MSPnet Blog: “High Stakes Accountability”
posted September 16, 2014 – by Rena Stroud
Earlier this year, Marc Tucker, President and CEO of the National Center on Education and Economy, posted a series of blog posts on accountability that were revised and rewritten into one cohesive report. Tucker suggests a complete redesign of the current system – which he reports has shown no signs of success (and actually significant harm) in its 10-year-tenure – with one that will improve the education system for all.
Tucker’s report contains several directives, including transforming the “blue-collar conception of teaching,” reimagining the content and timing of high stakes testing, and redefining the ramifications of student scores on those tests.
Though we could spend days debating the particulars of Tucker’s reimagined system, I’m going to focus on his proposed changes to high stakes tests. Tucker advocates replacing the current “low-level English and math literacy” tests whose scores capture a “very narrow slice of student accomplishment” with “very high quality assessments” given at “no more than three key points on the trajectory from grade one to the end of high school.” These tests would be designed to capture a more complete range of knowledge from a much wider list of subject areas (note that English and mathematics tests would be administered every other year to a sample of students). Instead of the traditional multiple choice format, tests would be comprised of “performance items, many of which would require the production of such things as extended essays, working robots, works of art and so on.” If you’re thinking this sounds expensive, you’re right: “These assessments would be expensive and time consuming to develop and administer, and they should be.”
It certainly sounds enticing – a way to truly get at essential skills – but is it feasible? Creating “very high quality assessments,” administering them, and scoring them in a way that would be useful to teachers and schools as well as policy makers is no small task. This is not to say they should not be undertaken, but rather a point made to stimulate discussion. Is this the right system, or at least a system that is worthy of exploration? And, if so, what would it take to make it work?
Let’s move on for a minute to discuss one piece of Tucker’s proposal that I found particularly interesting: the final high stakes test that is to be given – at the earliest – at the end of sophomore year of high school. If a student passes the test, he or she would have the option of leaving high school and moving on to college. They could also choose to stay in high school in order to better prepare themselves for more selective colleges (or for any other reason). What do you think of this idea? In addition to issues of social maturity (imagine a whole flock of 15- and 16-year-olds on a college campus!), what other problems might arise from such a system? Should we be concerned about the two years of material that students who pass the test and choose to move on are missing in their junior and senior years? What impacts might these tests, given the possibility of fleeing high school early, have on student motivation?
Accountability and high stakes tests are active, engaging, and ever-evolving topics. Join in the discussion to share your opinion. And check out last week’s MSP News for a link to Marc Tucker’s report as well as to several other articles on accountability.
Accountablity System
posted by: Pat Fitzsimmons on 9/17/2014 7:49 pm
Fraction calculation
posted by: Richard Askey on 10/7/2014 11:13 am
http://www.schoolinfosystem.org/pdf/2013/02/TIMSS_fraction_item_raskey 22013.pdf
Of the US states which took this test, Massachusetts had the best score, and it was significantly higher than Finland. However, the score for Finland was the third lowest and significantly below random guessing. South Korea had about twice as many students who knew how to do this than in Massachusetts. Someone might show these results to Vermont's Secretary of Education.
The form in which this was carried over was not correct. The url was broken at the wrong place. Type the full url if you want to see evidence that 42% of Finnish 8th grade students thought that the right way to compute 1/3 - 1/4 is (1-1)/(4-3), and see how some US states do on different ways to compute this. Maybe someone can fix the url on the version that appears on the screen.
post updated by the author 10/8/2014
High Stakes Accountability
posted by: Bud Meyers on 9/18/2014 8:28 am
Thanks for raising this!
posted by: Brian Drayton on 10/6/2014 11:24 am
1. It is hard and expensive to design good test items, especially if they are to provide evidence about higher-level capabilities. It's even more expensive to administer and then score such things.
These truths are often raised in a way that implies that this level of effort cannot be attempted. My question is, why not, if the quality of our children's education is so essential to economy, innovation, well-being, informed citizenship, etc.? By choosing cheaper "dipsticks," we are getting data that are not giving us the answers we want -- and then using those inadequate data to make all kinds of strategic decisions! Makes no sense to me.
2. "We." Is it in fact, on the whole, desirable that there be a uniform system across the country at the "grain size" of individual students at many grades, given that age-grades have only the roughest correlation with intellectual and other kinds of development? A teacher, parent, school, or district certainly needs prompt, useful diagnostics about individual students -- but these are at the level at which qualitative/contextual factors are also known, and so could (and sometimes do) contribute to appropriate use of the information. At higher levels, however, statistics come into play but always with the recognition that important information is lost along the way...