.DOCX

Unit 1 Discussion 1- Differences Between Univariate And Multivariate Statistics

Unit 1 Discussion 1- Research Questions and Data

Research Questions and Data

Resources

• Discussion Participation Scoring Guide.

• PhD Dissertation Guidebook: School of Business and Technology.

For this discussion activity, complete the following:

1. Briefly discuss the similarities and differences between univariate and multivariate statistics.

2. A researcher wishes to determine if level of education significantly affects income, cost of home, and family income among postal workers after adjusting for starting salary. Identify the dependent variables, independent variables, and covariates. Then use the decision-making tools on pages 22–23 of your Mertler and Vernatta Advanced and Multivariate Statistical Methods text to identify the most appropriate statistical test to conduct the analyses of the data for the research question.

3. Which decision-making tool are you most comfortable with and why?

pages 22–23

Determining the appropriate statistical technique relies upon the identification of the type of variables (categorical or quantitative) and the number of IVs and DVs, all of which influence the nature of the research questions being posed. This chapter introduced the statistical tests to be presented in the upcoming chapters. The statistical methods are organized under four purposes of research questions: degree of relationship, significance of group differences, prediction of group membership, and structure. Statistical tests that analyze the degree of relationship include bivariate correlation and regression, multiple regression, and path analysis. Research questions addressing degree of relationship all have quantitative variables. Methods that examine the significance of group differences are t test, one-way and factorial ANOVA, one-way and factorial ANCOVA, one-way and factorial MANOVA, and one-way and factorial MANCOVA. Research questions that address group differences have categorical IVs. Statistical tests that predict group membership are discriminant analysis and logistic regression. Research questions that address prediction of group membership have a categorical DV. Statistical tests that address the purpose of structure are factor analysis and principal components. Questions that address structure usually do not distinguish between independent and dependent variables.

Two decision-making tools are provided to assist in identifying which statistical method to utilize—the Table of Statistical Tests and the Decision-Making Tree for Statistical Tests. The Table of Statistical Tests is organized by the type and number of IVs and DVs, while the Decision-Making Tree for Statistical Tests is organized by the purpose of the research question.