The Case Against TAAS

K-12 Testing

The first major legal challenge to the current wave of state graduation tests went to trial in federal district court in San Antonio, Texas, early this fall. The Mexican-American Legal Defense and Education Fund (MALDEF) was attempting to prove that requiring students to pass the Texas Assessment of Academic Skills (TAAS) in order to obtain a diploma discriminates against Latino and African-American youth (see Examiner, Fall 1997).


The TAAS graduation exam is administered beginning in grade 10. Students can re-take any portions of the test that they do not pass. Other versions of TAAS are administered in elementary and middle schools, where they will soon become grade promotion hurdles.


Testimony and written reports submitted by plaintiffs’ expert witnesses proved that a far lower percentage of African American and Latino students than white students graduate. The TAAS increases grade retention, drives up the dropout rate, and increasingly controls teaching and instruction, turning many schools into test coaching programs.


In one report, Professor Walt Haney of Boston College demonstrated that the gap in the pass rates between white students and minority students is large enough to be defined as a “racially discriminatory impact” under federal civil rights law.


Haney also showed that the graduation rate for Latinos and African Americans dropped significantly, from about 60 percent to about 50 percent, after the state implemented the TAAS graduation requirement. After eight years, it still has not returned to the previous level. While the white graduation rate also declined initially, it rebounded to its previous level of 70 percent within a year. Each ten percent gap represents about 100,000 black and Hispanic students who do not graduate.


Haney reported that a review of State Board of Education minutes revealed that the Board chose the passing score arbitrarily, ignoring accepted professional measurement procedures. The score that was set had the effect of maximizing the adverse impact of the TAAS on black and Hispanic students.


Haney added that the use of cut-off scores to determine graduation is a violation of the Standards for Educational and Psychological Testing. Standard 8.12 states, “In elementary or secondary education, a decision or characterization that will have a major impact on a test taker should not automatically be made on the basis of a single test score.” The recent report from the National Academy of Sciences, High Stakes, takes the same position (see Examiner, Fall 1998).


Because the TAAS is used to hold schools accountable, schools do whatever they can to post higher scores. One effect, according to Haney, is that retention in grade 9 drastically increased as schools attempted to improve their grade 10 scores. In theory, retained students will do better on the TAAS with an extra year of schooling. Some may also drop out, thereby reducing the number of students who are likely to fail the test.


Haney also questioned the TAAS tests’ validity. He noted that the correlation between TAAS reading and math test scores was higher than that between the reading and writing tests, a very unusual result. In addition, there was a far lower correlation between student grades in English II (grade 10 English) courses and TAAS than researchers usually find between classroom grades and exam scores. Over one quarter of African American and Latino students who passed English II failed the TAAS reading test in the same semester, compared with only ten percent of white students. A slightly smaller disparity is evident on the writing test. As Haney points out, this could be due either to test invalidity or to a failure to provide the students with a fair opportunity to learn — or both. Either conclusion bolsters MALDEF’s claims against using the TAAS as a graduation requirement.


Haney and other researchers also found that the TAAS had damaging consequences for curriculum and instruction. (The Examiner will report on these findings in the next issue.)


State’s Defense
The state tried to prove that the score gaps between groups did not reach the legal level of “racially discriminatory impact,” and that in any event the test is necessary. They also argued that the test is valid and that the cut score was properly set. Holding more students back, the state claimed, is educationally beneficial since it means the students are getting an increased opportunity to pass the test. From this perspective, the lower graduation rate is simply an artifact of the additional time it is taking Latino and African American students to pass the tests. Moreover, the state contends that the racial gap in passing rates has been closing while the passing rates for all ethnic groups have been rising.


Even if the plaintiff’s powerful evidence about the damaging and discriminatory impact of the test does not prevail legally, advocates should use this information to debunk the myth of great improvement in Texas schools. For example, Texas reading results have not improved by even one point from 1992 to 1998 on the National Assessment of Educational Progress (NAEP).


Despite this dismal NAEP performance, on top of the discriminatory impact and curriculum damage from TAAS, Texas has emerged as a “poster child” for high-stakes testing for students and schools. Texas Governor and presidential candidate George W. Bush has even bragged that the TAAS is the antidote to the discrimination of low expectations. But as Haney and McNeil show, the TAAS institutionalizes low expectations, guaranteeing cycles of failure, as seen in dropout and retention figures. These issues must be taken to the public, as well as fought in court.


The case is GI Forum, et al., V. Texas Education Agency, et al., CA No. SA-97-CA-1278EP.