4 ms·

That cheating occurs is no surprise and is closely related to the question of assessing a programmer's output. Every time you instigate a method of measurement

That cheating occurs is no surprise and is closely related to the question of assessing a programmer's output. Every time you instigate a method of measurement, the programmer (or teacher) finds some way to increase the result without achieving the actual aim. In the case of a programmer, the number of lines of code output, or the number of functions written, or the number of bugs detected and fixed, can all be achieved with less "true value output" (whatever that means). The measurement can be gamed. Teachers in the UK have been accused of "teaching to the test" rather than "teaching for understanding" - but what more can be expected? When exams test for understanding rather than rote skills, only then will understanding be the aim of the teachers looking for the easy way out. Let me add that I know many dedicated teachers who do the right thing regardless of the threats posed. They encourage and enthuse their students, ranging widely over their subject to get the students engaged and interested. They know that this serves them better, and will, mostly, improve the test scores, even if not as much as "teaching to the test" would. They know they're doing the right thing, but come under enormous pressure to work more obviously towards improving test scores. I'm not surprised that many of them buckle under the strain and take less honorable courses. That's part of the tragedy - the destruction of otherwise excellent teachers.

alexandros 16y ago
I've written extensively on this phenomenon, take a look: http://lesswrong.com/lw/28r/is_google_paperclipping_the_web_the_perils_of/ http://lesswrong.com/lw/28r/is_google_paperclipping_the_web_...
guelo 16y ago
No, the tragedy is uneducated kids.
- RiderOfGiraffes 16y ago
  That's why I said "part" - but each gifted teacher lost to the profession leads to many, perhaps thousands, of less well educated children.
liedra 16y ago
See I think that these sorts of tests are looking at the wrong sorts of things. What they should be looking for is relative improvement rather than reaching a particular level. Especially in underprivileged areas, where the kids might be doing much more poorly than more privileged areas. Don't get me wrong, I think that if there are poor teachers then they should be dealt with, but I think that these standardised tests of kids across the board are not the way to go about it. I'm thinking maybe something like a "peer review" type system could be put in place that takes more into account than just the output of the kids on one test (the results of continuous assessment across the whole year, for example; teaching reviews, etc.). Of course this would come with its own issues, but it would certainly cut down on the pressure that comes from that one test!
- RiderOfGiraffes 16y ago
  Most teacher and school assessments in the UK are based exactly on that - achieved improvement versus expected improvement. Schools in underpriveleged areas and which perform poorly on standardised tests still themselves score well if the results are "good given the context and circumstances." The tragedy is that teachers who don't teach to the test and provide an excellent education that is tailored to the children they have are starting to worry about the security of their jobs, whereas poor teachers who have no real ability except to regurgitate the texts and drill the kids on process and recitation are relatively safe.
  - tokenadult 16y ago
    The tragedy is that teachers who don't teach to the test and provide an excellent education that is tailored to the children they have are starting to worry about the security of their jobs, whereas poor teachers who have no real ability except to regurgitate the texts and drill the kids on process and recitation are relatively safe. Evidence for this empirical statement? I have my doubts about this, because the best teachers I had in childhood (too few, but more than one) were able both to engage the pupils from the dullest to the brightest with deep, thoughtful questions and able to cover the fundamental facts of the lessons with efficiency.

alexandros 16y ago

I've written extensively on this phenomenon, take a look: http://lesswrong.com/lw/28r/is_google_paperclipping_the_web_the_perils_of/ http://lesswrong.com/lw/28r/is_google_paperclipping_the_web_...

guelo 16y ago

No, the tragedy is uneducated kids.

RiderOfGiraffes 16y ago
That's why I said "part" - but each gifted teacher lost to the profession leads to many, perhaps thousands, of less well educated children.

liedra 16y ago

See I think that these sorts of tests are looking at the wrong sorts of things. What they should be looking for is relative improvement rather than reaching a particular level. Especially in underprivileged areas, where the kids might be doing much more poorly than more privileged areas. Don't get me wrong, I think that if there are poor teachers then they should be dealt with, but I think that these standardised tests of kids across the board are not the way to go about it. I'm thinking maybe something like a "peer review" type system could be put in place that takes more into account than just the output of the kids on one test (the results of continuous assessment across the whole year, for example; teaching reviews, etc.). Of course this would come with its own issues, but it would certainly cut down on the pressure that comes from that one test!

RiderOfGiraffes 16y ago
Most teacher and school assessments in the UK are based exactly on that - achieved improvement versus expected improvement. Schools in underpriveleged areas and which perform poorly on standardised tests still themselves score well if the results are "good given the context and circumstances." The tragedy is that teachers who don't teach to the test and provide an excellent education that is tailored to the children they have are starting to worry about the security of their jobs, whereas poor teachers who have no real ability except to regurgitate the texts and drill the kids on process and recitation are relatively safe.
- tokenadult 16y ago
  The tragedy is that teachers who don't teach to the test and provide an excellent education that is tailored to the children they have are starting to worry about the security of their jobs, whereas poor teachers who have no real ability except to regurgitate the texts and drill the kids on process and recitation are relatively safe. Evidence for this empirical statement? I have my doubts about this, because the best teachers I had in childhood (too few, but more than one) were able both to engage the pupils from the dullest to the brightest with deep, thoughtful questions and able to cover the fundamental facts of the lessons with efficiency.