Studying Moderating Effects in Management Research: A Review of Problems in Hypothesis Formulation, Data Analysis and Results Interpretations

Moderators that specify contingencies and boundary conditions play a central role in management research, because moderators serve to refine theory for the explanation of complex human behavior.  While there are multiple reviews on the problems and recommended ways to study moderators in empirical research, many new problems have arisen because of the increasing complexity in management studies.  Moreover, previous reviews have focused on data analytical procedures but paid inadequate attention to hypotheses formulation.  Hence, this seminar will present ten problems identified in recently published articles in Journal of Organizational Behavior, including: (1) Main effects were inappropriately hypothesized before hypothesizing moderating effects; (2) Natures of the moderating effect were incorrectly specified; (3) Inadequate information on interaction term was provided which made accumulation of findings on moderating effects infeasible; (4) High/low levels of the independent variables and moderators were not adequately represented because of possible range restrictions; (5) Measurement errors were ignored in analyses; (6) Multiple moderating effects were incorrectly tested in separate models; (7) Interaction terms were inappropriately standardized; (8) Main effects were inappropriately interpreted when moderating effects were statistically significant; (9) Levels of moderator in  graphical representation and simple slopes analysis were arbitrarily selected; and (10) Testing significant difference in slopes for low and high levels of moderator.  Best-practice recommendations and exemplary studies for each of the problems will be provided.

Professor Gordon Cheun​

Gordon joined the University of Auckland Business School in January 2016 as Professor of Organisational Behaviour. He served as the Associate Vice President and Professor in the Department of Management at The Chinese University of Hong Kong (CUHK) before he joined the Business School. He obtained his BBA from CUHK and his PhD in management from Virginia Tech. He is well recognised internationally as an expert in structural equation modelling, especially in measurement equivalence/invariance, analysis of dyadic data, as well as estimation of moderating and mediating effects in complex latent variable models. He has published more than 20 articles in research methodologies, which have been cited over 9,300 times. He has twice received the Sage Best Paper Award from the Research Methods Division of the Academy of Management (2000 and 2009) and in 2008 the Best Published Paper Award in Organisational Research Methods.

He served as the Division Chair of the Research Methods Division of the Academy of Management in 2006/07 and as the Chair of the Research Methods Special Track at the Academy of International Business 2011 Conference at Nagoya, Japan. He is currently a member of the Editorial Board of Organisational Research Methods, a member of the International Scientific Advisory Panel for the Behavioral Sciences Institute (BSI) at Singapore Management University and of the International Advisory Board, Center for the Advancement of Research Methods and Analysis (CARMA) in USA.

Gordon is an enthusiastic teacher and has taught undergraduate level international business, management, and organisational behaviour, and graduate level research methods and structural equation modelling courses. He has received many teaching awards, including the Vice-Chancellor Exemplary Teacher Award (2000) and the Faculty of Business Administration Outstanding Teacher Award (2002/03) at CUHK.

About Academic Seminars

Our academic seminars are a forum for our academic staff to collaborate, share and discuss relevant research and trends with their peers and broader academic community.


Joyce Ackroyd Building (37), room 430