measurement
-The process of describing some property of a phenomenon of interest usually by assigning # in a reliable & valid way
concepts
generalized ideas that represent something of meaning
-age, sex, # of children are relatively stable
operationalization
a process that involves identifying scales that correspond to variance in the concept
-researchers measure concepts with operationalization
scales
provide a range of values that correspond to different values in the concept being measured
constructs
a term used for concepts that are measured with multiple variables
-multiple variables in one concept provide a more complete account
4 levels of scale management
nominal, ordinal, interval, ratio
nominal
-simplest type of scale
-qualitative
-can but doesnt have to be a # because no quantities are being represented
-used by marketing researchers quite often
-example: uniform #s, airport terminals, school bus #s
ordinal
-have nominal properties
-include ranking scales
-allow things to be arranged based on the concept they possess
-example: "win, place & show" in a horse race says they placed but not in a specific order
interval
-have both nominal & ordinal properties
-captures info about differences in quantities of a concept
-doesnt exactly represent some phenomenon
-example: fahrenheit temperature scale' "80 degrees is hotter than 40 degrees
ratio
-represents highest form of measurement
-represents absolute meaning
-provides iconic measurement
-0, therefore has meaning in that it represents an absence of some concept
mathematical & statistical analysis of scales
-researcher must know the meaning behind the # before useful conclusions can be drawn
-example: averaging the #s used to identify school buses is meaningless
discrete measures
-take on only one of a finite # of values
-most often used to represent a classificatory variable
-can possess ordinal & nominal properties
-the central tendency of a nominal or ordinal discrete measure is best captured by the mode
continuous measures
-assign values anywhere along some scale range in a place that corresponds to the intensity of some concept
-ratios measures are continuous measures
-the mean and standard deviation may be calculated from continuous data
reliable & valid measures
an attribute is a single characteristic or fundamental feature of an object, person, situation, or issue
-measures of attributes are often combined to represent some less concrete concept
index measures
multi-item instruments for measuring construct (or composite measures)
-assigns a value based on how much of the concept being measured is associated with an observation
-often formed by putting several variables together
reliability
an indicator of a measure's internal consistency
-a measure is reliable when different attempts at measuring something converge on the same result
internal consistency
represents a measure's homogeneity
-uses the split-half method for checking reliability
split-half method
take 1/2 of the items from the scale & checking them against the results from the other 1/2
-the 2 scales should correlate highly & produce similar scores
coefficient alpha
represents internal consistency by computing the average of all possible split-half reliabilities for a multiple item scale
-coefficient demonstrates whether or not the different items converge
coefficient value ranges
0.80-0.96 -- very good reliability
0.70-0.80 -- good reliability
0.60-0.70 -- fair reliability
below 0.60 -- poor reliability
test-retest reliability
involves administering the same scale or measure to the same respondents at 2 separate times to test for stability
-if the measure is stable over time, the test should obtain similar results
-represents a mesure's repeatability
-reliability is necessary b
validity
the accuracy of a measure or the extent to which a score truthfully represents a concept
-good measures should be both precise & accurate
3 approaches of establishing validity
-face: refers to the subjective agreement and reflects on the concept
-criterion: addresses the question "does my measure correlate with other similar measures?"
-construct
questionnaire
the primary tool for building responses to research questions
questionnaire design
one of the most critical stages in the survey research process
question relevancy
-all info collected should address a research question that will help the decision maker address the current marketing problem
-researcher should be specific about data needs
-research should be accurate
-one should use simple, understandable,
unbiased, u
open ended response
questions pose some problem or topic and ask respondents to answer in their own words
-most beneficial in exploratory research
-cost is usually higher
-interviewer bias may influence answer
fixed alternative response
(close ended)
questions give respondents specific limited-alternative responses and ask them to choose the one closest to their viewpoint
-takes less time
-requires less interviewer skills
-easier for respondent to answer
-each alternative should be mutua
types of fixed alternative questions
-simple dichotomy: questions present 2 alternatives
-multiple choice: asks about general frequency of occurrence
-frequency determination: questions ask for an answer about the general frequency of occurrence
-checklist- questions allow respondents to pro
leading questions
questions that suggest or imply certain answers
-"bandwagon effect
loaded questions
questions that suggest a socially desirable answer or are emotionally charged
-certain answers to questions are more socially desirable than others
avoiding ambiguity
questions can often be too ambiguous because they are too general
-example: are you satisfied with the present system of searching for material using a smartphone or computer?
avoiding assumptions
researchers should not place the respondent in a bind by including an implicit assumption in the question
-also should not assume that the respondent has previously thought about the issue
unaided recall questions
questions that give no clue as to the brand of interest
aided recall questions
questions that provide a clue to jog the respondent's memory
question sequence
the order of questions may serve several functions
order bias
-can result from an alternative answer's position in a set of answers or from the sequencing of questions
-tends to distort survey results
-can be caused by asking specific questions before asking about broader issues
funnel technique
asking general questions before specific questions in order to obtain unbiased responses
-allows researchers to understand the respondent's frame of reference before asking more specific questions
survey flow
refers to the ordering of questions
-often certain sections of a questionnaire are irrelevant to a particular respondent
-asking a question that does not apply to the particular respondent can cause biased answering or survey breakoffs
survey breakoff
respondent stops answering questions
multiple-grid question
(matrix table)
-presents several similar questions and corresponding response alternatives arranged in a grid format
survey technology
software programs like Qualitrics allows special features to facilitate design
advantages to survey technology
-response quality
-timing
-randomized assignment
physical features of survey technology
-tracking interest
-status bar
-prompting
-piping
tracking interest
heat map that tracks parts of an image or advertisement that captures the respondent's attention the most
status bar
visual indicator of questionnaire length
prompting
informs respondent they skipped a question or didn't provide enough info
piping
allows responses to a previous question to be inserted into later questions
pretesting & revising
allows researchers to determine if respondents have any difficulty
back translation
the process of translating a questionnaire from one language to another and having it translated back again by the second independent translator
sample
a subset or some part of a larger population
population
any complete group of sharing some common set of characteristics
population element
an individual member of the population
census
an investigation of all the individual elements making up the population-- a total enumeration rather than a sample
reasons to sample
-cuts costs
-reduces labor requirements
-gathers vital information quickly
-accurate & reliable results
defining target population
-once the decision to sample has been made, the first question concerns the target population
-household is the appropriate population element for consumer research
stages in the selection of a sample
-define target population
-select sampling frame
-determine if probability/nonprobability method is being used
-plan to select sampling units
-determine sample size
-select sampling units
-conduct field work
sampling frame
a list of elements from which the sample may be drawn
-also known as working population
-errors occur when certain sample elements are excluded or when the entire population is not accurately represented
sampling services
firms specializing in providing lists or databases of specific populations
-also known as list brokers
-example: Equifax city directory provides complete, comprehensive and accurate business and residential information
online panels
list of respondents who have agreed to participate in marketing research
-generally contains millions of potential respondents
-the more specific, the more expensive
sampling unit
a single element or group of elements subject to selection in the sample
-if target population has been divided into 2 units
primary sampling unit
designates units selected in the first stage of sampling
secondary sampling unit
describes units in successive stages of sampling
probability sampling
every population element has a known, nonzero probability of selection
-eliminates the bias in non-probability sampling because probability sampling is random
-example: simple random sampling
non-probability sampling
probability of any member of the population being chosen is unknown
-pragmatic, used in market research
convenience sampling
type of non-probability sampling by obtaining people/units that are conveniently available
-used to obtain a large # of completed questionnaires quickly and economically
judgement sampling
a non-probability technique in which an experienced individual selects the sample based on his/her judgement
-example: consumer price index
-often used by test market cities because their profile matches closely with the national profile
quota sampling
used to ensure that various subgroups in a population are represented on a pertinent sample characteristics to the exact extent that investigators desire
-interviewers have a quota to achieve
-may be appropriate when the researcher knows that a certain de
advantages of quota sampling
-speed of data collection
-lower costs
-convenience
snowball sampling
a variety of procedures in which initial respondents are selected by probability methods
-additional respondents are then obtained from info provided by initial respondents
-technique is used to locate members of rare populations by referrals
-possible bi
simple random sampling
a sampling procedure that assures each element in the population will have an equal chance of being included in the sample
-used when populations consist of a large # of elements
systematic sampling
an initial starting point is selected by a random process; then every #th number on the list is selected
-not actually a random selection procedure, yields random results if arrangement of items is listed at random
stratified sampling
a subsample is drawn using simple random sampling within each stratum
-provides a more efficient sample
-random sampling error will be reduced
proportional sampling
the # of sampling units drawn from each stratum is not allocated in proportion to the population size but is dictated by analytical considerations
-ensures adequate # of sampling units in every stratum
cluster sampling
classified as a probability sampling technique because of either the random selection of clusters or the random selection or elements within each cluster
-the primary sampling unit it so longer the individual element in the population (grocery stores) but
multistage area sampling
involves 2 or more steps that combine some of the probability techniques already described
-typically geographic areas are randomly selected in lower population units
-example: bureau of census provides maps, population info, demographic characteristics f
degree of accuracy in appropriate sample design
cost savings is a tradeoff for a reduction in accuracy
resources in appropriate sample design
if resources are restricted, certain options will be eliminated
time in appropriate sample design
constraints restrict sampling to simpler methods
advance knowledge of the population in appropriate sample design
lack of adequate lists can rule out some types of sampling
national vs local project in appropriate sample design
geographic elements will influence sample design