Vous êtes sur la page 1sur 1


Psychometrics is concerned with the theory and practice of mental and psychological measurement
and primarily dealing with the objective measurement of skills and knowledge, abilities and
attitudes and it also, is widely applied in educational assessment to measure abilities domains in
educational achievements. A discourse of psychometrics stresses the alteration of human behaviors
to facts and prioritizes high-stakes, point-in-time sampling, and standardization. In the
psychometric model, reliability is the accuracy of the transformation from real-world phenomenon
to a number, and validity is the stability of that numerical representation. Turning from conceptual
concerns to practices of assessment, psychometric discourse provided three key imperatives: to
identify sub-components of competence; to standardize assessments and take multiple samples.

The established of practices to convert human phenomena into numbers and it must be
acknowledged that physicians often work in teams and systems, rendering it impossible to attribute
quality of practice to a single person in assessment particularly links to patient outcomes propelled
the adoption of a number of new practices, including the pre-operative briefing. The need to
identify subcomponents that make up competence has led to efforts around the world to articulate
competence frameworks: for example, the CanMEDS roles, the ACGME competence framework,
and Tomorrows Doctor. Diagnostic, contextual, or inter-personal variables that might be part of
the authentic variability of real practices. Today there is a fundamental re-examination of the role
of subjectivity and an effort to rehabilitate subjective judgment.

Assessment research flourished with new attention given to the connection of expertise and
scoring; the psychometric properties of different kinds of assessment tools extrapolating from a
number of studies, they showed that MCQs, essays, oral examinations, OSCEs, reliability and
validity to ensure and improve the quality and fairness of assessment is invaluable.
An observation revealed to shift the focus of assessment from measurement of the outcome of an
educational activity to assessment as an integral part of education.
The dominance of psychometric standards as criteria for assessment quality filtered down from
large testing organizations.
Perhaps the most important is that it requires a numerical and reductionist approach, and that
aspects such as fairness, defensibility and credibility are by necessity mainly translated into
reliability and construct validity.

It is concluded that the strict psychometric model is no longer sustainable and we would like to
make a plea for a major revision of the statistical concepts and approaches to assessment.
Several concepts, approaches, statistics models concerning psychometrics are central to
discourse and some limitations of overemphasis on this paradigm are evident which have
significant consequences i.e clinical practice and expensive, high-stakes final examinations are
inadequate suited for assessment in workplaces.