During the 2008 school year, Silvia Bermudez, a teacher in a DPS elementary mixing dual-language instruction and a “modification” of the Montessori Method, attended a meeting with parents, teachers and the school principal, JT. The principal was proud of the CSAP scores which, in her opinion, proved that “our school’s program is a success. In fact, we are outscoring schools such as Dennison Elementary.”
While this news apparently was enthusiastically received by the parents, Mrs. Bermudez was more than a little stunned. She had already examined the CSAP scores of her school and compared them to other Montessori schools in the area, and knew that her school did not have higher scores in any area than Dennison. Of course, she remained silent.
An Hispanic parent spoke up and stated her concerns about her child’s CSAP scores, which were unacceptably low. The principal told the Hispanic parents “not to worry about those scores”, and that “the CSAP tests are inherently racist. Just ignore the results, because the students are doing well in their classes, as you can tell from their grades.”
Mrs. Bermudez, who is also Hispanic, was highly offended by the statement, because it seemed to presume that Hispanics are somehow inferior and incapable of performing the basic analytical exercises necessary to score well on the CSAP tests and other standardized tests such as the AP or SAT tests. It also struck her as ironic that, while the principal accepted CSAP scores when they could be used to show “success” for her program, she blatantly informed the parents of Hispanic children that the tests were “racist” and could be “ignored”.
An additional irony in the situation was that the principal had evidently instructed the department leads that they were absolutely not allowed to give grades at less than “acceptable”. Mrs. Bermudez had students who did not do their work and did not test well in the class, and she refused to change their grades. Sometime after submitting the students’ grades, “someone” in the office changed the students’ grades if any teacher graded them “too low”.
So the principal’s statement that the students’ classroom grades showed their actual achievement was false from the start. Children, who according to the evaluation of their certified and qualified teachers were not succeeding in their classroom responsibilities, could hardly be expected to succeed on the standardized tests. What’s more, the accusation that those students’ failing scores on the standardized tests was due to “inherent racism” in the standardized tests was especially pernicious; they implied that the entire system was biased against the students.
What must those parents think of the system in which they had enrolled their students? Such a statement, made by a trusted figure of authority, could have no other effect than to undermine the credibility of the system and sow seeds of distrust within the Hispanic community.
Her statement is shared by many people and dozens of articles on the subject are easily found on the internet. But are they accurate?
Standardized tests have been around for decades. According to Monty Neil, Interim Executive Director of The National Center for Fair and Open Testing (FairTest), “currently tests are put through various procedures to reduce that kind of bias. This didn’t used to be true; tests from the 1960s were blatantly racist for example.”
ETS, a non-profit organization that scores more than 50 million tests (such as TOEFL and GRE) annually, provides an excellent example of how a test question could appear innocuous to an average middle class Caucasian but could, in reality, be misleading to a minority student.
Consider this analogy question:
A) PEACH: RIPE, B) LEATHER: BROWN, C) GRASS: GREEN, D) ORANGE: ROUND, E: LEMON: YELLOW
The object is to find the analogy that is closest to the original posit. Since a Strawberry is a fruit that is typically “red”, therefore answer E is correct, since a lemon is a fruit that is typically yellow
However, most Hispanics are familiar with limes which are green, rather than yellow lemons. What’s more, the Spanish word for lime is “limón”, which closely resembles “lemon”, so the image that would most likely appear in their mind would be of a green lime, which conflicts with the yellow lemon image. It is therefore very probable that many of these students might choose C rather than E.
While this analogy does not reveal “racism” (rather, it demonstrates a cultural bias), it demonstrates how a simple question could be unfairly clear to children of some ethnicities while being “tricky” to others.
In the early days of standardized testing, few people realized the unfairness in these types of questions. But over recent decades, as awareness grew, the creators and scorers of standardized tests developed complicated review processes to identify potentially biased questions and remove them from the tests. One such review process is known as DIF.
So nowadays, before any test is given to students, the examinations undergo a stringent process to analyze them using focus groups. The results are carefully measured to identify any statistical variant that would indicate that a bias, and any question that might have a bias is removed from the test.
Still, even after cultural biases are removed from the examinations, there is a clear difference between the performance of various groups. Neil explained that “kids from lower economic standards and certain ethnic groups historically tend to score lower. There has been some gap narrowing, but you can see that kids that show up in school show differences in size of vocabulary reading readiness early stages of numeracy, counting and sorting into recognizable sets, depending upon their economic status and ethnicity. Clearly race and class are conflated in this country. So kids of color don’t do so well in these kinds of indicators, as poor kids tend to start out behind. Wealth is an advantage. It may not help you survive in the streets or the woods in a difficult environment, but it’s not a test of ability but of exposure. The tests measure some of that.”
Most importantly, Neil said, is that “a huge part of how a student does in testing relates not only to how they do in school but also what happens in her home or community.”
It should be clear by now that standardized exams are subjected to careful and considerate scrutiny to ensure that they are as equitable as possible. They are not perfect, but since we know that some administrators and teachers are willing to “fudge the grades” in order to make it easy for their students to “succeed” in their classes, some sort of independent, reliable mechanism must be put in place to determine if schools are doing what they should.
The primary failure of standardized exams is that they only prove that the students are able to riddle their way through written tests, and if they cannot, the assumption is that the school has failed. But what is happening in the community and in the homes is as important (or more) than what happens in the classroom.
As we have discussed in previous articles, it does not matter how good your teachers are if they are not supported by their principals. But not even the finest teaching and administrative staff can overcome the disadvantages of some communities without total support from the community, and a willingness therein to try to remedy complex and sensitive social issues. Principals are caught in a devilishly complicated political bind; they must try to improve student performance, but most often they incapable of telling the parents that the reason Johnny is failing is because neither Johnny nor his parents are working hard enough to acquire the skills they need to succeed.
Until we figure out how crack that nut, changing grades and dismissing tests as “racist” will never help the children to achieve real success.