Wildly divergent results

Using two different approaches, the difference in literacy-testing scores for prisoners is astonishing. According to Radio New Zealand;

A new test shows the literacy and numeracy problems of prisoners to be much worse than previously thought.

The Burt Word Recognition Test found 2% of prisoners need literacy help and 17% need numeracy help.

A new screening tool, developed by the Ministry of Education, was also trialed recently on 197 new prisoners.

It found 90% needed literacy help and 80% were not functionally numerate.

I had to read this twice, myself, thinking perhaps there was a mistake in the first figures. The first test finds 2 in 100 prisoners need help and the second test finds 90 in 100 need help. A few people should be scratching their heads over this and asking themselves how much time and money is being wasted administering tests which produce such wildly divergent results.

(The Burt test is used to assess the needs of offenders on release or serving community sentences, so it isn't just 'nice-to-know' stuff.)

(And yes, Adolf, the copy-writer needs help.)