Posted on 03/31/2008 9:34:20 AM PDT by GoldwaterInstitute
I'm pinging this because of some of the interesting information about NCLB presented in the article.
It's good to see schools take a stand and raise their standards.
I’ve had more than one educator from more than one school district also tell me that they teach to the test.
I'm not sure that's altogether bad...if the test actually measures what is supposed to be taught, then one ought to be teaching what will be on the test...that is to say, the topics that will be tested, as opposed to what the teacher thinks is "fun".
What ought not to be taught is the exact questions and answers that will be on the test.
then I guess the schools will have to stop admit illegal children who don’t speak english.....
Testing has to happen or there is no way to determine whether anything of value is being done.
There is no perfect test. But you have to have SOME test. Or you can doom the failing schools to failure forever since the people who run them will continue to rearrange the deck chairs and demand more money and we’ll continue to put children in prisons instead of colleges or in jobs.
Testing isn’t ‘designed’ to ensure that a certain number of people fail or succeed. That fact that they put out percentiles based on results in actually not something designed PRIOR to writing the test but give you a very good idea AFTER the testing of whether either the test was too hard or the students were not up to whatever level was expected.
For generations we’ve had standardized tests like the California Achievement Test. The fact that I could test 5-10 grades above my class level didn’t mean the test was flawed necessarily but it did give me an opportunity to show that I needed a little more challenge that I was getting. And some other kid is going to test a few years below his grade level and the teachers should know to spend more time on that kid and not me on any lesson (and this was when I had 40 or more kids in elementary classes).
If you understand a subject, you could design a test to verify that somebody else studying the subject actually know has a subjective understanding. That is the purpose of a test. Anything beyond that is social engineering and I’m likely going to be against it.
Question is, who is better suited to make the determination? Local communities or the federal government?
I agree with your sentiment but you are ignorant of the NCLB. The NCLB isn't a test. Every state gets to write it's own test. There is no national test, no national standard.
The article in question suggests that the states pre-existing test was too HARD for the students to meet the NCLB standards. Which sounds like whining to me.
In some cases, I'd submit that the federal government is at least no worse. If you'll recall, a lot of local communities have been known to give diplomas to kids who couldn't read or do basic math....
All academically successful children are homeschooled. The only thing a school does for an institutionalized child is send home a curriculum for the parents to follow. Some call this "after schooling".
This is true even for immigrant children. When ever I have met an academically achieving institutionalized immigrant child, the parent have told me that they have found someone to mentor their child. These are often older cousins, aunts, and uncles etc.
As a health professional with my own clinic I met many of these families. Without exception, the parents were "after schooling" or homeschooling. (By the way, homeschooling is a lot less work.)
All academically successful children are homeschooled. If they are institutionalized, this homeschooling is often called “after schooling”.
Yes, it absolutely is. The test is designed to ensure that a bell curve of results is achieved. Thus, some children MUST fail. The results must mimic the design or the test is useless.
In every standardized test, some questions are not figured in scoring. They are control questions designed to give information to the examiner. If many students pass, or if many students fail, that question it is thrown out. A question that has too many good results, or too many bad results tells the tester nothing. A question that allows one to distinguish among students is ideal.
A question that good students answer correctly while poor students don't, is a "good" question. A question that good students answer incorrectly but poor students answer correctly is a "poor" question. The examiner can distinguish the good from bad students because they've taken the rest of the test.
Have I convinced you yet that I know what I'm talking about?
In our homeschool we did not have “tests”. My goal was for the children to have 100% mastery. When 100% master was achieved then we moved on to the next level.
I know you’ve said this before about all academically successful children are homeschooled, but that is not true. It may be for most children, but not all.
My husband was not homeschooled at all. His parents pretty much ignored him (they got divorced when he was 12), and he was totally on his own at 17.
My husband did not like being poor, so he figured out how to put himself through college.
He is now director of engineering for a computer company in the Silicon Valley, and he will probably be a VP in the near future.
Where there is a will, there is a way. Some people will overcome adversity to become successful.
Actually, no. Did you work for a testing company or write or field test these for some company?
As an example, there is a great canard out there that the SAT test is designed to help white kids and harm black kids. And there are lots and lots of people who are completely convinced it is true. But I don’t take their word for it since their premise would require incredibly intelligent testing designers who were also racists and basically evil or an overwhelming amount of benign cultural bias that should be evident to everybody if exposed. (Which hasn’t been shown.)
If you’d like to point me to some specific reference for a specific test, or give me some credential you have personally which would give you access to the information on control question, I’d be perfectly happy to believe you.
But lets pretend for a moment that everything you believe is correct. What is the benefit to having half of the testing subject fail? The test then only finds a median and does not actually ‘test’ knowledge or competence at any level of understanding or study. If a test is primarily designed to pass half the students and fail half, what you are really doing is polling for a mid-point and not actually determining whether a group of subjects understands and can successfully answer questions about the subject.
You have to go back to the original word TEST. Lots of people think that its some kind of experiment. It just ‘tests’ the knowledge of the testee on a set number of questions relating to a particular subject(s). There are tests which everybody passes (think DMV).
In California, there is a graduation test to get your high school diploma. Do you believe that these tests are designed for half of the students to fail? Hopefully, you are grasping that there is some inherent illogic in this or every test would be viewed negatively.
Also, the Bell Curve is actually a means of scoring results on a percentage basis. It does not determine or question whether or not students ‘pass or fail’ a test. It is simply a method of placing test scores on a scale which puts an individual group of test subjects into order of performance. If everybody got 90% of more question right, it doesn’t mean everybody failed since some kids got 100% but it gives a means for you to differentiate between the kids who got 90% and the ones who got 100%. The Bell Curve is done AFTER test results have been completed and there are no Bell Curve applications to standardized tests. None. There is no Bell Curve for the CAT. None for the ACT or SAT. Since there is no ‘grade’ given for these, there is no purpose of applying a ‘Curve’.
These tests really only allow us to compare one child to another. So Johnny is better than Susie, and Susie is better than Cindy but we still really don't know if any of them are on grade level.
It is quite difficult to find information about tests that seems reasonable and reliable. But here is a straightforward link. http://www.edtech.vt.edu/edtech/id/assess/purposes.html
Norm-Referenced Measures (NRM) Most appropriate when one wishes to make comparisons across large numbers of students or important decisions regarding student placement and advancement. Norm-referenced measures are designed to compare students (i.e., disperse average student scores along a bell curve, with some students performing very well, most performing average, and a few performing poorly).
determine individual performance in comparison to others; standardized, comparisons among people
items produce great variance in scores, perhaps with less than 50% scoring correctly
item analysis used to select those items that were answered correctly by those scoring high on a test but incorrectly by those scoring low on a test (a positively discriminating item)
it is inappropriate to use NRMs to determine the effectiveness of educational programs and to provide diagnostic information for individual students; items cover a broad range of content and often represent a mismatch between what is taught locally and what is taught in other states
http://www.edtech.vt.edu/edtech/id/assess/diagnostics.html
Item Analysis for Norm-Referencing Item analysis is an important technique for perfecting norm-referenced tests that discriminate between high and low scorers. You begin by choosing some standard for high and low scorers on a test. For example, high scorers = students with test scores in the top 1/3 of the class, and low scorers = students with test scores in the bottom 1/3 of the class.
In the example below, 12 students are in the high and low groups. On a specific test item, if the 12 students in the high group respond to the item correctly (answer B) and the 12 students in the low group respond to the item incorrectly, you have a perfectly discriminating item.
You can compute a discrimination index for an individual test item as follows: subtract the number of students in the low group responding to the item correctly from the number of students in the high group responding to the item correctly (12-0). Divide this figure by the number of students in one group (12). The value of the example above is 1.0, or a perfectly discriminating item.
Designers of norm-referenced tests typically seek items in the range of .35 to .70.
If most students in both the high and low groups respond to an item correctly, your discrimination index might be .14. This is a red flag that the test item is too easy.
If more students in the low group respond to an item correctly than students in the high group, your discrimination index would be negative (-.08). This is a red flag that the test item is flawed.
My child has a reading disability. I've been battling his teachers for 2 years because I have noticed a drop in his classroom test scores. I look at the tests and think, "Good grief, if you want to know if the child understands photosynthesis, just ask him to explain it!" The tests questions tend to be written in such a complicated way, explicitly designed to be confusing. Now you aren't measuring whether my kid knows about photosynthesis. You're measuring his ability to read and understand your way of asking questions. That is fine, and that is important, but it isn't testing what he knows about science.
I've just found out that an edict came down from the school district. Classroom tests must mimic the form of standardized tests. So, what are they really testing?
A criterion referenced test, on the other hand, is to determine mastery of a particular topic or subject at a particular grade level, and wouldn't necessarily show a bell-shaped curve if the students were of at least average intelligence and had been properly taught.
Note, I'm not a test designer either...
No school flunks because of a couple of kids. But the act was designed to prevent massive failures like the systems which existed in many larger metro areas.”
The school I work at is 80% Hispanic, and 100% Free/Reduced Price Lunch. Many of the students are “high-mobility” or frequent movers. One of the other schools in my district has the same stats I mentioned for mine, and also a 60% turnover rate per year. In some neighborhoods, that's all you can get unless you bus kids in from elsewhere.
I don't mind the accountability provisions of NCLB, but I don't see how ALL students will be reading at grade level can be accomplished. That is period, not just within the specified timeline. NCLB allows for (IIRC) 2% of the kids to be in special ed or ELL. In my school, my students alone are 2.2% of the school population, and none of them are ELL.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.