The core issue I was trying to highlight is that the summary says "in all countries," which is misleading. It might be better to say "in a little over half of the countries" or even "in the majority of countries." The PISA 2012 results document uses the wording "many countries and economies" to describe the situation—never "all."
However, there's another problem with the summary sentence I quoted. The PISA 2012 results has this to say about reading:
By contrast, girls outperform boys in reading almost everywhere. This gender gap is particularly large in some high-performing countries, where almost all underperformance in reading is seen only among boys.
So the other part of that same sentence from TFS, "boys generally perform a bit better than girls, but this applies only to math," is extremely misleading due to its incompleteness, in addition to being horribly structured. (I trust you understand why this is a problem for journalistic honesty.)