SATs, contractors and doing it on the cheap

Off the topic of ELT, but still educationally related - The SATs(Standard Attainment Tests), sat by 11 and 14-year-olds in the UK. There's been a bit of a kerfuffle, to put it mildly, at the delay in releasing the results this year. There's been more disgruntlement at schools, where teachers have been resending marked papers because of inaccuracies in the scoring. So, what's going on?

the papers have been sent off to be marked to a company contracted to do it, presumably the cheapest one available.

Its name?
ETS Europe.

The marking of the SATs has been subcontracted to a company not based in the UK. In other words, British students' work is being checked and marked for accuracy in somewhere other than the UK.

I can't begin to describe how shocking I find this. It's wrong on so many levels. Now, if it was just a case of a multiple choice paper being fed into a computer, I could accept that. If it was just checking the result of a maths paper, where you can only have a correct or wrong answer, I could just about accept that too. But it seems (and honestly, I couldn't be more glad if I were wrong on this point) that the entire lot is being sent off. OK, you can make a point abour objectivity in marking: The examiner will have a set number of descriptors against which he/she will check the submitted work, and based on that assign a mark. However, I can see so many ways that marking will be inaccurate.
Just a few examples:
Orthography. The way that UK kids are taught to write is significantly different from the way it is done in other EU countries. This is a fairly neat example:
, and it's written by an adult! Imagine an 11-year-old's being deciphered by an examiner.

Cultural and Social mileau. Taken out of context, how can anything relating to a culture or society be accurately interpreted, let alone assigned a score in an exam?

Examiner's L1/L2 competency. No matter how good the examiner's English may be, nevertheless they will be marking and interpreting at one remove - that is, they will have to decode the information, recode into L1, interpret according to two sets of cultural and possibly sociolinguistic filters, then assign a mark and re-encode into English. There is no way it can be done entirely fairly, as any examiner in this situation will use affective filters in the process.

All I can say is that it's WRONG. Totally bloody WRONG.