New State Test Format Should Show Precisely How California Students Measure Up

The credibility of student standardized-achievement test results was shaken several months ago by the report of a West Virginia citizens’ group that surveyed all 50 states. It found that no state was “below average” at the elementary level on any of the six major nationally normed, commercially available tests. This finding suggests a mythical world like Garrison Keillor’s Lake Wobegon, where “all the women are strong and all of the children are above average.”

Americans, struggling to compete with Europe and Asia, are counting on their schools to produce a generation of higher-performing graduates. To judge the progress of this country’s children and the performance of its schools, parents and taxpayers look increasingly to objective external tests. Evidence that these quality indicators are not trustworthy is so unsettling that the secretary of education convened an emergency meeting to consider the West Virginia study.

Here in California, Gov. George Deukmejian’s proposed 1988-89 education budget would restructure the California Assessment Program (CAP) to give a true picture of the performance of districts, schools, teachers and students. The $750,000 appropriation would pay the development costs to convert the CAP into a CAS--a Comprehensive Assessment System that would give school districts the option to replace the current patchwork of tests.

CAP has a high degree of security and credibility, but it does not have national norms and it does not give individual student or teacher results. Most school districts supplement CAP with commercial, standardized tests and therefore inherit their problems of credibility and security. The secret to having all of the children “above average” is in choosing the reference group or norm group to which a set of scores will be compared. For a decade, elementary student achievement has been going up nationally. Because many test norms are 5 to 10 years old, the rising tide has lifted many students above averages of the past. These gains are real, but they beg the question of a current comparison. The parents of today’s sixth-graders want to know how they rank with their peers, not with the sixth-graders of a decade ago.

There is also a choice of norms among types of schools--like urban, suburban and non-public. Some districts or schools may choose a lower-performing norm group to make themselves look better. The type of reference group is usually indicated in the fine print, but many observers never pick that up.

Test security is a problem. When test items become easily available, the likelihood of cheating rises. Instead of teaching the skill of two-digit multiplication, a teacher can teach the answer to 22 x 33. In a state like California, with 10,000 sixth-grade teachers, not everyone will be ethical. Most commercial tests have only two versions, and test copies are readily available. Districts that purchase the tests sometimes have lax security standards, and copies wander through the schools and into other districts.

Deukmejian’s Comprehensive Assessment System proposal would convert CAP from a sample where not every student is tested to an individual test that has national as well as state norms. Both state and national norms would be updated annually. Parents, board members and school personnel could then be sure that comparisons were “apples to apples.”

Another advantage of the Comprehensive Assessment System is the saving in testing time. Many districts give both CAP and a commercial test, as well as a basic competency test required by law. CAS would roll all three testing purposes into one test, saving time and class disruption.

CAS will help the public and state evaluate schools in another important way. Schools with concentrations of low-income and minority youths often find it hard to compete with schools serving middle- and upper-class communities. The experience of doing good work and achieving ambitious objectives yet still ending up on the bottom half of the list is disheartening. To take into account the student clientele, “comparison bands” were invented some years ago. Schools are grouped by socio-economic characteristics and then compared on student achievement. Therefore, it is possible for a school with below-average student achievement to be above its comparison band. This confuses some parents and offends others, who see this lower standard as relegating their students to second-class status in which they are “the best of the worst.”

To resolve this dilemma, CAS offers a “value-added” indicator. By comparing the annual test scores of children who have been in a school for an entire year or more, conclusions may be drawn about how much they have learned. The “value-added” to the fourth grade in School A can then be compared to that of other schools without regard to race or class.

Public confidence in schools will grow when people get reliable data and it shows that progress is being made. Current testing programs encourage confusion and uncertainty about student performance. The governor’s Comprehensive Assessment System proposal is a cost-effective way to create tests that we can count on to tell the truth about the performance of our students, teachers, schools and school districts.