Market Research Chapter 24: Multivariate Statistical Analysis

Front 1

What is multivariate data analysis?

Back 1

Research that involves three or more variables, or that is concerned with underlying dimensions among multiple variables, will involve multivariate statistical analysis.

Front 2

What marketing problems involve multivariate data analysis?

Back 2

Most psychographic research and most research that seeks to identify variable market segments

Front 3

Variate

Back 3

A mathamatical way in which a set of variables can be represented with one question.

A linear combination of variables, each contributing to the overall meaning of the variate based upon an empirically derived weight

Front 4

What are the two very basic classifications of multivariate data analysis procedures?

Back 4

Dependence methods and interdependence methods

Front 5

Dependence Techniques

Back 5

Explain or predict one or more dependent variables. Needed when hypotheses involve distinction between independent and dependent variables

Front 6

Name some examples of dependence techniques?

Back 6

Multiple regression analysis
Multiple discriminant analysis
Multivariate analysis of variance
Structural Equation of modeling

Front 7

Interdependence Techniques

Back 7

Give meaning to a set of variables or seek to group things together.

Used when reaserchers examine questions that do not distinguish between independent and dependent variables.

Front 8

What are some examples of interdependence Techniques?

Back 8

Factor Analysis
Cluster Analysis
Multidimensional scaling

Front 9

True/False: The nature of the measurement scales will determine which multivariate technique is appropriate for the data

Back 9

True

Front 10

Selection of a multivariate technique requires consideration of what?

Back 10

Selection of a multivariate technique requires consideration of the types of measures used for both independent and dependent sets of variables

Front 11

Nominal and ordinal scales are metric or nonmetric?

Back 11

Nonmetric

Front 12

Interval and ratio scales are metric or nonmetric?

Back 12

metric

Front 13

General Linear Modeling

Back 13

A way of explaining and predicting a dependent variable based on fluctuations (variation) from its mean due to changes in independent variables

Front 14

Multivariate Analysis of Variance (MANOVA)

Back 14

A multivariate technique that predicts multiple continuous dependent variables with multiple categorical independent variables

Front 15

What are the steps involved in interpreting N-way (Univariate) ANOVA?

Back 15

1)Examine overall model F-Test Result. If significant, proceed

2)Examine invididual F-Tests for invididual Variables

3)For each significant categorical independent variable, interpret the effect by examining the group means

4)For each significant, continous covariate, interpret the parameter estimate (b)

5)For each significant interaction, interpret the means for each combination

Front 16

Discriminant analysis

Back 16

A statistical technqiue for predicting the probablity that an object will belong in one of two or more mutually exclusive categories (dependent variable) based on several independent variables

Front 17

Factor analysis

Back 17

Statistically indentfies a reduced number of factors from a larger number of measured variables

Front 18

What are the two types of factor analysis?

Back 18

Explaratory factor analysis (EFA)

Confirmatory factor analysis (CFA)

Front 19

Exploratory Factor Analysis

Back 19

Performed when the researcher is uncertain about how many factors may exist among a set of variables

Front 20

Confirmatory Factor Analysis

Back 20

Perfomred when the researcher has strong theoretical expecations about the factor structure bfore performing the analysis

Front 21

Factor Loading

Back 21

Indicates how strongly a measured variable is correlated with a factor

Front 22

Data Reduction Technique

Back 22

Approaches that summarize the informaiton from many variables into a reduced set of variates formed as linear combinations of measured variables

Front 23

The rule of parsimony:

Back 23

an explanation involving fewer components is better than one involving many more

Front 24

Creating Composite Scales with Factor Results

Back 24

When a clear pattern of loading exists, the researcher may take a simpler approach by summing the variables with high loadings and creating a summated scale

Front 25

What does low loadings suggest about a variable?

Back 25

Very few loadings suggest a variable does not contirbute much to the factor

Front 26

How is the reliability of each summated scale tested?

Back 26

The reliability of each summated scale is tested by computing a coefficient alpha estimate

Front 27

Communality

Back 27

A measure of the percentage of a variable's variation that is explained by the factors

Front 28

What does a relatively high communality indicate?

Back 28

A relatively high communality indicates that a variable has much in common with the other variables taken as a group

Front 29

Eigenvalues

Back 29

a measure of how much variance is explained by each factor

Front 30

Factor Rotation

Back 30

a mathamatical way of simplifying factor analysis results to better identify which variables "load on" which factors

Front 31

What is the most common procedure for factor rotation?

Back 31

Varimax Rotation

Front 32

Cluster Analysis

Back 32

A multivariate approach for grouping observations based on similarity among measured variables

Front 33

Multidimensional scaling

Back 33

Measures objects in multidimensional space on the basis of similarity of objects

Front 34

Structural Equations Modeling (SEM)

Back 34

Combines an inter and dependence technique to allow testing theory by providing an omnibus assessment of fit centered around a goodness of fit test

Front 35

Partial Least Squares (PLS)

Back 35

Combines a factor analytic and regression approach to provide path estimates to a model but not a goodness of fit. Usefull when there is only a small amount of data existing or when the measurment quality is not particular strong.