Meritocracy, while ideologically attractive, is practically unsound due to fundamental flaws in its conceptual framework. In the first of what I hope will become a short series, I summarize arguments against overreliance on examination scores.
Let’s set the grounds for discussion straight. What is meritocracy? Dictionaries and encyclopedias such as Wikipedia happily proclaim that it is the practice of ranking people according to their abilities rather than by more arbitrary means such as social pedigree. But what constitutes ability, or merit, or talent to begin with? Upon further inspection, the definition of meritocracy falls apart and reduces to stating the need for a concrete, tangible designation of merit. It is at this point where things fall apart, for merit is an abstract concept for which nobody has the faintest idea how to define. And without a rigorous definition of merit, there can be no self-consistent instantiation of meritocracy.
On some level there seems to be no problem at all. After all, the problem of separating the wheat from the chaff is what examinations were designed for. Given the obvious impracticality of trying out all candidates in some role, it makes sense to come up with something that is a useful predictor of future success in that role. And thus was born the examination system, something that Singapore inherited from both Imperial China and the British Empire. And since time immemorial, Singaporeans have had to deal with examinations as some arcane coming-of-age ritual, a ring of fire that everyone has to pass through in order to become a useful and productive citizen. In Singapore (and indeed many other countries, most notably the UK), at least, examinations form the inextricable core of our meritocratic society: they form the operational definitions for merit.
This is all well and good, assuming that examinations do indeed do their job well. However, there is increasing evidence from the field of educational psychology that examinations are riddled with all sorts of statistical biases that make them highly unreliable predictors of future success. The brouhaha with the University of California (UC) system in 2002 with Scholastic Aptitude Test (SAT) scores is perhaps only the most recent high-profile case. Specifically, a UC study showed that SAT scores are as good as useless for predicting future GPAs, giving admissions officers essentially no information about a prospective student over what the high school GPA predicts.