TPRS gone wrong

I was looking for other blogs to look at and add to my blog, and I stumbled upon this TPRS teacher's blog:

http://blogs.glnd.k12.va.us/teachers/jrooke/

It shows my major, major complaint with TPRS (which I mostly love and borrow from a lot, btw, as you can see in previous posts). Look at all the "we translated" "students translate" assessment! With all the research showing that translation isn't good for bilingual proficiency, why does TPRS look like this? Why, for a method that's so incredibly communicative and includes the brilliance of circling questions, do TPRS teachers insist on translation in their assessment?

This is my goal in my curriculum development journey: eradicate translation from my assessment, evaluate whether this works, and show other teachers how to do the same!