MSPAP failings

January 03, 2001|By Bill Evers

STANFORD, Calif. -- What should parents and taxpayers expect of a state-mandated test given to hundreds of thousands of public school students?

A reasonable expectation would be a test that measured student achievement against clear expectations for learning, one that had been carefully checked for accuracy, one that was graded fairly rather than capriciously. Regrettably, Maryland's state school assessment, the MSPAP test, falls woefully short on all of these important counts.

I was the chair of a content review panel that reached this conclusion after being sequestered in a windowless room in the Maryland State Department of Education and permitted to read through nearly 200 white plastic binders containing MSPAP test items.

The panel consisted of a mathematician from the University of Rochester, a scientist from California State University, a reading specialist from the Oregon Research Institute, a writing specialist from the University of Wisconsin and myself, a political scientist from Stanford University's Hoover Institution.

It became quickly evident to our panel that students do not need to have learned much of anything of substance from their lessons to do well on the MSPAP. Students and teachers start the test day with a scripted "pre-assessment activity," during which background knowledge and vocabulary are revealed. Additional knowledge needed to answer the questions is taught on the spot in MSPAP resource readings.

Official Maryland policy is thus for teachers to "teach to the test" whatever content is needed 20 minutes prior to the test itself! State education officials seem to have little interest in finding out what academic content students have learned before the test day, and that discourages schools from teaching solid academic content in the classroom.

Aside from this policy of "just in time" teaching, the test and its grading guidelines had an abundance of errors.

Science questions frequently had experiments that would not work as claimed or were potentially hazardous to students. In one history question, the wording of the U.S. Constitution was confused with the wording of the Declaration of Independence.

Further, the sample student answers used to train the graders of the exams placed little value on accuracy. One sample had Mahatma Gandhi as a woman; another had the Pilgrims landing in Maryland -- and these errors were not flagged. Spelling and grammar are never graded on a no-excuses basis, and top-scoring responses were riddled with misspellings such as "lerned," "permition," "becouse" and "seid"(in grade 5). The MSPAP does not adequately test the components of good writing, and reading skills are not directly tested.

One of the most disturbing aspects of the MSPAP tests is the manner in which they are graded.

For example, fifth-grade students had to read a passage about an activity from their test booklets and were asked what the author could have done to help a reader better understand the passage, other than adding an illustration. A sample student answer, which was well stated and grammatically correct, was summarily given zero credit.

Another student gave an answer that was grammatically incorrect and did not adhere to the question ("The author could of drawed an illustration") and was given a grade that indicates a "satisfactory" answer. The top score was given to an answer that was unintelligible and also non-responsive to the test question ("The author could of gave an example of the data chart"). While the MSPAP test content is confidential, this information should give a sense of the troubling nature of the grading system.

It is alleged that the MSPAP challenges students to use "higher-order thinking skills," but our panel found no evidence to support this. In fact, the MSPAP is pitched well below the level of testing found in states such as Virginia and California or in America's top economic competitors, including Japan and Germany. Maryland students are not tested on their deep thoughts but rather on their ability to spit back what the MSPAP feeds them on test day.

A great mystery to our panel is why Maryland parents and taxpayers tolerate this test, which is riddled with factual errors, graded unfairly and doesn't even cover the content in the state's own academic standards.

It is neither an adequate gauge of what the students have learned from their schoolwork nor a test of their critical thinking.

Our panel conducted the first independent evaluation of the content of the MSPAP tests, but our detailed 300-page report has been embargoed by the Maryland State Department of Education. Maryland citizens deserve to know the truth, and we encourage the state agency to release our complete report.

Bill Evers is a political scientist at Stanford University's Hoover Institution.

Baltimore Sun Articles
Please note the green-lined linked article text has been applied commercially without any involvement from our newsroom editors, reporters or any other editorial staff.