Saturday, February 18, 2012

The achievement gap, II: how our schools are working hard to make it go away

If you're concerned about achievement gaps of the sort recently reported on by the Times, you could either (re)instate rigorous, structured, direct instruction in line with the latest findings in cognitive science research, teaching each child in his or her Zone of Proximal Development, i.e., at his or her instructional level, with proper scaffolding, and furnishing each classroom with teachers who've mastered both their content areas and these best practices. Or you could:

I. Eliminate the ability of academically advanced students to get ahead in the classroom by:

1. implementing low level, one-size-fits-all instruction (for which there's no better model than Investigations math)
2. eliminating grade acceleration and individualized instruction
3. eliminating gifted programming or making it about time-consuming projects that supplement existing assignments rather about academic challenges that replace these assignments.

II. Reduce the ability of students to get ahead on their own time by:
1. assigning tons of homework of the low-ratio-of-learning-to-effort variety 
2. including massive summer projects and one-size-fits all reading lists.

III. Reduce the ability of grades to reflect achievement differences via"grade compression" and inflexible "rubrics" that:
1. employ subjective grading standards (elevating "creativity" and "engagement" over correct answers, clarity, articulateness, and solid analysis) 
2. take points off for unexplained answers, however correct 
3. give partial credit for "explained" incorrect answers 
4. keep the purely academic demands/expectations of assessments and assignments as low as possible 
4. minimize the opportunity for students to demonstrate work that exceeds those demands/expectations 
5. even if students find a way to demonstrably exceed expectations or go above and beyond academically, don't give them any extra points for it 
6. deploy "wild card" variables that partially randomize who gets what grade (e.g., trick questions; unclear directions; trivial requirements like including today's date on the title page of your report or using the word "I" in your science project abstract; rather than collecting homework, leaving it up to the students to turn it in and giving out zeroes for things not turned in on time) 
7. assign heterogeneous-ability group projects and give everyone in the group the same grade

IV. Reduce the ability of NCLB tests to reflect achievement differences, via:
1. low academic ceilings 
2. partial credit for explained incorrect answers; points off for unexplained correct answers (as above) 
3. wild card variables (as above)

V. Lobby colleges to pay less attention to high-ceiling standardized tests like the SATs and the Achievement Tests, and more attention to grades and "leadership" activities.



But then the next question becomes how to eliminate the growing achievement gap between U.S. students and those from other developed countries.

1 comment:

Barry Garelick said...

assigning tons of homework of the low-ratio-of-learning-to-effort variety

A discussion raging over at Joanne Jacobs' blog
regarding the seven "myths" of learning recently talked about in the Washington Post, has some commenters asserting that the traditional model of teaching consists of a low-ratio-of-learning-to effort. The debate comes down to whether the math problems in traditional texts from previous eras as you've shown in this blog are inferior to those methods used in programs like Investigations, Connected Math, or the various look-alikes.