CUNNINGHAM: ESSAY TESTS DON'T WORK THERE IS NO GOOD WAY TO SET PERFORMANCE STANDARDS EVEN WITM M-CHOICE STANDARDIZED TESTS NO GOOD BEYOND K-5 \doc\web\99\09\cunntest.txt So at least we're in agreement that the very notion of a criterion based test with a high pass point for all students is a bad idea. My conclusions: Expectation based on proven student averages - good. Expectation based on continuous improvement well beyond current levels - bad Expectation of "proficiency" from all students - bad "world class standards" - bad assessment of content not yet taught - bad Use of test to determine proficiency - bad Use of test to determine rank order of general ability - good So it is good to use Stanford 9 to establish the average is about 40-50 percentile, and pick out which students are at top and bottom. It is _not_ good to use Stanford 9 or ITBS as WA's Terry Bergeson said "60th percentile is still nowhere near where we want to be". It's not the job of the state to require all students perform at the 85th percentile. If you ask me 40-50th percentile isn't bad, especially when a third of the kids are Latino with ESL problems. From: "George K. Cunningham" To: , , , Subject: Re: [fair-diversity] Re: CA Testing ALL 8th graders on algebra? Date sent: Wed, 7 Jul 1999 13:23:57 -0400 In discussing opposition to standards-based education reform, you must decide whether it is always bad or whether it is just the way it is being implemented. The solution to fixing it put forth by one segment of the education community is to implement it with alternative, authentic, performance-based assessment methods. When you get past the educababble, what this means is old fashioned essay question. After trying these methods for most of this decade, the verdict is now in. It doesn't work. For evidence of this, check out the May issue of Phi Delta Kappan where many of the original advocates of this approach explain why it doesn't work very well. There are some good points to the STAR system (California's system). The use of off-the-shelf standardized tests makes sense for assessing reading and basic math skills in the elementary schools. This is the most economical, reliable and valid way of conducting such assessments. There is no other construct measured better than reading comprehension using the conventional objective methods found on standardized tests. I don't believe standardized achievement tests are particularly useful for students in middle and high schools. Their focus is not on reading, and the curriculum is so diverse across students in a school, schools, districts, and states that standardized achievement tests can not assess what students have been taught and instead become measures of high level thinking, which can not be taught. Here is where the STAR augmentation comes into play. It is a set of items intended to specifically assess California's content standards. They use sound objective testing methods. The intention is to expand this approach, which I would support. California is also developing a rival system called the California Applied Assessment System (CAAS). If it ends up being similar to the STAR program this will be progress, but if it heralds the return of the failed CAS system it will be a disaster. STAR is a creature of the California Board of Education, established to the dismay of the California Department of Education. CAAS will be under the control of the CDE. The biggest unanswered question is how to set performance standards. This is the level of performance required for a student to be said to be performing at an acceptable level. Cut-scores are an example of performance standards. At the present time there is no accepted, correct way of setting performance standards. There aren't even any good proposed ways of doing this. The most carefully researched and expensively implemented are the performance standards used by NAEP. These have been savaged by almost everybody. George K. Cunningham Professor University of Louisville -----Original Message----- From: arthur hu To: mike_mckeown@qm.salk.edu Cc: gkcunn01@ULKYVM.LOUISVILLE.EDU Date: Tuesday, July 06, 1999 7:10 PM Subject: Re: [fair-diversity] Re: CA Testing ALL 8th graders on algebra? Thanks, that makes somewhat more sense, but somehow I still think that you folks have merely evolved OBE from fuzzy standards to traditional academic standards. It's still a mutation of OBE. George, do you agree at all that these new STAR tests are simply replacing peformance based fuzzy outcome tests with multiple choice tests, but keeping the "raising the bar to world class standards" OBE ideas? I see the education dissidents falling into line supporting yet another variatoin of "standards based" education. Send reply to: fair-diversity@egroups.com Date sent: 5 Jul 1999 14:13:30 -0700 From: "Mike McKeown" To: arthurhu@halcyon.com Copies to: "education-consumers@lists.dunde" , fair-diversity@egroups.com, h-bd@egroups.com, wa-ed-deform@egroups.com Subject: [fair-diversity] Re: CA Testing ALL 8th graders on algebra? > RE>CA Testing ALL 8th graders on algebra? > 7/5/99 > > "All" students in grades 2-7, and in grade 11, take both the SAT-9 and a grade specific test linked to the California Standards. > > Students in grades 8, 9 and 10 "all" take the SAT-9. Those in algebra > I, geometry or algebra 2, or "integrated" math courses covering the same content in different order, take a test linked to the course they have taken, e.g. algebra I students take the algebra I test, no matter what grade they are in, first year integrated students take the integrated I test no matter grade they are in. > > Most of the data that has been reported for grades 8, 9, and 10 is limited to those students who took algebra in grade 8, geometry in grade 9 and algebra 2 in grade 10, or the matching years 1, 2, and 3 of integrated. The reported scores contain the 35 questions of the course-specific exam plus 15 "foundation" questions that were answered as part of the SAT-9. > > The scores in these grades reflect how the top cohort of each school or district is doing in a course for which the participating students were deemed ready by the district or school staff. Scores on the 8th grade algebra test then reflect the degree of preparation of the entering students as well as the quality of the teaching and learning in the algebra classes. Districts that place students in algebra before they have mastered the prerequisite skills will have a high fraction of underperforming studen > Mike > -------------------------------------- > Date: 7/5/99 1:51 PM > To: Mike McKeown > From: arthurhu@halcyon.com > Mike, are all 8th graders tested on algebra or integrated math > regardless > of whether they have take algebra? Traditional sequance is algebra > 1 in 9th grade , algebra 2, trig and precalculus with calculus in > college. Are you on board this movement to move everything down > one grade when it hasn't been proven you can make even more > than 50% of kids pass algebra anywere in the US at 9th grade? > > > ------------------------------------------------------------------------ > FreeShop is the #1 place for free and trial offers and great deals! > Try something new and find out how you could win two round-trip tickets > anywhere in the U.S.! http://clickhere.egroups.com/click/368 > > > eGroups.com home: http://www.egroups.com/group/fair-diversity > http://www.egroups.com - Simplifying group communications > > > > Arthur Hu arthurhu@halcyon.com Education Deform Critic Index: http://www.leconsulting.com/arthurhu/index/edreform.htm listserver: http://www.egroups.com/list/wa-ed-deform