Juggling matric pass stats

Ve vill tell you, ja?There’s an interesting observation by Naledi Pandor, our education minister, in this story about the declining pass rate for matrics. The article notes that pass rates across the country have declined, year-on-year, but on a growing base. “The national pass rate this year is 65.2 percent compared to 66.5 percent last year, but 368 217 passed Grade 12 this year, a huge increase from the 272 488 who passed in 1998,” the story says.

“I find it strange that, despite the fact that more children have passed, we say we have done badly this year,” it quotes Pandor, in response.

But wait a second. The pass rate compares 2007 with 2006, while the matric population compares 2007 with 1998. This is a problem. So let’s complete the comparison. In 1998, the pass rate was only 51%. By that measure, 2007 is considerably better than 1998 in both absolute and relative terms, and shows an increase in the number of full-time students from 534 290 to 564 750. I couldn’t find the number of students who sat the 2006 exams, but approximate numbers in the media suggest a small improvement (of about 2 500, or 0.7%) in absolute numbers of matric passes between 2006 and 2007, despite the decline in the pass rate.

I’m no fan of our lengthy, superfluous and failing experiment in “outcomes-based education”. The pass rate should concern us, and the fact that it declined in the last three years is troubling. So, even more significantly, should matric standards. However, if schools are reaching more kids that previously didn’t go to school, perhaps the decline in pass rate isn’t indicative of declining quality of education. As long as it isn’t an excuse for failing to take the overall pass rate seriously, or a cover for a decline in matric standards, we should probably concede that Pandor’s department has achieved at least one thing: increasing the absolute number of successful matriculants South Africa produces.

Reaching any further conclusions, whether positive or negative, requires more complete data. If anyone knows where I can find such data, send me a link. I’d be interested to spend some time on it.

Similar spikes: