|Monday, May 30, 2016|
|8:00 AM–8:50 AM |
|Area: OBM; Domain: Theory|
|Instruction Level: Intermediate|
|CE Instructor: Julie M. Slowiak, Ph.D.|
|Chair: Julie M. Slowiak (University of Minnesota Duluth)|
|BRADLEY E. HUITEMA (Western Michigan University)|
|Dr. Brad Huitema is professor of psychology, a member of the general faculty, and statistical consultant for businesses and several units of Western Michigan University. His 1980 text, The Analysis of Covariance and Alternatives, has been well received in the field. He is a member of the editorial board for the Journal of Applied Behavior Analysis and the Journal of Behavioral Assessment. He is also a referee for several journals including The American Statistician; Behavioral Research Methods, Instruments, and Computers; the Journal of Psychopathology and Behavioral Assessment; Psychological Bulletin and Psychometrika. His research interests include time series analysis, evaluation of preventitive health practices, and single organism and quasi-experimental design. His international travel includes teaching several short courses in Mexico.|
A bevy of current methodological topics relevant to psychological researchers working in diverse areas is presented. Many of them are widely misunderstood. Among these topics are: (1) the recent efforts to ban null hypothesis significance tests (NHST), (2) interpretation problems surrounding both conventional (i.e., Fisher and Neyman-Pearson) and Bayesian statistical inference, (3) the relationship between p-values and replication, (4) unjustified hand wringing regarding effect declines (often claimed to cast doubt on scientific methods in general) that appear to occur in many or most areas of science, (5) the increasing salience of single-case research designs and associated analyses, and (6) the stunning effect of complex black-box statistical methods on judgment. The historical context for several of these topics is presented in order to facilitate an understanding how we have ended up in our current state.
|Target Audience: |
Behavioral researchers with a strong interest in research methodology.
|Learning Objectives: At the conclusion of the presentation, the participant will be able to: (1) describe at least one reason that treatment effects tend to decline; (2) provide an example of a major public health problem solved with a very small n design; (3) identify a study in which a single-case design is very convincing.|