South Africa is a country with a notoriously dysfunctional school system. Up until quite recently however the sole criterion for publicly evaluating the success of government's educational policies was the annual matric pass rate. There has thus been an ever-pressing temptation for ministers of education at national and provincial level, and the officials and principals under them, to try and manipulate the results in an effort to secure increasing pass rates.
One of the ways this has traditionally been done is for schools to "cull" their weaker pupils before they reach matric, thereby inflating the pass rate. In a January 2012 paper Dr Stephen Taylor noted that one way one could control for this tendency - in evaluating trends in the matric pass rate - is by dividing the numbers of pupils passing matric in a given year by the number of Grade 10 pupils two years previously (see here - PDF).
In other words what is the "pass rate" when measured, not against the number of Grade 12 pupils who wrote the exam, but rather against the number of pupils in Grade 10 two years previously (before a whole lot of them dropped out the system)?
This is not, Taylor notes, a perfect measure as there are high levels of grade repetition in Grade 10, compared to earlier grades, "which causes a bulging of grade 10 enrolments." It is the trend then which is important, rather than the absolute figure.
What emerges then when one applies this form of measurement to the 2012 matric results?
The matric pass rate last year increased to 73.9% from 70.2% in 2011 (see Table 1). The ratio of successful government matriculants to Grade 10s in public schooling (two years prior) was 37.5% - up from 35.2% the year before (see Table 2). Thus the trend of an increasing matric pass rate holds, even if one uses 2010 Grade 10s, rather than just those who ultimately wrote the 2012 matric, as one's base.