MKTG 461
Terms
undefined, object
copy deck
- Validity
- The degree to which a research instrument serves the purpose for which is was constructed; it also relates to the extent to which the conclusions drawn from an experiment are true.
- Construct validity
- The degree to which researchers measure what they intended to measure, and to which the proper identification of the independent and dependent variables were included in the investigation.
- External validity
- The extent to which the measured data results of a study based on a sample can be expected to hold in the entire defined target population. In addition, it is the extent that a causal relationship found in a study can be expected to be true for the entire defined target population.
- Internal validity
- The certainty with which a researcher can state that the observed effect was caused by a specific treatment; exists when the research design accurately identifies causal relationships.
- Experimental design reliability
- The degree to which the research design and its procedures can be replicated and achieve similar conclusions about hypothesized relationships.
- Extraneous variables
- All variables other than the independent variables that affect the responses of the test subjects. If left uncontrolled, these variables can have a confounding impact on the dependent variable measures that could weaken or invalidate the results of an experiment.
- Independent variable
- An attribute of an object whose measurement values are directly manipulated by the researcher, also referred to as a predictor or treatment variable. This type of variable is assumed to be a causal factor in a functional relationship with a dependent variable.
- Dependent variable
- A singular observable attribute that is the measured outcome derived from manipulating the independent variable(s).
- Control variables
- Extraneous variables that the researcher is able to account for according to their systematic variation (or impact) on the functional relationship between the independent and dependent variables included in the experiment.
- Area sampling
- A form of cluster sampling where clusters are formed by geographic designations such as cities, subdivisions, and blocks. Any geographic unit with boundaries can be used, with one-step or two-step approaches.
- Cluster sampling
- A method of probability sampling where the sampling units are selected in groups (or clusters) rather than individually. Once the cluster has been identified, the elements to be sampled are drawn by simple random sampling or all of the units may be included in the sample.
- Convenience sampling
- A method of nonprobability sampling where the samples are drawn based on the convenience of the researcher or interviewer; also referred to as accidental sampling. Convenience sampling is often used in the early stages of research because it allows a large number of respondents to be interviewed in a short period of time.
- Judgment sampling
- A nonprobability sampling design that selects participants for a sample based on an experienced individual's belief that the participants will meet the requirements of the research study.
- Nonprobability sampling
- Sampling designs in which the probability of selection of each sampling unit is not known. The selection of sampling units is based on the judgment or knowledge of the researcher and may or may not be representative of the target population.
- Probability sampling
- Sampling designs in which each sampling unit in the sampling frame (operational population) has a known, nonzero probability of being selected for the sample.
- Quota sampling
- The selection of participants based on specific quotas regarding characteristics such as age, race, gender, income, or specific behaviors. Quotas are usually determined based on specific research objectives.
- Snowball sampling
- A nonprobability sampling method that involves the practice of identifying a set of initial prospective respondents who can, in turn, help in identifying additional people to be included in the study.
- Sampling
- The process of selecting a relatively small number of elements from a larger defined group of elements so that the information gathered from the smaller group allows one to make judgments about that larger group of elements.
- Simple random sampling (SRS)
- A method of probability sampling in whichEvery sampling unit has an equal, nonzero chance of being selected. Results generated by using simple random sampling can be projected to the target population with a prespecified margin of error.
- Stratified random sampling (STRS)
- A method of probability sampling in which the population is divided into different subgroups (called strata)and samples are selected from each stratum.
- Sampling error
- Any type of bias in a survey study that is attributable to mistakes made in either the selection process of prospective sampling units or determining the size of a sample required to ensure its representativeness of the larger defined target population.
- Construct development
- An integrative process of activities undertaken by researchers to enhance understanding of what specific data should be collected for solving defined research problems.
- Constructs
- Hypothetical variables comprised of a set of component responses or behaviors that are thought to be related.
- Systematic random sampling (SYMRS)
- A method of probability sampling that is similar to simple random sampling but requires that the defined target population be naturally ordered in some way.
- Operationalization
- The process of precisely delineating how a construct is to be measured. The variables are specified in such a manner as to be potentially observable or manipulable.
- Discriminatory power of the scale
- The scale's ability to significantly differentiate between the categorical scale responses (or points).
- Internal consistency reliability
- The extent to which the items of a scale represent the same domain of content and are highly correlated both withEach other and summated scale scores. It represents the degree to which the components are related to the same overall construct domain.
- Scale reliability
- The extent to which a scale can produce the same measurement results in repeated trials.
- Split-half test
- A technique used to evaluate the internal consistency reliability of scale measurements that have multiple attribute components.
- Equivalent form reliability
- A method of assessing the reliability associated with a scale measurement; the researcher creates two basically similar yet different scale measurements for the given construct and administers both forms to either the same sample of respondents or two samples of respondents from the same target population.
- State-of-behavior data
- Raw responses that represent an individual's or rrganization's current observable actions or reactions or recorded past actions/reactions.
- State-of-being data
- Raw responses that are pertinent to the physical and/or demographic or socioeconomic characteristics of individuals, objects, or organizations.
- State-of-intention data
- Raw responses that represent an individual's or organization's expressed plans of future actions/reactions.
- State-of-mind data
- Raw responses that represent the mental attributes of individuals which are not directly observable nor available through some type of external source.
- Affective component
- That part of an attitude which represents the person's feelings toward the given object, idea, or set of information.
- Cognitive component
- That part of an attitude which represents the person's beliefs, perceptions, preferences, experiences, and knowledge about a given object, idea, or set of information.
- Conative component
- That part of an attitude which refers to the person's behavioral response or specific action/reaction toward the given object, idea, or set of information; it tends to be the observable outcome driven by the interaction of a person's cognitive and affective components toward the object or behavior.
- Constant sums rating scale
- A scale format that requires the respondents to allocate a given number of points, usually 100, among several attributes or features based on their importance to the individual; this format requires a person to value each separate feature relative to all the other listed features.
- Likert scale
- A special rating scale format that asks respondents to indicate the extent to which they agree or disagree with a series of mental belief or behavioral belief statements about a given object; it is a cognitive-based scale measurement.
- Paired comparison rating scale
- A scale format in which preselected groups of product characteristics or features are paired against one another and the respondents are asked to select which feature in each pairing is more important to them.
- Rank-order rating scale
- A scale point format that allows respondents to compare their responses to each other by indicating their first preference, then their second preference, then their third preference, etc., until all the desired responses are placed in some type of rank order, either highest to lowest or lowest to highest.
- Semantic differential scale
- A special type of symmetrical rating scale that uses sets of bipolar adjectives and/or adverbs to describe some type of positive and negative poles of an assumed continuum; it is used to capture respondents' cognitive and affective components of specified factors and create perceptual image profiles relating to a given object or behavior.
- Staple scales
- Considered a modified version of the semantic differential scale; they symmetrically center the scale point domain within a set of plus (+) and minus (-) descriptors.
- Anonymity
- The assurance that the prospective respondent's name or any identifiable designation will not be associated with his or her responses.
- Questionnaire
- A set of questions and scales designed to generate enough raw data for accomplishing the information requirements that underlie the research objectives.
- Structured questions
- Questions that require the respondent to make a choice among a limited number of prelisted responses or scale points; they require less thought and effort on the part of the respondent; also referred to as closed-ended questions.
- Unstructured questions
- Question/scale formats that require respondents to reply in their own words; this format requires more thinking and effort on the part of respondents in order to express their answers; also called open-ended questions.
- Interviewer's instructions
- The vehicle for training the interviewer on how to select prospective respondents, screen them for eligibility, and conduct the actual interview.
- Screening forms
- A set of preliminary questions that are used to determine the eligibility of a prospective respondent for inclusion in the survey.
- Test-retest approach
- A procedure used to assess the reliability of a scale measurement; it involves repeating the administration of the scale measurement to either the sample set of sampled respondents at two different times or two different samples of respondents from the same defined target population under as nearly the same conditions as possible.
- Coefficient alpha
- A widely used measurement of the internal consistency of a multi-item scale in which the average of all possible split-half coefficients is taken.
- Coding
- The activities of grouping and assigning values to various responses from a survey instrument.
- Cross-tabulation
- The process of simultaneously treating (or counting) two or more variables in the study. This process categorizes the number of respondents who have responded to two or more questions consecutively.
- Data entry
- The direct inputting of the coded data into some specified software package that will ultimately allow the research analyst to manipulate and transform the raw data into data structures.
- Data validation
- A specific control process that the researcher undertakes to ensure that his or her representatives collected the data as required. The process is normally one of recontacting about 20 percent of the selected respondent group to determine that they did participate in the study.
- Analysis of variance (ANOVA)
- A statistical technique that determines whether two or more means are statistically different from each other.
- Hypothesis
- A yet-unproven proposition or possible solution to a decision problem that can be empirically tested using data that are collected through the research process; it is developed in order to explain phenomena or a relationship between two or more constructs or variables.
- Null hypothesis
- A statement of the perceived existing relationship between either two questions, dimensions, or subgroupings of attributes as being not significantly different; it asserts the status quo condition, and any change from what has been thought to be true is due to random sampling error.
- Type I error
- The error made by rejecting the null hypothesis when it is true; represents the probability of alpha error.
- Type II error
- The error of failing to reject the null hypothesis when the alternative hypothesis is true; represents the probability of beta error.
- Dummy variables
- Artificial variables introduced into a regressionEquation to represent the categories of a nominally scaled variable (such as sex or marital status). There will be one dummy variable for each of the nominal categories of the independent variable and the values will typically be 0 and 1, depending on whether the variable value is present or absent for a particular respondent (e.g., male or female).
- Multicollinearity
- A situation in which several independent variables are highly correlated with each other. This characteristic can result in difficulty in estimating separate or independent regression coefficients for the correlated variables.
- Multiple regression
- A statistical technique which analyzes the linear relationships between a dependent variable and multiple independent variables by estimating coefficients for the equation for a straight line.
- Coefficient of determination (r2)
- A statistical value (or number) that measures the proportion of variation in one variable accounted for by another variable; the r2 measure can be thought of as a percentage and varies from .00 to 1.00.
- t-test
- A hypothesis test procedure that uses the t-distribution: t-tests are used when the sample size of subjects is small (generally less than 30) and the standard deviation is unknown.
- z-test
- A hypothesis test procedure that uses the z distribution; z-tests are used when the sample size is larger than 30 subjects and the standard deviation is unknown.
- Multivariate techniques
- Statistical procedures used when there are two or more measurements on each element and the variables are analyzed simultaneously. Multivariate techniques are concerned with the simultaneous relationships among two or more phenomena.
- Cluster analysis
- A multivariate interdependence technique whose primary objective is to classify objects into relatively homogeneous groups based on the set of variables considered.
- Discriminant analysis
- A multivariate technique for analyzing marketing research data when the dependent variable is categorical and the independent variables are interval.
- Discriminant score
- In discriminant analysis, this represents the score of each respondent on the discriminant function.
- Factor analysis
- A class of statistical procedures primarily used for data reduction and summarization.
- Factor loadings
- Simple correlations between the variables and the factors.
- Perceptual map
- A graphic representation of respondents' beliefs about the relationship between objects with respect to two or more dimensions (usually attributes or features of the objects).
- Appendix
- A section at the end of the final research report used to house complex, detailed, or technical information.
- Believability
- The quality achieved by building a final report that is based on clear, logical thinking, precise expression, and accurate presentation.
- Credibility
- The quality that comes about by developing a final report that is accurate, believable, and professionally organized.
- Executive summary
- The part of the final research report that illustrates the major points of the report in a manner complete enough to provide a true representation of the entire document.
- Hypertext markup language (HTML)
- The language used to create Web pages for communicating the research results as well as other information on the Internet.
- Limitations
- A section of the final research report in which allExtraneous events that place certain restrictions on the report are fully communicated.
- Powerpoint
- A specific software package used to develop slides for electronic presentation of the research results.
- Psychogalvanometer
- A device that measures a subject's involuntary changes in the electronic resistance of his or her skin, referred to as galvanic skin response (GVR).
- Pupilometer
- A device that observes and records changes in the diameter of a subject's pupils. Changes are interpreted as the result of unobservable cognitive activity.
- Voice pitch analyzer
- A computerized system that measures emotional responses by changes in the subject's voice.
- Eye tracking monitor
- A device that observes and records a person's unconscious eye movements.
- Level of significance
- The amount of risk regarding the accuracy of the test that the researcher is willing to accept.
- Regression coefficient
- The statistical measure of the slope coefficient (b) of an independent variable (x) that tells how much the researcher can expect the dependent variable (y) to change, given a unit change in (x).
- Unexplained variance
- In multivariate models, it is the amount of variation in the dependent construct that cannot be accounted for by the combination of independent variables.
- Chi-square (X2) Analysis
- The standardized measurement of the observed difference squared between two frequency distributions that allows for the investigation of statistical significance in analyzing frequency distribution data structures.
- Explained variance
- In multivariate methods, it is the amount of variation in the dependent construct that can be accounted for by the combination of independent variables.
- Open-ended questions
- Open-ended questions should begin with words such as "why" and "how" or phrases such as "What do you think about . . ." Open-ended questions should lead students to think analytically and critically. Ultimately, a good open-ended question should stir discussion and debate in the classroom sparking enthusiasm and energy in your students.
- Closed ended questions
- Closed ended questions are those questions, which can be answered finitely by either “yes†or “no.†Also known as dichotomous or saturated type questions. Closed-ended questions can include presuming, probing, or leading questions. By definition, these questions are restrictive and can be answered in a few words.
- F-test
- The test used to statistically evaluate the difference between the group means in ANOVA.
- Five common problems associated with the marketing research report
-
(1) lack of data interpretation,
(2) unnecessary use of multivariate statistics,
(3) emphasis on packaging rather than quality,
(4) lack of relevance, and
(5) placing too much emphasis on a few statistical outcomes. - Four primary objectives of the marketing research report
-
(1) To effectively communicate the findings of the marketing research project
(2) To provide interpretations of those findings in the form of sound and logical recommendations
(3) To illustrate the credibility of the research project
(4) To serve as a future reference document for strategic or tactical decisions.