questions first • Carefully design branching questions • Ask for classification information last • Place difficult or sensitive questions near the end |
|
Hawthorne Effect
|
humans act differently if they know they are being observed
this is not what marketing researches want to happen |
|
hypothetical construct
|
a concept used in theoretical models to explain how things work. hypothetical constructs include such things as attitudes, personality, and intentions, concepts that cant be seen but that are useful in theoretical explanations
|
|
implicit alternative
|
A problem that occurs when a question is not framed so as to explicitly state the consequences, and thus it elicits different responses from individuals who assume different consequences.
|
|
Internet Based Questionnaires
|
o Sampling control—list of email addresses, younger people, advantage to mail is the intended respondent will respond, least expensive
o Information control—pictures and graphics, sequenced o Administrative control—quickest turnaround, |
|
Intentions
|
Anticipated or planned future behavior
marketers often need this type of info to assess demand for a good/service estimating accurately is difficult because often times what people say they will do is not what they will actually do. therefore, results are not always trustworthy, and this method should be used with caution |
|
Interpreting Rating Scales
|
using normative standards
do not look at the average score in isolation from the actual ratings because it may be misleading if you are trying to determine the real meaning of a summary statistic, you need to compare it to the actual ratings. the average may imply that the respondent has neutral feelings but when you actually look at the responses you find out that the majority of the responses were positive but there were a few negative outliers that brought the average down time based norms: perform tracking and them compare the new data to previous data collected |
|
Interval
|
Allows you to compare relative interval size and gives more insight into the difference between each level
you can find mean ratings you can find the percentage or frequency count per scale point ex. rating people give for each song |
|
Itemized
|
Individuals must indicate their ratings of something by selecting the response category that best describes their position on the position of object (very satisfied, somewhat...)
ex general itemized scale summated ratings (likert scale) semantic differential snake diagram |
|
item non response
|
a source of non sampling error that arises when a responded agrees to an interview but refused, or is unable, to answer specific questions
|
|
Knowledge/ Awareness
|
insight into, or understanding of, facts about some object or phenomenon
marketers want to know what individuals believe about products, brands, companies, ads, etc. |
|
leading questions
|
A question framed so as to give the respondent a clue as to how he or she should answer.
|
|
level of disguise
|
degree to which respondents are informed about who is sponsoring the study and the actual purpose of the study
if they are told less about the study, it has a high disguise. if they are told a lot about the study is has a low disguise in the same questionnaire the level of disguise may change. it may start with broad questions first (high disguise) to more specific and purpose related questions toward the end (low disguise) |
|
level of structure
|
the degree of standardization in the questions asked and the response alternatives.
if it is low there would be a lot of open ended questions if the structure is high there are more likely to be a fixed set of answers |
|
lifestyle (psychographics)
|
involves a customer's way of life, what interests them, and what they like
provides: better understanding about how customers live and customer interests lead to more effective marketing strategies Bill Clinton video about how he considered psychographics while running for election |
|
Mail Questionnaire
|
o Sampling control—one or more accurate mailing lists of people in population of interest, purchased from a list company, little control in securing a response from the intended respondent
o Information control —no interviewer bias, include visuals, anonymity offered to the respondents, do not allow clarification of questions, respondents can look at the whole questionnaire before answering it o Administrative Control—low response rates, anonymity, low cost, quality control, take longer to implement |
|
the management problem
|
you cant always observe data relative to a hypothetical construct
|
|
Measurement
|
rules for assigning numbers to objects in such a way as to represent quantities of attributes
|
|
Measuring Ad Effectiveness
|
unaided recall
aided recall recognition |
|
Measuring variables other than attitudes
|
personality: the normal pattern or an individuals behavior. measure using LIKERT scale
psychographics: characteristics of a consumer's lifestyle. measure using a LIKERT SCALE intentions: expected or planned future behavior. measure with a GENERAL ITEMIZED scale motives: the reason why someone acts the way they do. measure with a GENERAL ITEMIZED scale whenever possible you should use the same type of scale to measure different variable |
|
methods of assessing attitude
|
observation of behavior
indirect techniques performance of objective tasks physiological reactions self report measures |
|
Motivation
|
a need, want, drive, wish, desire, impulse, or any other state that moves you and channels your behavior toward a goal;
why you behave as you do |
|
Multichotomous Question
|
- A fixed-alternative question in which respondents are asked to choose the alternative that most closely corresponds to their position on the subject.
- Example: Age How old are you? Less than 20 20-29 30-39 40-49 50-59 60 or over |
|
Nature of Administration
|
interview (personal or telephone) or self administered surveys (mail or internet)
personal or telephone interviews are administered by an interviewer where as mail or internet surveys are self administered |
|
Nominal
|
comes from the latin word meaning name. it exists in name only; it is not actual
can help categorize or identify membership in a category you can also discover the percentage or frequency of each category. the most frequent category is called the mode because it occurs the most often you cannot calculate the mean bc there is no quantitative data available from this ex genre of music |
|
Normative Standards
|
a comparative standard used to provide meaning to raw scale scores
|
|
observational study:
row of computers and let the subjects browse through the website while you monitor what they do |
this is a contrived setting bc they are not in a "natural" setting like their home
this is undisguised bc they would know they were being observed it would be unstructured bc observers would be simply observing clicks and how long they pause on certain pages but there would be no specific things to look for the observer would be a human. you could possibly use an eye camera to monitor where the subject looks on the screen which in this case the observer would be a mechanical device |
|
observed response =
|
truth + systematic error+ random error
|
|
obtaining primary data: communication and observation
|
-Communication- a method of data collection involving questioning of respondent’s to secure the desired information, using a data collection instrument called a questionnaire Advantages: versatility, speed, cost Disadvantage: tempted to impress and not tell the truth -Observation - a method of data collection in which the situation of interest is watched and the relevant facts, actions, or behaviors are recorded Advantages: Objectivity and accuracy Disadvantages: limited to present behavior; forced to wait for events to occur |
|
ordinal
|
the numerical data only reflects the order of the data
it does not tell you the difference between the different items that are ordered; it simply lists the order in which they occur you can calculate percentages for each rank level you can calculate the median rank of the data ex number of each song in a list, or the order in which the song is played |
|
Personal Interviews
|
o Sampling control—some cases a random sample but usually want a list called population elements to draw from
o Information control—can be conducted using questionnaires with any degree of structure; great strength is the ability to rephrase. A lot of information can be gathered, can be subject to bias o Administrative control—cost more to conduct, and take a lot of time |
|
Personality
|
normal patterns of behavior exhibited by an individual; the attributes, traits, and mannerisms that distinguish one person from another
|
|
population-based norms-
|
comparing scores to the summary statistics of scores obtained by similar entities
|
|
Potential Sources of Variation in Scores
|
true differences in the characteristic being measured
differences in other relatively stable characteristics of the individual differences due to temporary personal factors differences due to situational factors differences due to variations in administration difference due to the sampling of items difference due to lack of clarity of the measuring instrument differences due to mechanical factors |
|
Predictive Validity
|
The usefulness of the measuring instrument as a predictor of some other characteristic or behavior of the individual
|
|
Procedure for Developing a Questionnaire
|
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 question 6. determine question sequence 7. determine physical characteristics of questionnaire 8. develop recruiting message or script 9. reexamine steps 1-8 and revise if necessary 10. pretest questionnaire and revise if necessary |
|
Random Error
|
error in measurement due to temporary aspects of the person or measurement situation and which effects the measurements in irregular ways
|
|
randomized response model
|
• An interviewing technique in which potentially embarrassing and relatively innocuous questions are paired and The question the respondent answers is randomly determined but is unknown to the interviewer.
|
|
Ratio
|
a zero means an absolute or natural absence of value , which gives a lot more information than in an interval where 0 is a relative reference point
you can compare the absolute magnitudes of numbers, meaning it is more exact than in interval comparing frequently used to analyze purchase behavior bc it is easy to quantify you can calculate the mean, median, and mode ex. length of each particular song |
|
recall loss
|
a type of error caused by a respondent’s forgetting that an event happened at all
|
|
recognition
|
"do you remember seeing this ad for Dell computers?"
|
|
Reliability
|
Ability of a measure to obtain similar scores for the same object, trait, or construct across time , across different evaluators, across the items forming the measure
|
|
respondent's willingness to produce a response
|
3) Amount of work involved in producing an answer
4) The inability to express an answer 5) Sensitivity of the issue 6) When an issue is embarrassing |
|
response latency
|
observers strength of knowledge, beliefs, or prefrences by recording the amount of time it takes to respond to a questions. the quicker you answer, the more you know about the question. this is a form of indirect observation
|
|
reverse scaling
|
reverse scaling: a technique in which some of the items on a multi item scale are written so that the most positive responses are the the opposite end of the scale from where they normally appear
this is an attempt to reduce RESPONSE SET BIAS which is a problem that arises when respondents answer questionnaire items in a similar way without thinking about the items. so, by putting answers in an unlikely spot it would (in theory) force the respondent to pay more attention to their answers. however, this may not work bc if people are tired or bored or just dont care about the survey then they may still answer incorrectly |
|
Scale Design Considerations
|
how many items should you use to measure one construct?
how many scale positions should you use? should you consider adding any additional options? should reverse scaling be used? |
|
scales of measurement
(from highest level of measurement to lowest) |
ratio
interval ordinal nominal |
|
Self Report Categories
|
graphic
itemized comparative |
|
Different types of self-report rating scales
|
general itemized-ratings scale, Likert scale, semantic differential scale, graphic ratings scale, and constant sum scale -
|
|
Selection Criteria
|
Consider the perspectives of the:
managers, respondents, modes of administration |
|
self report measures
|
a methods of assessing attitudes in which individuals are asked directly for their beliefs about or feeling toward an object or class of objects
|
|
Semantic Differential
|
the subjects are asked to check which cell between a set of bipolar adjectives or phrases describes their feelings toward the object
precise and concise adjectives are used this in an INTERVAL scale the challenge is to find two ends of the spectrum with bipolar adjectives |
|
Semantic Differential Scale Question
|
Rate Central Perk on each of the following characteristics by placing an x on the blank that best describes your evaluation ex insufficient seating vs sufficient seating
|
|
Snake Diagram
|
takes the results from semantic differential statements and connects them, thereby depicting the profile of the objects being evaluated. it is called a snake diagram based on the zig zag lines from connecting the semantic results
|
|
Sources of Random Error
|
the person's mood states, health conditions, fatigue levels, can change a persons answers to certain questions
the differences in procedures of different interviewers can cause random error social influences may cause random error. ex. if your parents/friend are watching you, you may answer differently than if you were alone physical conditions may cause random error such as weather. if you feel comfortable you may give different answers than if you weren't time pressures may cause random error bc if someone is rushed they may not answer completely the methods for recording the response may cause random error |
|
Sources of Systematic Error
|
Leading questions are sources bc the wording will lead the participants to give answers that are different than what they would have otherwise responded with
incompleteness is another source bc if the questions do not address all the important issues it will not provide correct data if your questions cover irrelevant issues there will be error bc the person doesnt care or know about the questions. they will just become noise to the participant and they may not answer correctly honesty is another source bc the willingness of a person to reveal honest answers will affect the amount of error |
|
Split Ballot
|
- A technique used to combat response bias in which one phrasing is used for a question in one-half of the questionnaires while an alternate phrasing is used in the other one-half of the questionnaires.
- Example: Do you think gasoline will be more expensive or less expensive next year than it is now? More expensive Less expensive Do you think gasoline will be less expensive or more expensive next year than it is now? More expensive Less expensive |
|
Structured
|
this is slightly different from using structure in collecting data through communication approach
the problem has been defined precisely enough so that the behaviors that will be observed can be specified beforehand, as can the categories that will be used to record and analyze the situation observers will often times have a check list of things to look for or may simply write small paragraphs about what they observed in terms of what they note and record |
|
Summated Ratings (Likert Scale)
|
a self report technique for attitude measurement in which respondendts indicate their degree of agreement or disagreement with each number of statments
advantage is that is uses more detailed statements than other types of measures, however this can be a disadvantage if the survey is over the phone bc a long detailed statement may be confusing to the participant helps you evaluate strengths and weaknesses of each item based on how strongly people agree/disagree with each statement you can find the summated score and then the average score you can get a general idea of a consumers overall attitude of something it is an INTERVAL scale bc there are equal levels of intervals between each answer choice |
|
Summated or Likert Scale Question
|
Please rate the Central Perc on each of the statements listed. use the scale provided to indicate your level of agreement/disagreement with each statement
strongly agree, somewhat, neither, somewhat, strongly disagree |
|
systematic error
|
error in measurement that is also known as constant error since it affects the measurement in a constant way. the amount of error would stay the same each time
ex you are trying to measure something but the measurements on the tape are wrong by the same amount for each interval. therefore, when you take your measurements they will all be off by the same amount |
|
Telephone Interviews
|
o Sampling control—generate a list of companies or businesses, don’t rely on phone books, random digit dialing, plus-one sampling, in-bound telephone surveys (punch buttons)
o Information Control—both fixed and open ended questions, computer assisted interviewing (computer manages and you use keyboard) o Administrative Control—more costly compared to mail or internet, there is higher response rates, interviewers are trained (not as intense as personal), unlike personal there are many interviewers, |
|
Telescoping
|
a type of error resulting from the fact that most people remember an event as having occurred more recently than it did
|
|
time-based norms-
|
tracking scores for an entity over time; using ratings from earlier time periods as norms for interpreting rating from future time periods
|
|
Trail run: use some sort of internet program or service that records the data about the clicks on the page, how long the person stayed on each page, and other details about their time browsing the website
|
this would require mechanical observation to record info about the clicks
this would be in a natural setting bc it would be in the subjects own home or wherever they access internet it would be disguised bc the subjects would not know they were being observed is there was not a trail website, the same thing could be studied but with competitors web pages |
|
Types of mechanical observers
|
video camera
galvanometer voice pitch analysis eye camera barcode scanner response latency |
|
unaided recall
|
"for what type of products and brands do you remember seeing ads"
|
|
Understanding Motivation
|
Marketers need to understand so they can anticipate consumer needs and then offer products and services to satisfy those needs
researchers interested in motives because of 2 things: they help to determine why people behave as they do; they tend to be more stable than behavior understanding what drives a person's behavior makes it easier to understand behavior itself |
|
Undisguised
|
Subjects know they are being observed
|
|
Undisguised Structured
|
Form:
standard questions/ responses characteristics: simple administration simple analysis suitable for facts of clear cut opinions due to forced alternatives |
|
Undisguised Unstructured
|
Form:
Non standard questions and responses characteristics: flexible difficult interpretation interviewer influences better for exploratory research |
|
Unstructured
|
the problem has not been specifically defined so a great deal of flexibility is allowed for the observers in terms of what they note and record
|
|
Validity
|
The extent to which differences in scores on a measuring instrument reflect true differences among individuals, groups, or situations in the the characteristics that it seeks to measure, or true differences in the same individual, group, or situation from one occasion to another, rather than systematic or random errors. (correctness of a measure)
|
|
video camera
|
observes overt behavior by recording it for later human viewing and observation. a problem is you cannot capture all aspects of the behavior that you may want to know about
|
|
voice pitch analysis
|
observes emotional arousal by computer analysis of changes in relative vibration frequency of human voice in response to some marketing stimuli
|
|
Which method of administration would be best to use for the restaurant questionnaire?
|
either a personal interview or telephone interview would probably be best
however there is a trade off between them telephone interview costs less time and money, but the risk of confusion when involving complex info and asking for detailed responses of the phone |
|
3 important decisions in collecting information by communication
|
level of structure
level of disguise nature of administration |
|
how many items should you use to measure one construct?
|
global measures: a measure designed to provide an overall assessment of an object or phenomenon, typically using one or two items
composite measures: a measure designed to provide a comprehensive assessment of an object or phenomenon with items to asses all relevant aspects or dimensions |
|
how many scale positions should you use?
|
make sure to include a sufficient number of scale points
should there be a neutral point? you must consider that if the interviewees dont have strong enough opinions on the topic they may respond to too many questions with a neutral response and as a result you will not get high quality data |
|
should you consider adding any additional options?
|
ex should you add "i dont know" option? if the respondents are knowledgeable enough then no, you should not provide this option
|
|
3. Determine content of individual questions
|
do you have enough questions to adequately cover the issue?
can respondents provide desired information? do they have the knowledge? how to overcome this is use a 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. often called SCREENING QUESTIONS will respondents provide desired information? is the question necessary? extremely important when dealing with sensitive questions ways to avoid sensitive questions: add a small level of disguise to the question another way is to ask for general rather than specific info provide a counter bias statement. ex. "studies show that a majority of credit card holders can afford to pay no more than the minimum required payment each month. in how many of the last 6 months have you paid only the minimum?" you might be in the danger of leading the respondent to give you a false answer |
|
5. Determine Wording of Each Question
|
make sure questions are not worded in a leading manner, or in a way that gives the respondents a clue as to how he/she should answer; a likert scale can help you solve this issue bc you give the respondents equal opportunity to agree or disagree, and then you can use more specific statements to see which parts they agree or disagree to
avoid generalizations and ambiguous terms; avoid this by using a specific time fram. ex. instead of saying in a typical month...ask last month break double barreled questions into separate questions; double barreled questions call for two responses i none question and create confusion for the respondent avoid assumed consequences which are problems that occur when a question is not framed so as to clearly state the consequences and therefore generates different responses from individuals who assume different consequences |
|
10. pretest questionnaire and revise if necessary
|
conduct initial pretest using personal interviews with people from the same population you plan to study
observe for difficulties and obtain comments concerning any problems from both respondents and interviewers revise and repeat the first two steps conduct more pretest interviews using the mode of administration planned for study (again with people from the same population you plant to study) revise questionnaire again to overcome problems repeat the last two steps until no more problems are encountered |
|
structure and disguise in questionnaires
|
Structured and Undisguised
Form · Standardized questions · Standardized responses · Example: Fixed alternative questions Characteristics · Simple administration · Simple analysis · Suitable for facts or clear cut opinions due to forced alternatives Unstructured and Undisguised Form · Non-standardized questions · Non-standardized responses · Example: Depth interviews Characteristics · Flexible · Difficult interpretation · Interviewer influenced · Better for exploratory research Structured and Disguised Form · Standardized questions · Standardized responses Characteristics · Simple administration · Simple analysis · Difficult interpretation · Least used method Unstructured and Disguised Form · Standardized stimuli · Non-standardized responses · Example: Projective Techniques Characteristics · Difficult analysis · Subjective interpretation · Suited for explo |