D.C. Public Schools Revises Controversial Teacher Evaluation System Implemented By Rhee
Teachers in D.C. Public Schools who perform well will see some easing from a controversial evaluation process introduced by former DCPS Chancellor Michelle Rhee.
The 290 teachers in the school system who earned ratings of "highly effective" for the last two years and who scored a 3.5 average on their first two evaluations this school year can opt out of the final three observations by district officials that take place over the year, The Washington Examiner reports.
DCPS evaluates its teachers and staff on a system it calls IMPACT, which assesses teacher performance according to student achievement, instructional expertise, collaboration and professionalism. Teachers are observed five times a year and receive scores on a scale of 1 to 4, with a score of 1 being "ineffective," and 4 being "highly effective."
D.C. Schools Chancellor Kaya Henderson already relaxed IMPACT rules in July by allowing some teachers who receive two consecutive "minimally effective" scores to keep their jobs -- those teachers would have been dismissed under IMPACT's original provisions, The Washington Post reported.
IMPACT has been a controversial system since its inception. Advocates tout its process of holding teachers accountable and note the program's incentives for improved teacher performance. Critics have argued that it judges teachers unfairly -- in which student scores on standardized tests could count for as much as half of a teacher's evaluation. Some have argued that the system could drive teachers to cheat to improve scores that contribute to evaluations, annual goals and bonuses.
DCPS and the U.S. Department of Education are investigating system schools regarding questionable scoring patterns and alleged cheating incidents among teachers during Rhee's tenure, between 2008 and 2010.
In July, DCPS fired more than 400 teachers as a result of poor IMPACT scores, while 663 Washington Teachers' Union members received bonuses for earning high scores.