2016 Vol. 80(2) 156-168
Editor:
John A. Palmer, Ph.D.
Copyright:
Parapsychology Press
Citation
Kennedy, J. E. (2016). Is the Methodological Revolution in Parapsychology Over or Just Beginning? Journal of Parapsychology, 80, 156-168.
Article
Is the Methodological Revolution in Parapsychology Over or Just Beginning?
J.E. Kennedy
Significant results from parapsychological experiments using standard psychological research methods motivated psychologists to recognize some widespread methodological deficiencies and the need for preregistered well-powered confirmatory research. Psychological researchers have not yet recognized several other common methodological weaknesses that can be expected to cause this cycle to be repeated. When confronted with the choice between psi versus overlooked methodological deficiencies, psychologists will recognize the need for methodological improvements. These overlooked methodological factors include: (a) deficient study registration practices, (b) bias from dropouts and incomplete data, (c) the need for software validation, (d) measures to prevent experimenter fraud, (e) appropriate statistical methods for confirmatory research, (f) failure to consider inferential errors with Bayesian analyses, (g) the weaknesses of retrospective meta-analysis and strengths of prospective meta-analysis, and (h) problems from statistical dependence for the outcome variables in statistical analyses. Psychological and parapsychological researchers can easily avoid this inefficient process of methodological evolution driven by controversies about parapsychological findings. Research practices that address these methodological deficiencies are available and will eventually be recognized as needed for psychological and parapsychological research. Recommended practices for addressing these methodological weaknesses are described.
Keywords:
research methodology, incomplete data, software validation, experimenter fraud, statistical power