Methodological Workshop: How to study differences in perceptions? – The use of Polynomial Regression and Response Surface Methodology

Welcome to a methodological workshop

How to study differences in perceptions? – The use of Polynomial Regression and Response Surface Methodology

Jeff Edwards, Belk Distinguished Professor of Organizational Behavior, University of North Carolina, USA
Organized by: The Procome research group, Medical Management Centre, Karolinska Institutet
Date: 12th March 2018, 9-17.00
Place: Karolinska Institutet, Stockholm (Solna Campus, the conference room will be announced later)
Registration: malin.pettersson@ki.se, Free of charge.

The workshop gives a hands-on experience on statistics that can be used to study congruence (i.e. similarity, agreement, fit) between two constructs, for instance between ideal and perceived situations or a manager’s and a team’s perceptions. Difference scores have traditionally been used in these types of comparisons, but have several methodological limitations. The workshop presents polynomial regression and response surface methodology, which overcomes the limitations associated with difference scores. These methodologies can be used to examine a wide range of congruence hypotheses. Polynomial regression and response surface methodology is often used for in person-environment fit research, self-other agreement studies in leadership research and perceptual distance studies when comparing leader’s and their subordinates ratings of their work context.

This workshop will review problems with difference scores, introduce polynomial regression and response surface methodology, and illustrate the application of these methods using empirical examples. Participants will be provided with data to analyze during the workshop with the goal of creating a hands-on experience that will clarify and reinforce the methods discussed. Participants should bring a laptop with access to SPSS or STATA.

Biography
Professor Jeff Edwards is the Belk Distinguished Professor of Organizational behavior at the Kenan-Flagner Business School at the University of North Carolina at Chapel Hill. His research focuses on individual and organizational change, person-organization fit, stress and coping in organizations, and strategies that jointly promote employee well-being and corporate competitiveness. He also conducts methodological work concerning the design, measurement and analysis of research in organizations especially related to difference scores, polynomial regression and measurement and construct validation. 

Key references
Edwards, J.R., Alternatives to difference scores: Polynomial regression and response surface methodology. Advances in measurement and data analysis, 2002
Edwards, J.R., The study of congruence in organizational behavior research: Critique and a proposed alternative. Organizational Behavior and Human Decision Processes, 1994. 58(1)

See also: https://www.youtube.com/watch?v=9ejTsYgReeg