Applied Statistical Procedures

Course dates

Wellington: 20–24 November 2017
Auckland: 19–23 February 2018

Instructor

Dr Gordon Emmerson

Prerequisites

While this is an intermediate course, research language and concepts will be taught to encompass both qualitative researchers with little knowledge of quantitative research and quantitative Researchers who wish to review or broaden their understanding of the range of Research Procedures and how to use them. The level of this course falls between courses such as Introduction to Statistics and Structural Equation Models using Amos. Applied Statistical Procedures covers a range of the most commonly used statistical procedures from Chi-Square to Factor Analysis. It further defines research methodologies that so participants can match their research design with their research needs, i.e. when and how to determine causal relationships, and how to evaluate and report attitudes and behaviours. The course is taught from an applied perspective with many examples, and questions are encouraged. SPSS will be used to practise procedures, but no prior knowledge is required.
 
You will be exposed to a variety of research scenarios and to the logic of statistical procedure selection and application. The target audience ranges from qualitative researchers wanting to gain quantitative skills, to quantitative researchers wanting to broaden their understanding across procedures, or to become more comfortable with covariance prior to taking on the likes of Structural Equation Modelling.

Course outline

On completing this course you should be able to read and understand literature where these procedures are reported, select the appropriate statistical procedure for research, run the procedure, and report the results from an informed base of understanding.
 
Day 1: The context of quantitative research in relation to qualitative research. The language of quantitative research, and the required fundamentals of SPSS.
 
Day 2: Reliability, Correlations, Controlling for Confounding Variables, Chi Squares, and T-tests.
 
Day 3: ANOVA, ANCOVA, Factoral ANOVA, MANOVA and Non-Parametric Tests.
 
Day 4: Simple Regression, Multiple Regression, Discriminate Analysis and Factor Analysis.
 
Day 5: Testing Normality, Data Transformations, Validity, Reporting and Ethics, individual sessions.
 
The procedures that will be covered will include:

  • Frequency based statistics of Chi-Square Goodness of Fit and Chi-Square Test of Association.
  • Parametric test of difference statistics of T-tests, ANOVA, ANCOVA, MANOVA, and MANCOVA. Factorial analysis with multiple independent variables will also be covered along with Repeated Measures ANOVA.
  • Non-Parametric test of difference statistics of Mann-Whitney, Wilcoxon, Friedman's Analysis of Variance, and Kruskal Wallis.
  • Statistics to predict and to explain variance of Simple Regression, Multiple Regression, Discriminant Analysis and Multiple Discriminant Analysis.
  • Data reduction techniques of Factor Analysis.
  • Power, data and statistics that are most powerful, and techniques for increasing statistical power.
  • How to determine the best procedure for the demands of the research.
  • Data transformation to increase power and allow parametric procedures to be employed when data can be appropriately adjusted.
  • Important interplays between effect size and significance.
  • Integration of statistical results into reports.

References

Aron A, Coups E, Aron EN (2013). Statistics for the Behavioral and Social Sciences: Pearson New International Edition: A Brief Course. Pearson Higher Education.

Hair JF (2010). Multivariate data analysis. Pearson College Division.

  

Register for the course