June 3, 2012
DC CHANCELLORS RHEE AND HENDERSON’S “PREDICTED” vs ACTUAL ACHIEVEMENT GAINS IN 2010 AND 2011.
By Erich Martel
Retired Washington, DC High School Teacher
DCPS failed to meet “predicted gains” in student achievement and retention of teachers promised by Chancellor Michelle Rhee in 2009-10 as a condition for the three-year grant of $64.5 million from the Walton, Broad, Arnold and Robertson foundations. These are the same predictions of improvement that were submitted to the US Department of Education of the DC “Race to the Top” application.
The foundations’ grants were to be used for incremental teacher salary increases, teacher bonuses and “Performance-Based Excessing.”
Rhee claimed, and the foundations agreed, that students’ scores would rise, if she could “excess” teachers with effective evaluations, in addition to firing teachers with “ineffective” evaluations and award bonuses to teachers with “highly effective” evaluations. She claimed that the new IMPACT teacher evaluation system (funded by Fight for Children and facilitated by Cate Swinburn, more on whom below) would result in improved test scores and enable her to replace bad with good teachers. As a result, students’ test scores would rise.
Almost none of their predictions happened.
According to letters from each foundation to DC Public Education Fund Executive Director Cate Swinburn (who facilitated these and other foundation grants) and Swinburn’s letters to Rhee and DC Chief Financial Officer Natwar Gandhi (http://npe.educationnews.org/Review/Resources/DCPEFtoDCPS.pdf), these carrot and stick provisions would result in improved student test scores.
The results (shown in detail below) show that Rhee and Henderson failed to meet their goals:
For School Year 2009-10, Rhee promised sixteen (16) “predicted gains” – None were met (0/16)
For School Year 2010-11, Rhee promised twenty (20) “predicted gains” – Two were met (2/20)
Two year TOTAL: (2/36) or about 5.6% (not 56%, but less than 6 percent!)
*An excessed teacher is one with an “effective” evaluation whose budgeted position has been eliminated. Teachers who are excessed are notified in early May. The teacher has no right to vacancies in his/her certification area and must compete with other excessed teachers and uncertified new hires for vacancies in the teacher’s area of certification. Principals are not required to explain their decisions. Teachers not placed have several limited options, most of which involve leaving DCPS. This “excessing” process is completely separate from the end-of-year, final evaluations in late June or early July. Teachers found “ineffective” or “minimally effective” for the second time, are immediately terminated. Thus, in any given school year, all returning teachers have one of three effective ratings.
For “student achievement growth,” Rhee listed 16 “predicted gains” for 2010 and 20 for 2011.
Number of Predicted Gains Met: 2010: 0 out of 16 2011: 2 out of 20 (in bold blue)
Number of Target Scores/Gaps Met: 2020: 0 out of 16 2011: 0 out of 20
The Results, in detail:
Average DCPS NAEP Scale Scores, changes from 2009 to 2011
2009 [2011 Predicted] [2011 Actual]
Category Score Gain Score Gain Score
NAEP Elementary Math 220 +5 225 +2 222
NAEP Secondary Math 251 +5 256 +4 255
NAEP Elementary Reading 202 +6 208 -1 201
NAEP Secondary Reading 244 +4 248 -7 237
DC CAS: Percent of students scoring Proficient or Advanced
2009 [2010 Predicted] [2010 Actual ] [2011 Predicted] [2011 Actual]
Score Gain Score Gain Score Gain Score Gain Score
DC CAS Elem Math 48.0% +5% 53.0% -4.8% 43.2% +5% 58.0% –0.9% 42.3%
DC CAS Second Math 48.8% +5% 53.8% -5.0% 43.8% +5% 58.8% +2.6% 46.4%
DC CAS Elem Reading 39.6% +5% 44.6% +4.5% 44.1% +5% 49.6% -1.1% 43.0%
DC CAS Second Reading 40.5% +5% 45.5% +2.8% 43.3% +5% 50.5% +0.9% 44.2%
Closing the White-Black Achievement Gap
[2010 Predicted] [2010 Actual ] [2011 Predicted] [2011 Actual]
2009 Gap Chng Gap Chng Gap Chng Gap Chng Gap
Elementary Math 46.9% -5% 41.9% +4.3% 51.8% -5% 36.9% +4.0% 55.1%
Secondary Math 49.7% -5% 44.7% +1.7% 51.4% -5% 39.7% -2.2% 49.2%
Elementary Reading 46.7% -5% 41.7% +3.3% 50.0% -5% 36.7% +3.9% 53.9%
Secondary Reading 53.2% -5% 48.2% – 0.9% 52.3% -5% 43.2% -3.2% 49.1%
Closing the White-Hispanic Achievement Gap
[2010 Predicted] [2010 Actual ] [2011 Predicted] [2011 Actual]
White-Hispanic 2009 Gap Chng Gap Chng Gap Chng Gap Chng Gap
Elementary Math 34.8% -5% 29.8% +4.5% 39.3% -5% 24.8% -3.6%% 35.7%
Secondary Math 31.4% -5% 26.4% +11.0 42.4% -5% 21.4% -5.6% 36.8%
Elementary Reading 40.1% -5% 35.1% +3.4% 43.5% -5% 30.1% -0.7% 42.8%
Secondary Reading 41.8% -5% 36.8% +9.2% 51.0% -5% 31.8% -12.4%38.6%
Closing the Achievement Gap Between Economically Disadvantaged (i.e., poor) and Non-Disadvantaged
[2010 Predicted] [2010 Actual ] [2011 Predicted] [2011 Actual]
2009 Gap Chng Gap Chng Gap Chng Gap Chng Gap
Elementary Math 26.2% -3.5% 22.7% +1.9% 28.1% -3.5% 19.2% -1.0% 27.1%
Secondary Math 21.6% -3.5% 18.1% +3.5% 25.1% -3.5% 14.6% -2.6% 22.5%
Elementary Reading 28.2% -3.5% 24.7% +2.9% 31.1% -3.5% 21.2% -2.1% 29.0%
Secondary Reading 25.3% -3.5% 21.8% +2.6% 27.9% -3.5% 18.3% -2.0% 25.9%
I realize those tables are not very readable. Erich is going to send me them as Excel files, which I can post in a readable manner.
In short, Rhee and Henderson predicted all sorts of stunning successes — and met only 2 out of 36 goals.
LikeLike
Gary- Please tweet this out so I can re-tweet later. Can you add a Tweet button to your blog. Please!
LikeLike
I think the most concerning about this is that it has been their deliberate strategy to excess “effective” teachers and hope they don’t return. Even if you accept that IMPACT is a good evaluation tool, which is not proven, this is not reasonable. I can understand the desire to remove truly bad teachers but good ones? I also know of highly effective teachers who have had a hard time getting placements. Our sarcastic sad joke is botox and hair dye are more important in getting hired in the current DCPS culture than being “highly effective.”
I’m not at all surprised that the “Lake Wobegon” strategy of all our teachers being above average isn’t working. Much has been promised, little of quality has been delivered.
LikeLike