For about a decade, academic psychology finds itself in a crisis. Replication of the vast majority of research findings fails, the field is plagued by a bewildering methodological sloppiness and several cases of outright fraud surfaced. This book argues that one of the major causes of the crisis in psychology often goes unnoticed and is situated at the level of measurement methods. While the method sections of myriads of research papers claim that the validity of the measurement instruments used is 'acceptable', 'good' or even 'excellent', every thorough analysis of validity and reliability of nomothetic measurement procedures leads up to profound skepticism. At first sight, the use of numbers gave psychology an aura of sophistication and exactness, but on a closer examination, it rather puts psychology at risk to become a pseudo-science. The author illustrates in a very tangible way that a variety of factors not-intended-to-be-measured impacts on measurement outcome and that this renders most types of statistical inference ineffective. He concludes that a reorientation towards single case research, a re-appraisal of narrative and qualitative description and a measurement paradigm centred on quantification of formal characteristics of language might attune psychological methods better to the complex and dynamic nature of its objects and contribute to a true overcoming of the replicability crisis.
