Psychometrics

[Mark]

For a group dedicated to personality there is precious little about
personality theories and their use amongst recent postings. There are
also falling numbers of postings so maybe it is time to liven things
up a little with a little controversial discussion.

Psychometrics have steadily infiltrated industry over the last century
to the point where some surveys suggest that 70% of employers now use
psychometric testing as an aid to recruitment. I think that
psychometrics are the modern day phrenology and bring psychology into
disrepute. To back up this provocative statement I offer the
following evidence.

Psychometrics begins with an assumption that we all have the same
personal qualities but we each have different quantities of these
identical qualities. In Raymond Cattell's case he originally
suggested 16 personality factors which he derived from factor
analysis. Warrn Norman suggests only 5 personality dimensions.
Eysenck suggests 16 bipolar dimensions - the main two being introvert/
extravert and stable/neurotic.

This is analogous to a chocolate cake recipie with flour, sugar,
butter, eggs and cocoa powder all in varying quantities. If this is
the case, how do we make a coffee and walnut cake? Do we add more
cocoa or less? Clearly the qualities of different ingredients
influence the final flavour of the cake. Psychometrics relies on the
same ingredients in varying quantities. George Kelly suggest that we
each create our own 'recipies' by adding different ingredients. He
calls them 'personal constructs'.

Anyone care to defend psychometrics?

If a job isn't worth doing, it isn't worth doing well.

Mark