1. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-1 Chapter 1 Introduction

2. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-2 LEARNING OBJECTIVES: Upon completing this chapter, you should be able to do the following: 1.Explain what multivariate analysis is and when its application is appropriate. 2.Define and discuss the specific techniques included in multivariate analysis. 3.Determine which multivariate technique is appropriate for a specific research problem. 4.Discuss the nature of measurement scales and their relationship to multivariate techniques. 5.Describe the conceptual and statistical issues inherent in multivariate analyses. Chapter 1 Introduction

3. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-3  What is it? Multivariate Data Analysis = all statistical methods that simultaneously analyze multiple measurements on each individual or object under investigation.  Why use it? Measurement Measurement Explanation & Prediction Explanation & Prediction Hypothesis Testing Hypothesis Testing What is Multivariate Analysis?

4. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-4  The Variate  Measurement Scales Nonmetric Nonmetric Metric Metric  Multivariate Measurement  Measurement Error  Types of Techniques Basic Concepts of Multivariate Analysis

5. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-5  The variate is a linear combination of variables with empirically determined weights.  Weights are determined to best achieve the objective of the specific multivariate technique.  Variate equation: (Y’) = W1 X 1 + W2 X 2 +... + Wn X n  Each respondent has a variate value (Y’).  The Y’ value is a linear combination of the entire set of variables. It is the dependent variable.  Potential Independent Variables: X1 = income X2 = education X3 = family size X4 = ?? The Variate

6. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-6 Types of Data and Measurement Scales Data MetricorQuantitative NonmetricorQualitative NominalScale OrdinalScale IntervalScale RatioScale

7. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-7  Nonmetric Nominal – size of number is not related to the amount of the characteristic being measured Nominal – size of number is not related to the amount of the characteristic being measured Ordinal – larger numbers indicate more (or less) of the characteristic measured, but not how much more (or less). Ordinal – larger numbers indicate more (or less) of the characteristic measured, but not how much more (or less).  Metric Interval – contains ordinal properties, and in addition, there are equal differences between scale points. Interval – contains ordinal properties, and in addition, there are equal differences between scale points. Ratio – contains interval scale properties, and in addition, there is a natural zero point. Ratio – contains interval scale properties, and in addition, there is a natural zero point. NOTE: The level of measurement is critical in determining the appropriate multivariate technique to use! Measurement Scales

8. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-8 All variables have some error. What are the sources of error? All variables have some error. What are the sources of error? Measurement error = distorts observed relationships and makes multivariate techniques less powerful. Measurement error = distorts observed relationships and makes multivariate techniques less powerful. Researchers use summated scales, for which several variables are summed or averaged together to form a composite representation of a concept. Researchers use summated scales, for which several variables are summed or averaged together to form a composite representation of a concept. Measurement Error

9. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-9 In addressing measurement error, researchers evaluate two important characteristics of measurement: Validity = the degree to which a measure accurately represents what it is supposed to. Validity = the degree to which a measure accurately represents what it is supposed to. Reliability = the degree to which the observed variable measures the “true” value and is thus error free. Reliability = the degree to which the observed variable measures the “true” value and is thus error free. Measurement Error

10. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-10  Type I error, or , is the probability of rejecting the null hypothesis when it is true.  Type II error, or , is the probability of failing to reject the null hypothesis when it is false.  Power, or 1- , is the probability of rejecting the null hypothesis when it is false. H 0 true H 0 false Fail to Reject H 0 1-   Type II error Reject H 0  Type I error 1-  Power Statistical Significance and Power

11. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-11  Effect size: the actual magnitude of the effect of interest (e.g., the difference between means or the correlation between variables).  Alpha (  ): as  is set at smaller levels, power decreases. Typically,  =.05.  Sample size: as sample size increases, power increases. With very large sample sizes, even very small effects can be statistically significant, raising the issue of practical significance vs. statistical significance. Power is Determined by Three Factors

12. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-12

13. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-13 Impact of Sample Size on Power

14. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-14 Rules of Thumb 1–1 Statistical Power Analysis Researchers should always design the study to achieve a power level of.80 at the desired significance level. Researchers should always design the study to achieve a power level of.80 at the desired significance level. More stringent significance levels (e.g.,.01 instead of.05) require larger samples to achieve the desired power level. More stringent significance levels (e.g.,.01 instead of.05) require larger samples to achieve the desired power level. Conversely, power can be increased by choosing a less stringent alpha level (e.g.,.10 instead of.05). Conversely, power can be increased by choosing a less stringent alpha level (e.g.,.10 instead of.05). Smaller effect sizes always require larger sample sizes to achieve the desired power. Smaller effect sizes always require larger sample sizes to achieve the desired power. Any increase in power is most likely achieved by increased sample size. Any increase in power is most likely achieved by increased sample size.

15. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-15 Dependence techniques: a variable or set of variables is identified as the dependent variable to be predicted or explained by other variables known as independent variables. Dependence techniques: a variable or set of variables is identified as the dependent variable to be predicted or explained by other variables known as independent variables. oMultiple Regression oMultiple Discriminant Analysis oLogit/Logistic Regression oMultivariate Analysis of Variance (MANOVA) and Covariance oConjoint Analysis oCanonical Correlation oStructural Equations Modeling (SEM) Types of Multivariate Techniques

16. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-16 Interdependence techniques: involve the simultaneous analysis of all variables in the set, without distinction between dependent variables and independent variables. Interdependence techniques: involve the simultaneous analysis of all variables in the set, without distinction between dependent variables and independent variables. oPrincipal Components and Common Factor Analysis oCluster Analysis oMultidimensional Scaling (perceptual mapping) oCorrespondence Analysis Types of Multivariate Techniques

17. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-17 Selecting a Multivariate Technique 1.What type of relationship is being examined – dependence or interdependence? 2.Dependence relationship: How many variables are being predicted? What is the measurement scale of the dependent variable? What is the measurement scale of the dependent variable? What is the measurement scale of the predictor variable? What is the measurement scale of the predictor variable? 3.Interdependence relationship: Are you examining relationships between variables, respondents, or objects?

18. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-18 Two Broad Types of Multivariate Methods: 1.Dependence – analyze dependent and independent variables at the same time. 2.Interdependence – analyze dependent and independent variables separately.

19. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-19 Multiple Regression and Conjoint Discriminant Analysis and Logit MANOVA and Canonical Correlation, Dummy Variables MetricNonmetricMetricNonmetric Metric Nonmetric Factor Analysis Cluster Analysis Nonmetric MDS and Correspon- dence Analysis Selecting the Correct Multivariate Method SEMCFA SeveralDependentVariables OneDependentVariable Metric MDS MultivariateMethods DependenceMethods InterdependenceMethods Multiple Relationships - StructuralEquations

20. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-20 Multiple Regression... a single metric dependent variable is predicted by several metric independent variables. Y X1X1X1X1 X2X2X2X2

21. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-21 Discriminant Analysis Discriminant Analysis What is it? What is it?... single, non-metric (categorical) dependent variable is predicted by several metric independent variables. Why use it? Why use it? Examples of Dependent Variables: Gender – Male vs. Female Gender – Male vs. Female Culture – USA vs. Outside USA Culture – USA vs. Outside USA Purchasers vs. Non-purchasers Purchasers vs. Non-purchasers Member vs. Non-Member Member vs. Non-Member Good, Average and Poor Credit Risk Good, Average and Poor Credit Risk

22. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-22 Logistic Regression A single nonmetric dependent variable is predicted by several metric independent variables. This technique is similar to discriminant analysis, but relies on calculations more like regression. A single nonmetric dependent variable is predicted by several metric independent variables. This technique is similar to discriminant analysis, but relies on calculations more like regression.

23. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-23 MANOVA Several metric dependent variables are predicted by a set of nonmetric (categorical) independent variables.

24. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-24 CANONICAL ANALYSIS Several metric dependent variables are predicted by several metric independent variables.

25. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-25... is used to understand respondents’ preferences for products and services.... is used to understand respondents’ preferences for products and services. In doing this, it determines the importance of both: In doing this, it determines the importance of both: attributes and attributes and levels of attributes levels of attributes... based on a smaller subset of combinations of attributes and levels.... based on a smaller subset of combinations of attributes and levels. CONJOINT ANALYSIS

26. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-26 Typical Applications:  Soft Drinks  Candy Bars  Cereals  Beer  Apartment Buildings; Condos  Solvents; Cleaning Fluids CONJOINT ANALYSIS

27. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-27 Structural Equations Modeling (SEM) Estimates multiple, interrelated dependence relationships based on two components: 1.Measurement Model 2.Structural Model

28. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-28 Exploratory Factor Analysis... analyzes the structure of the interrelationships among a large number of variables to determine a set of common underlying dimensions (factors).... analyzes the structure of the interrelationships among a large number of variables to determine a set of common underlying dimensions (factors).

29. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-29... groups objects (respondents, products, firms, variables, etc.) so that each object is similar to the other objects in the cluster and different from objects in all the other clusters.... groups objects (respondents, products, firms, variables, etc.) so that each object is similar to the other objects in the cluster and different from objects in all the other clusters. Cluster Analysis

30. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-30 Cluster Analysis of “Eating Out” Questions 1.I eat at fast food restaurants at least once a week. 2.I prefer restaurants with the highest quality food. 3.I prefer restaurants that have quick service. 4.I prefer to eat at restaurants that have a nice atmosphere. Objective: Identify groups that maximize ratio of Objective: Identify groups that maximize ratio of between groups variance large between groups variance large within groups variance small within groups variance small = 17 7-point Agree/Disagree Scale

31. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-31 Constructs... Trust Trust Commitment Commitment Cooperation Cooperation Locus of Control Locus of Control Job Satisfaction Job Satisfaction Turnover Turnover High Medium Low OrganizationalCommitment Example: Cluster Analysis... “Polar Extremes” = remove middle group(s)

32. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-32 Multidimensional Scaling... identifies “unrecognized” dimensions that affect purchase behavior based on customer judgments of:... identifies “unrecognized” dimensions that affect purchase behavior based on customer judgments of: similarities or similarities or preferences preferences and transforms these into distances represented as perceptual maps.

33. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-33 Correspondence Analysis... uses non-metric data and evaluates either linear or non-linear relationships in an effort to develop a perceptual map representing the association between objects (firms, products, etc.) and a set of descriptive characteristics of the objects.... uses non-metric data and evaluates either linear or non-linear relationships in an effort to develop a perceptual map representing the association between objects (firms, products, etc.) and a set of descriptive characteristics of the objects.

34. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-34 oEstablish Practical Significance as well as Statistical Significance. oSample Size Affects All Results. oKnow Your Data. oStrive for Model Parsimony. oLook at Your Errors. oValidate Your Results. Guidelines for Multivariate Analysis

35. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-35 Stage 1: Define the Research Problem, Objectives, and Multivariate Technique(s) to be Used Multivariate Technique(s) to be Used Stage 2: Develop the Analysis Plan Stage 3: Evaluate the Assumptions Underlying the Multivariate Technique(s) Multivariate Technique(s) Stage 4: Estimate the Multivariate Model and Assess Overall Model Fit Overall Model Fit Stage 5: Interpret the Variate(s) Stage 6: Validate the Multivariate Model A Structured Approach to Multivariate Model Building:

36. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-36 Variable Description Variable Type Variable Description Variable Type Data Warehouse Classification Variables X1Customer Typenonmetric X2Industry Typenonmetric X3Firm Sizenonmetric X4Regionnonmetric X5Distribution Systemnonmetric Performance Perceptions Variables X6Product Qualitymetric X7E-Commerce Activities/Websitemetric X8Technical Supportmetric X9Complaint Resolutionmetric X10Advertising metric X11Product Linemetric X12Salesforce Imagemetric X13Competitive Pricingmetric X14Warranty & Claimsmetric X15New Productsmetric X16Ordering & Billingmetric X17Price Flexibilitymetric X18Delivery Speedmetric Outcome/Relationship Measures X19Satisfactionmetric X20Likelihood of Recommendationmetric X21Likelihood of Future Purchasemetric X22Current Purchase/Usage Levelmetric X23Consider Strategic Alliance/Partnership in Futurenonmetric Description of HBAT Primary Database Variables

37. Copyright © 2010 Pearson Education, Inc., publishing as Prentice-Hall. 1-37 Multivariate Analysis Learning Checkpoint 1. What is multivariate analysis? 2. Why use multivariate analysis? 3. Why is knowledge of measurement scales important in using multivariate analysis? important in using multivariate analysis? 4. What basic issues need to be examined when using multivariate analysis? when using multivariate analysis? 5. Describe the process for applying multivariate analysis. multivariate analysis.