The USA Today had an interesting article on teachers/schools/districts manipulating/cheating on student test scores in various ways…http://www.usatoday.com/news/education/2011-03-06-school-testing_N.htm?loc=interstitialskip.
Apparently, multiple research studies in different states have been conducted revealing that statistically improbable test score improvements at certain schools or in certain classes occasionally occur that look awfully fishy and would be almost impossible to achieve in reality – thereby raising “red flags” and the question of impropriety. However, I could have told those researchers that education statistics should often be taken with a grain of salt based on my very minimal and brief experience working at an educational non-profit and as a teacher – no expensive research projects needed. Education statistics are all too often a load of garbage and can be manipulated in a variety of ways (like most other kinds of statistics, come to think of it). Whether it is teachers making up baseline test scores, tests being conducted in non-standardized environments (e.g. students get a different length of time to take pre vs. post-tests, students’ attitudes are different on the Friday before winter break vs. the Monday of the second week of school, etc.), or the more blatant ways of “cheating” (e.g. teachers giving students the exact answers to test questions on study guides, administrators going back and correcting student answers, etc.), all sorts of inconsistencies and outside factors come into play when analyzing education statistics.
This is only one reason of many why I think it is silly and misguided to tie teachers’ salaries to students’ test scores, hand out funding to schools based solely on test score improvements, etc., etc. – the pressure to cheat or fake or manipulate or bend or fudge education statistics only increases when these added ‘incentives’ or ‘dis-incentives’ are thrown into the equation.