Practical stats, part 2: understanding and reporting regression analyses and multivariate ANOVA models
Regression analyses and models constructed by multivariate analysis of variance (ANOVA) are often appropriate for investigating relationships between more than two variables in cohort or group comparison studies. The results of these procedures can seem esoteric if we have no personal research experience with the procedures. Trying to learn about them from books or the Internet can also generate a substantial amount of frustration because most texts are technically written with assumptions that readers have at least some background in statistics, and that they are reading because they want to learn how to use these procedures for analyzing their own data immediately.
This workshop is offered without these assumptions. It is aimed at the lay audience with a view to helping them to understand the purposes and interpretation of these procedures and principles for their presentation.
Developer and facilitator: Darko Hren
Purpose: To help manuscript editors and translators understand the logic and purposes of regression analysis and multivariate ANOVA models. To explain and demonstrate principles for interpretation and presentation of the results of these procedures.
Description: The workshop will be structured as a mixture of interactive lecturing and practical work. Each concept will first be explained and examples discussed. Practical exercises will be given along the way to put the knowledge into practice.
Structure: The workshop will consist of two modules:
- Regression analysis: What is it, when is it used and why? What needs to be reported and what does the report mean? Logistic regression analysis and linear regression analysis: what is the difference? Practice in reporting regression analysis.
- Multivariate ANOVA models: Basic two-way between-groups ANOVA; mixed between-within subjects ANOVA (ANCOVA). When and why are these analyses used? What needs to be reported and what does the report mean? Practice in reporting multivariate ANOVA and ANCOVA models.
Outcome skills: After the workshop, participants will know when it is appropriate to use different regression analysis and multivariate ANOVA models. They will know how to interpret the results of these procedures and the best ways of presenting them in the manuscript.
Pre-meeting information: This workshop will be taught with a presumption that participants know and understand different scales of measurement and study designs through personal work experience or such as were discussed in the METM09 workshop Study designs in medical research: reporting structures and roles in knowledge-building, or the METM08 knowledge update Study designs in medical research: concepts and terminology.
To acquire or review a basic understanding (enough to follow the workshop) of scales of measurement, read http://allpsych.com/researchmethods/measurementscales.html or http://en.wikipedia.org/wiki/Level_of_measurement.
For a more detailed explanation, read http://www.wadsworth.com/psychology_d/templates/student_resources/workshops/stat_workshp/scale/scale_01.html.
To learn more about study designs start from http://en.wikipedia.org/wiki/Study_design and then follow the links to the different types of studies: randomized controlled trial; cohort study, case-control study, cross-sectional study.
About the facilitator: Darko Hren is an assistant professor at the University of Split Faculty of Humanities and Social Sciences, where he teaches social psychology, educational psychology, and basics of scientific methodology and communication. He was a statistical editor for the Croatian Medical Journal from 2002 until 2010. He gave earlier versions of his popular workshops on statistics at METM06, METM07, METM08, METM10, METM12 and METM14, and led the local organizing team for METM08 in Split.