Academic journal article The Journal of Parapsychology

Is the Methodological Revolution in Psychology over or Just Beginning?/ Ist Die Methodologische Revolution in der Psychologie Vorueber Oder Beginnt Sie Gerade Erst?/ ?Ha Terminado la Revolucion Metodologica En la Psicologia O Apenas Principia?/ Est-Ce la Fin Ou le Commencement De la Revolution Methodologique En Psychologie?

Academic journal article The Journal of Parapsychology

Is the Methodological Revolution in Psychology over or Just Beginning?/ Ist Die Methodologische Revolution in der Psychologie Vorueber Oder Beginnt Sie Gerade Erst?/ ?Ha Terminado la Revolucion Metodologica En la Psicologia O Apenas Principia?/ Est-Ce la Fin Ou le Commencement De la Revolution Methodologique En Psychologie?

Article excerpt

Parapsychologists have often argued that their research methods have been as good as the standard methods used in psychological research (Carderia, Palmer, & Marcusson-Clavertz, 2015, p. 1). With the publication of Bern's (2011) paper on "feeling the future," psychological researchers finally paid attention to this point and agreed. However, their conclusion was not that the research provided evidence for psi, but that the methods for psychological research needed to be substantially improved. With considerable publicity and declaration of a methodological revolution (Wagenmakers, 2015), psychological researchers discovered the need for preregistered well-powered confirmatory research that has long been the accepted practice for clinical trials.

However, psychological researchers are still struggling with methodological practices. Several key methodological issues have yet to be recognized and addressed. Psychological and parapsychological researchers can expect additional repetitions of this recent experience. When confronted with the choice between attributing experimental results to psi or to previously overlooked methodological deficiencies, psychological researchers will become advocates for additional methodological improvements.

An alternative strategy is to skip the inefficient cycles of implementing better methodological practices when skeptical psychologists become trapped by a choice between improved methodology and psi. Several methodological factors that psychologists and parapsychologists will eventually have to address can be easily foreseen and are discussed below. Methodological practices that address most of these factors have been implemented as standard procedure in regulated medical research. These practices are described.

Standards for Study Preregistration

Although the value of preregistration for confirmatory studies is now widely recognized, psychologists are still struggling with pitfalls in the registration process. Key practices for confirmatory study registration include that: (a) the registry specifies requirements for the registration information, (b) all key methodological decisions that could affect the study outcome are registered, (c) each registration is independently reviewed for consistency and completeness, (d) the study registration is made irreversibly public before data collection begins, and (e) the registrations can be easily found and accessed when searching for studies on certain topics (Watt & Kennedy, 2015, also see the comments for that online article). The KPU Study Registry (2012) implements all of these practices, but, as yet, study registries for general use by psychologists do not incorporate all of these practices. Parapsychologists who use nonoptimal study registration can expect to have their findings eventually challenged when psychologists recognize the weaknesses of their registration practices.

Dropouts and Incomplete Data

Dropouts and incomplete data can introduce bias into experimental results but are often given little or no attention by researchers. Participants who are doing poorly in an experiment may tend not to complete the experiment or tend to make invalid responses. In general, dropouts and other types of incomplete data cannot be assumed to be independent of the experimental intervention and procedures--and therefore are confounding factors that can potentially cause bias. In addition, ad hoc retrospective decisions about handling incomplete data can create bias. The possibility that an ostensible finding is entirely or partially due to bias from incomplete data needs to be addressed.

In clinical trials the "intent-to-trea" (or "intention-to-treat") principle requires that every participant who was randomized is included in the primary analysis, whether or not the participant complied with the protocol (U.S. Food and Drug Administration, 1998; Gupta, 2011). Intent-to-treat is a fundamental principle guiding study conduct and analysis and is generally considered necessary to avoid bias. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.