The EFPA Test-Review Model: When Good Intentions Meet a Methodological Thought Disorder

The European Federation of Psychologists’ Associations (EFPA) has issued sets of test standards and guidelines for psychometric test reviews without any attempt to address the critical content of many substantive publications by measurement experts such as Joel Michell. For example, he has argued th...

Full description

Bibliographic Details
Main Author: Paul Barrett
Format: Article
Language:English
Published: MDPI AG 2017-12-01
Series:Behavioral Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-328X/8/1/5
Description
Summary:The European Federation of Psychologists’ Associations (EFPA) has issued sets of test standards and guidelines for psychometric test reviews without any attempt to address the critical content of many substantive publications by measurement experts such as Joel Michell. For example, he has argued that the psychometric test-theory which underpins classical and modern IRT psychometrics is “pathological”, with the entire profession of psychometricians suffering from a methodological thought disorder. With the advent of new kinds of assessment now being created by the “Next Generation” of psychologists which no longer conform to the item-based, statistical test theory generated last century, a new framework is set out for constructing evidence-bases suitable for these “Next Generation” of assessments, which avoids the illusory beliefs of equal-interval or quantitatively structured psychological attributes. Finally, with no systematic or substantive refutations of the logic, axioms, and evidence set out by Michell and others; it is concluded psychologists and their professional associations remain in denial. As with the eventual demise of a similar attempt to maintain the status quo of professional beliefs within forensic clinical psychology and psychiatry during the last century, those following certain EFPA guidelines might now find themselves required to justify their professional beliefs in legal rather than academic environments.
ISSN:2076-328X