The single word ‘test’ sets off certainty and abuse.  Toss in the word ‘evaluation,’ especially ‘teacher evaluation’ and the argument becomes furious.

analyzing 8-week tests

For example, based on the once a year test, California schools received their Annual Yearly Progress scores mid-August and on September 13 Academic Performance Index scores, statistically calculated mainly from the test assessment.  Some schools were grinning and some were down so far it looked like up.

At the same time California legislative bill SB1381 is ready to be signed by the governor which over time will change the test results for schools because Kindergarten students must be 5 years old by September 1 rather than December 1 (with possible waivers, of course).  This change introduced over three years is guaranteed to revise the test scores for even the most low-performing third graders in the next few years.  The older the student, the more likely he or she is to understand how to perform.

Why the fury?

Read any newspaper, education magazine, or online journal to read a long list of reasons one test is an unreliable measure of a student or teacher.  Here are three often named: scores can bounce for a student from one year to the next; short tests every 8 weeks or so assesses what students are learning and provides opportunity to revise teaching; the tests used for AYP and API do not “measure the social skills that are crucial to early learning.” See Daniel Leonhardt’s article “Stand and Deliver” in The New York Times Magazine, September 5, 2010.

The Congressional Edujobs bill with money being sent to states will allay some anxiety during this year as more teachers are not worried about their positions and thus not so vehement about tests-whichever exams are used.

In addition, Race to the Top guidelines and funds for states is a good thing overall.  At least a set of proposals has been generated and states are now addressing the education problems that in the past have been enumerated until one’s eyes glaze over.  No district is asked to choose one over another way to save low-performing public schools.  The models that eventually show the most improvement in student achievement will likely combine several of the many models available.

One sure thing, however, is the chance to revise each state’s testing program.  Keeping in mind the long list of problems with the current tests, it seems valuable to devise a system for the state that will assess the achievement success of students and provide support for learners from the analysis of reliable assessments.  It may be that lots of short assessments (like old-fashioned spelling tests or brief math operations weekly assessments) will turn out to be the most useful.

Anxiety is using one exam a year to label students as well as use that score to evaluate teachers.  A few teachers unhelpful to students may be identified.  However, if the school does not receive the resources to improve, what good is it to castigate a particular school, its teachers and students?

Here is where small grants like those saluted in the current issue of the California Teachers Association magazine California Educator are important as well as financially well-liked in a state with a continuing budget crisis.  Teachers can develop a program that suits their own school’s difficulties, then apply and receive a grant to implement the plan.  Of course, concerns arise like does the small plan allow replication, does it become an institution for the school, does the entire school support the plan.

The struggle is faced in California as well as states all over the country: teachers must be accountable, the latest term for being responsible in elementary school for the success of 20-30 students a year.

A system of testing, if it doesn’t assess what teachers are being asked to do, is going to be seen as an obstacle, something to defend against, so that it takes up a lot of thinking time that one would hope was being used for instruction.