A Practitioner’s Guide to the Galaxy…of Statistical Methods:
Primer on Developments from the Last Two Decades and a Look Ahead
Presenters: Dan J. Putka, HumRRO
Larry J. Williams, Wayne State University
Coordinator: Robert E. Gibby, Procter & Gamble
Over the past 2 decades there have been numerous developments in statistical methodology that can be of great use to I-O psychologists in practice and academe. Historically, our formal graduate curricula have been limited in terms of their ability to incorporate new and emerging statistical methods (e.g., Aiken, West, Sechrest, & Reno, 1990; Aiken, West, Millsap, 2008). As a result, there is non-trivial burden put on I-O psychologists to remain cognizant of such developments in statistical methodology and understand how those developments may be beneficial to their work.
This workshop aims to provide attendees with a basic primer on developments in statistical methodology from the last 2 decades. Our focus will not be on providing a detailed tutorial on how to use various methods, nor will it be overly quantitative in nature. Rather our focus will be on breadth of coverage, conveying developments in plain English, and offering attendees a clear schema for understanding recent developments from the perspective of traditional methods covered in standard I-O curricula. We plan to not only address developments involving methods that have become more frequently used in I-O work over the last 2 decades (e.g., SEM, multilevel modeling, missing data techniques), but also methods emerging from other disciplines that may have value for our field (e.g., mixture modeling - latent class procedures, Bayesian model averaging, random forests), but have yet to be widely disseminated. The workshop is targeted towards practitioners looking for a refresher on the quantitative training they had in graduate school.
This workshop is designed to help attendees:
1. Describe basic relations between new and emerging statistical methods and traditional statistical methods.
2. Describe the types of questions (both practical and scientific) modern methods can be leveraged to answer.
3. List the advantages that modern methods can provide over traditional methods.
4. Select the scope/type of data needed to capitalize on modern methods.
5. Apply best-practice resources to implement modern methods (e.g., existing software, web tutorials).
Dr. Dan J. Putka is a principal staff scientist at HumRRO. His work commonly involves the development and evaluation of selection systems and the study of personnel issues and policy for the U.S. military. Dan has extensive experience modeling large complex sets of archival organizational data for the purpose of informing practical problems (e.g., identifying precursors of turnover, counterproductive behavior, and job attitudes). In addition to his service to clients, Dan has remained an active member of the scientific community, publishing his research in book chapters, as well as journals such as Organizational Research Methods, the Journal of Applied Psychology, and SIOP’s I-O Perspectives, and serving on the editorial board of Organizational Research Methods. His published research has addressed myriad issues, including reliability estimation, modeling of error in ratings, content-oriented validation, differential prediction, and multimodel inference. In recognition of his contributions to military research, Dan received the 2008 Arthur W. Melton Early Career Achievement Award from Division 19 of APA. He received his PhD in industrial-organizational psychology, with a specialization in quantitative methods, from Ohio University.
Dr. Larry J. Williams joined the faculty of Wayne State University in January 2010. Dr. Williams served as the founding editor of Organizational Research Methods (ORM), and he also has served as chairperson for the Research Methods Division (RMD) of the Academy of Management. Professor Williams established and currently serves as director of the Center for the Advancement of Research Methods and Analysis (CARMA). In 2004 Dr. Williams was recognized by the Southern Management Association as an author of 2 of the 6 most highly cited articles in the 30 year history of the Journal of Management, and he was also elected to be a member of the Society for Organizational Behavior. In 2005, Dr. Williams was selected to be the recipient of the 2005 Distinguished Career Contributions Award by the Academy of Management’s Research Methods Division. In 2008, Professor Williams was recognized as one of the 150 most-cited authors in the field of management (1981-2004) in an article published in the Journal of Management. He was elected a Fellow of the Society for Industrial and Organizational Psychology in 2010.