Tuesday, September 1, 2009

Too Much VUSD Testing?

Theme Skills Tests, a test too far?

In the North County Times recently was an article by Stacy Brandt entitled,
“REGION: Testing improvements attributed to teachers, increase of data.”
http://www.nctimes.com/news/local/oceanside/article_cc88423e-15a4-5eb6-8755-71f8722880f7.html?mode=story

Stacy Brandt wrote something which sounds good to the public and benefits district’s administration’s need for data but is NOT likely to be in the best interests of our students in VUSD. Below printed in red is the part of the article I question:

Teachers are becoming more familiar with what students are expected to know in each grade level, which has also helped to bring up test scores, said Sally Bennett-Schmidt, assessment coordinator with the San Diego County Office of Education. State education officials rewrote the tests in 2003 to align them closer to academic standards that were adopted a few years before.

Technology has also made it easier for teachers and administrators to receive and understand test scores more quickly, Bennett-Schmidt said. This means struggling students can get help sooner.

For example, teachers in the Vista Unified School District can use the results to benchmark tests quarterly to figure out which students need help on a particular topic and give those students extra attention. Principals can also use the data to decide which students should get more intensive, one-on-one help.

We can all agree that assessment is a good. Without testing it would be far more difficult to analyze the effectiveness of various teaching strategies. But we need balance in all things including testing.

Student testing comes at a price. For every hour of testing more than one hour of instruction is lost. The time lost beyond the time taking test is lost due to need for practice for test and give students instructions for taking the test.

Teaching time is finite.

Two or even three days are lost each time students must take a standardized, bubble in, test.

In addition is what I call the Ennui Factor. The more standardized bubble in tests students take the less “special” the test is and the less ‘focus’ the students have. The students are bored by taking these kinds of tests. A greater and greater percentage of students do not see the point in making an effort on the test. These bored and over tested students just randomly bubble in rather than make an effort. I have known students who made patterns for amusement with the filled in bubbles without even looking at the questions. We do not want students so tired of bubble in tests that by the time the California Standards Test in the Spring rolls around that they no longer care about trying to do their best. Our poor Spanish speaking students must also take a CELD Test. Just too many bubble in tests.

There is a further penalty regarding teacher effectiveness that has cost that is real but difficult to quantify. It is the intangible loss of quality as teachers must devote ever more teacher preparation time to the Scan Tron machine (when it works) and to uploading the results via their computers to Edusoft. This prep time lost is quality time that teachers need to evaluate how the lessons of the day are going, decide if and what changes are needed to increase effectiveness, correct daily work, quizzes and tests. These are the tasks that allow teacher to pinpoint students and concepts that need more work and to increase the teacher's effectiveness as a conveyor of knowledge. When decisions to increase student testing are made the costs must be calculated into the decision or at some point, increased testing actually decreases student performance.

This teacher time penalty is much worse for the poor first and second grade teachers whose students do not have the hand coordination to bubble in the test answers themselves. These poor teachers must transfer by hand all student answers to test bubble in answer sheets. Teachers are given NO help to do this laborious job. Such a system begs for errors in transcription as teachers become tired.

Besides what is the point? All teachers constantly are monitoring their students by grading quizzes and correcting daily class work. The results of the giant bubble in test are not a surprise to teachers and are not necessary for teachers to know who and what needs re-teaching. Teachers already have a multitude of ways to monitor students. Bubble in testing is really only useful for district administration to monitor schools, individual teachers and students. There is benefit for district administration but really little or none for teachers and students. However the reality is the less the district admin needs for their admin purposes the better for student learning.

We have may have reached the point of lowered student achievement because of too much testing—Testing Overkill—at least as far as K-6 is concerned. Recently the district has required that teachers increase the number of district required tests from three Benchmark Tests one each time their is a report card. This is actually six benchmark tests, three in language arts and three in math. Now with the additional requirement of the Theme Skills Tests, there are at least three more bubble in tests during the school year, making a total of nine for every student.

Most teachers would agree that Benchmark Tests are important if for nothing else than giving students practice for the bubble California Standards Test required by the state. Each of the six Benchmark Tests must be scanned into the scantron machine and then uploaded into Edusoft for the district administration to monitor. These are time consuming processes. This extra work must be done by the classroom teacher as the district provides no paid help for these extra jobs.

By requiring even three or possibly four more Theme Skill Tests in addition to the Bench Mark tests that also need to be scanned and uploaded by overworked teachers, the district administration has crossed a line from effective testing with real benefits for students to overkill with likely reductions in real learning for our students.

No Child Left Behind requires test scores to improve. This newTheme Skill Test requirement is likely to mean test scores will drop not rise. Let’s hope district administration will re-think this new requirement. Over testing is as detrimental to better student test scores as overbaking is to a cake.

NOTE: Scantron has a fairly high error rate. In my experience poorly maintained machines and tests that are bubbled in less than perfectly can result in an error rate as high as 2 to 4%.

I used scantron for all my chapter tests when I taught science. On several occasions I checked the scantron scoring by hand scoring tests after running them through the scantron and found that the machine had marked more incorrect than were actually incorrect on as many as one in every three or four tests. Part of the fault was the student had not marked bubbles as darkly as needed but for some of the scantron answer sheets I saw NOTHING wrong with the way in which the student had marked the answer, yet the machine still marked the correct answer as wrong.

I made up for the problem by running all tests twice and by readjusting my grading curve to reflect the scantron error rate.

No one should rely on the scantron test results to be error free.

No comments: