Laurie said in a recent blog comment:
“…I have long wondered about the irony of measuring language acquisition using NUMERICAL data…”.
I replied, as a comment, and make it a blog entry here because then it is more searchable:
Laurie I think it’s time for an investigation. I’m not wondering anymore. The manner in which we have invested in numerical assessment is cavalier, at best. So what if we’ve been doing it for many decades – we still need to look at it. Let’s compare the din of language, that is, the immeasurably complex possibilities of almost infinite possibilities of words in different orders, to an ocean. Words float all over the place, the flotsam and jetsam of language. They come and go in mind boggling combinations, like these here in this sentence. Now, are we really going to try to assign a number to those possibilities? I think that the only way to get accurate data is to create a body of evidence that has two aspects, one visible and one emotive. If a child studies a language for four years, let’s say, and we have, ON FILM, a set of eight benchmark interviews with each child (a pretest and a postest for each year), then, by the end of the course of study, we can see that the children either can or cannot understand and express themselves in the target language. Secondarily, they can either write or read in the language as well. This is a visual, not data driven, way of seeing if the child has learned anything. (This is where the ruse occurs – as long as teachers can create tests, they can teach to those tests and fool people. But film doesn’t lie and that is why I propose this visual assessment idea and the dumping of the data format). The emotive part, not scientific at all, to me, is the most indicative of success. Ask the kids if they feel that they are learning. Look at them in class (1st year students excepted – basically they are still like small children, getting input, so that their neurology is much less visible than CI trained kids at higher levels). If an administrator walks in, of course, they change the energy. Those walk in observations by administrators reveal nothing, except that teachers who are normal human beings tend to shut down some of their light when being observed (judgement has no place around languages). But, after four years, ASK THE KIDS if they are happy about the time they spent. Walk into my classroom tomorrow. Look at Chris. He is a language sponge who is a funny person. He has fun in French class. He is about six months away from speaking a ton. Look at Janine, who commented here yesterday. She sits up, squared shoulders, clear eyes, locking on to every syllable, decoding way more than 90% of what I am saying. Look at Trey. He astounds kids from other French 1 backgrounds with his output. Heads spin when he speaks. His questions in French reflect instant and complex higher order processing – this is a French 2 kid insisting on asking complex questions in the target language! (and guess what? the grammar is there!) But on the assessment instrument last year, nobody caught that. They asked him to describe some kids next to a school bus and he did, scoring something like 36 correct answers of 40 on the overall test. But that is not nearly as accurate an assessment of what he has learned as his speaking in class. He, admittedly, is rare, a supertalent, because most of my current kids in level 2 won’t begin strong output until April, but 30% of them, the ones with the discipline to truly listen in class for three years, will pass the AP French exam next year with scores of 3 or better (end of their 3rd year as juniors). I bet $5,000 on that with someone, for real, and I will pay up if the kids don’t deliver. But they will do it easily, because they will have heard and read French solidly for three years in ways that are musical, story driven, poetry driven, and therefore meaningful to them, so that the output will be there – output that will be human and not robotic. Trey will be a language output symphony by then. CI works, outhshines in the long run, when taken out of the data machine school scene thing. Why take the AP exam, then, if it a data gathering test? Great question. Reason one is that that exam becomes less and less stupid each year. They are getting better over there at the College Board. Secondly, and mainly, because Krashen trained teachers can play in any pool. We can play the data game for the data robots who then approve of us or not depending on who wrote the test and how robotic the administrator is, but we can also actually teach for acquisition. Our CI/TPRS trained kids can swim with the big sharks. Their 3’s and 4’s on the AP next year won’t even reveal all they know, which (here is the point of this ramble and it’s your fault Laurie) is because WE CAN’T MEASURE WHAT OUR KIDS ARE REALLY GETTING IN STEPHEN KRASHEN/BLAINE RAY BASED CLASSROOMS….O.K. I’m done.
P.S. Another point must be made – when 96% of kids (national statistic) are gone in level 4, with only a handful of white obedient girls left in the fourth year classes, we need to look at anything that will bring reform. And, as we get stronger and stronger (have a good week, y’all!), and we have 96% or thereabouts of our students still with us after four years, because they weren’t tested to death with left brain whips, then people will start to appreciate the affect – yes I mean affect and not effect – that data driven programs have on kids. All they want to do is learn the language and have fun. We don’t have to make it so complicated and serious.