High-Stakes Testing in Our Schools: A New Report from California

Dana L. Grisham

I was galvanized by the May 2001 issue of California Educator, which featured an article entitled “ZIP Codes Shouldn’t Determine Our Students’ Future” (available online). In it, Frank Wells reports on an “exhaustive” study undertaken by the California Teachers’ Association (CTA) to compare various indicators of the highest and lowest scoring schools on the Academic Performance Index (API). The study tells us how students in California are performing but, more important, it informs us of inequities about which we may be unaware.

For those not familiar with the California context, the Academic Performance Index is the heart of California’s Public Schools Accountability Act. Its purpose is to measure the academic performance of the state’s schools. The state has set an “interim” target score for all schools in the state. A school’s growth is measured by its progress toward that score. (For more information see the California Department of Education Web site.)

In California, students are tested every year from grades 2 to 11 using the Stanford Achievement Test (SAT-9), part of the Standardized Testing and Reporting (STAR) program conceptualized by educational reformers. Other components of STAR are not available yet so, for now, the API uses SAT-9 scores as its only measure. Scores are published at the Department of Education Web site.

Figure 1

bar graph image from the California Educator article


Figure 1 shows several bar graphs of what the CTA study found (reproduced with permission from the online version of the California Educator article). Not surprisingly, the greatest disparity between the highest and lowest scoring schools on the API occurred in the area of socioeconomic status (SES) (see the bottom of the figure). In the bottom 10 percent of elementary schools, 94 percent of children attending were characterized as living in poverty, while in the top-scoring 10 percent, only 7 percent of children were living in poverty. The percentages of poor children enrolled declined slightly for middle and high schools scoring in the lowest 10 percent, but stayed at 7 percent for the highest scoring schools.

Another unsurprising finding was the difference in the number of English language learners enrolled in the highest and lowest scoring schools. For example, in the lowest scoring elementary schools, 62 percent of students were classified as English language learners, while in the highest scoring schools, the percentage declined was only 4. Taking a test in a second language is problematic, since the language issue tends to confound the results. Does the child know the content but not have the language to express it? Or does the child not know the content? How much of the test itself has to do with mainstream culture and language? How much has to do with learning English rather than learning content?

Teacher preparation and experience is a third major difference between highest and lowest scoring schools. In the lowest scoring schools, about 25 percent of teachers have an emergency credential (25% at elementary, 27% at middle school, and 21% at high school). A mere 4 percent of teachers in the highest performing schools were not fully credentialed. In November last year, I wrote an editorial in this e-journal that highlighted one of the unintended consequences of class-size reduction in California: that many of the most vulnerable children were being taught by the least prepared teachers as class-size reduction resulted in a movement of teachers from areas of highest need.

The May 2001 themed issue of Educational Leadership addressed the question “Who is teaching our children?” Accountability advocates such as Chester Finn called for more alternative routes into teaching. Others, like Barnett Berry, pointed out that there are no shortcuts to preparing good teachers. Berry also presented evidence that a substantial number of teachers (18% in California) are already “prepared” through alternative routes. There appears to be a little schizophrenia going on when it comes to teacher preparation.

Some states are implementing high-stakes testing at all levels of teacher education. New legislation in California, for example, calls for accountability measures at the end of the undergraduate years. A second accountability measure occurs at the end of the single permitted year of teacher preparation. A third high-stakes assessment is planned for the end of the two new years of induction training.

Even before the new credentialing standards become law, the teacher candidate must also take and pass several tests to earn his or her credential. At the elementary level, for example, candidates must pass the California Basic Educational Standards Test (CBEST) and, unless on a waiver program recognized by the state, the Multiple Subject Assessment for Teachers (MSAT), prior to beginning their teacher preparation programs. Another high-stakes test, the Reading Instruction Competence Assessment (RICA) must be passed before certification can occur. This latter assessment was administered for the first time in October 1998 after every teacher education institution in California had undergone an extensive, high-stakes program review by the California Commission on Teacher Credentialing. The objective of the review was to ensure that RICA standards were being taught.

At the same time that credential candidates were being asked to demonstrate their suitability to teach via these assessments, other people were teaching on emergency credentials. These people had only to possess an undergraduate degree and pass the CBEST. Needless to say, many of the people with emergency credentials wound up teaching in hard-to-staff schools, usually in high-poverty areas or the inner city, as the figure above shows.

An interesting relationship between traditional and year-round school schedules is shown in Figure 1. This relationship came as a surprise to me, though it probably shouldn’t have. Fifty-eight percent of the lowest performing elementary schools were on year-round schedules, while only 3 percent of the highest performing schools were on such schedules. I began to think about why this might be. The twin specters of power and money emerged. California has an abysmal record when it comes to funding schools, as shown in Figure 2 (reproduced with permission from the Education Data Partnership).

In times of scarce resources, parents who can afford them provide additional resources for their children’s schools. Schools with older and more crowded facilities are probably not in the more affluent areas. Not shown in Figure 1 but mentioned in the CTA report is the size of the schools in the top and bottom 10 percent. Average enrollment in the elementary schools in the lowest 10 percent was 568, while in the top-scoring 10 percent it was 411.

    graph of California's expenditure per each K to 12 student

As stated in the CTA report, it’s easy to see where the lowest performing schools are concentrated. Fifty percent of the schools scoring in the lowest 20 percent on the API were in rural areas of the San Joaquin and Imperial valleys (agricultural regions with many migrant workers), while the other 50 percent were in the inner cities of San Diego, Long Beach, Los Angeles, Oakland, and San Francisco.

Finally, I looked at the mobility percentages and tried to figure out why they were not as different as one might expect. The lowest scoring 10 percent of elementary and middle schools had 20 percent of their students described as highly mobile; at the high school level, 15 percent of the students in the lowest scoring schools were highly mobile. Mobility at the highest scoring schools was lower, but the differences were much less glaring than on other measures. Could it be money and power again? Do low SES families move because they must? Do high SES families move because they can?

Talk of leveling the playing field has been around for as long as I can remember, but it is mostly talk with little action. It’s much easier to test than to teach. It’s also much easier to administer tests and then blame the victim than it is seriously to address the issues that make inequalities stand out so starkly.

I invite your comments! Join our Online Communities and share your ideas and experiences on these issues.

To print this editorial, point and click your mouse anywhere on its text; then use your browser’s print command.

Citation: Grisham, D.L. (2001, July/August). High-stakes testing in our schools: A new report from California Reading Online, 5(1). Available: http://www.readingonline.org/editorial/edit_index.asp?HREF=/editorial/july2001/index.html

Reading Online, www.readingonline.org
Posted July 2001
© 2001 International Reading Association, Inc.   ISSN 1096-1232