Bill Pfeifer asks about the test \doc\web\97\07\family.txt Thanks a lot for lending me an ear > Date: Sun, 28 Sep 1997 16:13:57 -0800 > From: family > To: Arthur Hu > Subject: Re: Sec. Riley's Speech to National Press Club > Arthur, it is good to hear from you. I saw your recently published > article-good job. I have been swamped and unable to get to all my email > to respond. I will attempt to do so now below your original question. > > Arthur Hu wrote: > > > What's your opinion of the new Washington test which currently > > only "passes" the top 15%, and throws questions with 3 view > > drawings with dimensions and hidden lines and doing college > > level statistics and probability with frequency histograms at > > 4th graders, and then promises that we'll pass 100% of students > > to that standard? > > Judging from the tone of your question, it appears that you do not like > the recent Washington Test. I will try to respond given the limited > information I have. > > 1)The Washington test- I have not seen the test and therefore cannot > criticize or justify any comments. Our recently formed Charter School > and it's teachers have voiced interest in getting the test to give to > our students. I would be interested in getting a copy, or samples if > that is all that is available, to review. Do you have any? I'm told that it's locked up tight, they won't let anybody look at the original test. I'll try to see if they can at least release a skills inventory, but I'll bet they don't have one, given that the one required for the 5 problems I saw on the web and in the Seattle Times, the skills required aren't in the Addison Wesley text my district uses, and therefore not likely to be an any competive books either. There is no mention of using similar triangles to convert the size of a shadow of one object that you can measure into one of a much taller object, no instructions on how to interpet a oblique drawing and top-view with dashed / hidden lines and dimensioning arrows, no example of a frequency histogram with unlabled intermediate tic marks. There are no problems which require more than 2 or 3 simple steps, or require a written explanation of a complex multi-step problem. How they can expect children to master an apparently unlimited knowledge set when they can't even get them to master the old standards is sheer insanity. How did they get a committee to establish that content that is not contained by any of the common 4th grade text books constitutes "essential skills"? > > 2)Passing the top 15%- This may or may not be a concern depending on the > rest of the facts. How far away from passing were the other 85%? Is the > level of material appropriate and what we want to focus our attention > on? Does the test include the basics that are needed for a firm > foundation to build upon? As you know tests drive curriculum. Only 20% passed the math, only 15% "passed" every area. If you did not meet "the standard", you did not pass. Terry Bergstrom said that the idea is to attach mastery of this test to the high school diploma, logically, this would replace the current system under which you can pass with a "D" grade with what appears to correspond to a B+ or an A-, or else you flunk. > > 3)Three view drawings with dimensions and hidden lines- > Without seeing what you are referencing it is hard to address. I don't > see a problem with three view drawings so long as it does not make up > the entire test. Is this to figure area, perimeter, volume etc.? Hidden > lines - If intentional deception is taking place I am against it. > However if for example in a perimeter problem, there are lines without > dimensions that must be extrapolated from other information given, I see > no problem. I didn't get 3 view drawings explained to me until 7th grade. I didn't learn about oblique view until high school. The sharpers kids will be able to figure this out since it's common to anybody that does any reading of stuff like Popular Science, but any kid who hasn't been exposed to the stuff (namely, girls, the poor, and minorities) are going to be clueless unless it's in the curriculum. The book they already have has so much stuff jammed into it, there's only one or two pages on each topic. How in the world are they going to fit in jr high level shop and statistics? > > 4)College level statistics and probability with frequency histograms- > Obviously if this is truly college level it is inappropriate. However > since no statistics and probability was ever taught in elementary > schools in the past this, I am sure, is a reflection of the NCTM math > standards. Some of this is fine but not at the exclusion of basics and > at an appropriate level. It would be interesting to see an item analysis > of the test. I have seen a trend to teach more statistics and exclude > the basics with many students not knowing that 3/4 is .75 etc. > Actually the AW book does have some simple statistics in it, and some graphs in it. Again, the kids that have been exposed to popular science and check out Time Life science books like I did are going to be in good shape, but the kids that haven't had it spelled out to them are going to be up a creek without a paddle. Unless they're testing content that's on the curriculum, this new asessment isn't any different than an IQ test, and a measure of how much their parents have exposed their kids to stuff that isn't covered in school. > 5)Promises to pass 100% to that standard- Promises are rare in education > and I would like to see their action plan to do so. What would be > sacrificed to carry out that action plan would be the next question? So > long as we are talking about academics and not a social agenda, it is > beneficial and we can debate the appropriate level and content that is > needed. There is no action plan other than "we'll revamp the curriculum, and teach them the skills they need to pass this test" Promises like this aren't that rare, President Clinton has set a goal of 100% of Americans reading at grade level by grade 3, from the perspective of anyone with even a mimimal grounding in testing and statistical populations, any one who makes such a statement should flunk even basic statistics. Riley of Bellevue schools said that the fact that Somerset elementary passed 70% over standard was proof that every student could pass this test, even though Somerset scored higher than 98% of the schools listed in the Times, and had 90% over average on the CTBS, and had only 2% free/red lunchers. He also said he believed that every high schooler should be able to master basic calculus. Do you really believe this "every child can be taught to master what is expected of them" philosiphy that is the foundation of Objective Based Education and Performance Based Testing? > > I would appreciate more information and your perception on the > Washington Test. If we have a bad or overzealous test it doesn't totally > negate the need for a state or national test. I would rather see a > slightly overzealous test than a weak watered down test, or affective > test. The perpetuation of grade inflation with no test gives the student > and parents the false sense that everything is fine until it is to late. In my view, the national 50th percentile is doing just fine (well, maybe not in California if whole reading has done as much damage as they say it has). Just about the only populations in the world that do better than white Americans are the Asians and Asian Americans, and guess what their educational foundations are built on? High conventional test scores, and massive rote memorization of facts and content. Americans aren't being beat by people who have demonstrated superiority in performance based tests, they have been beaten by kids who have aced their basic facts, and IQ tests that aren't anywhere as complex as some of these test questions. The AW book appears to cover more material, not less than I got when I was a 4th grader (it's got cones, spheres, and cylinders, mental math, and some decent estimation techniques). As long as grades are returned to a criterion, and not slavishly norm-referenced definition, I don't have a problem with grades. Compared to the old fashioned tests, the new tests don't really measure anything different than the old tests, the assumption that the kids that are good at "problem solving" skills are somehow different from kids that ace the "basics" is obviously wrong, and I believe rooted in the assumption that since minorities and the poor did badly on the old test that there was some other "factor", namely thinking skills that weren't bein measured by the old test. Except the new test hammers the poor and minority even worse, since multiple choice questions are much easier, and the kids at the top have an even bigger advantage when it comes to content that isn't covered by the curriculum and complex thinking skills. You don't master complex thinking until you've mastered the basiscs, the assumption that you can skip the basics by moving straight onto these mental obstacle courses is really warped. > > I look forward to your response, and any further information on how to > get a copy, or examples, of the test. > > Dr Bill Pfeifer > > Here's an excerpt from what I've posted on my web index, it has pointers to examples at Sample Questions for Washington State: http://csl.wednet.edu/Web%20page/Assessment/Pilot4/Pilot_menu.html I'm attaching a repeat of the 1st article with my ascii interpretation of the bird feeder question. http://www.leconsulting.com/arthurhu/test.htm#reform (deleted, see web page)