Courtesy: Granada Hills Charter
Students learn about polynomials in Janie Dam's Algebra 2 class.

Do you count on EdSource’s education coverage? If so, please make your donation today to keep us going without a paywall or ads.

The overtesting of students increasingly has become one of the political battlefields in our public school system.

It sat among the five concerns listed on the United Teachers of Los Angeles’ Beyond Recovery Platform. Yet as reported by EdSource recently, we have a clearer understanding of the impact of the Covid-19 pandemic on students’ academic progress thanks to California Smarter Balanced and National Assessment of Educational Progress (NAEP) standardized test scores. The alarming drops in both math and English language arts performance reinforce the importance of having objective student data available to clearly identify the gaps in learning that need to be closed.

Standardized tests have been with us for thousands of years. The first recorded occurrence can be found in ancient civil servant selection, as early as 2200 B.C. Among things candidates for government jobs were mandated to demonstrate was proficiency in archery. Applicants were handed three arrows to shoot at a human-sized and shaped object while riding on horseback. If all three arrows hit the target, a perfect score was earned; two reaching the target was graded as good; and if only one arrow met the target, it was a pass.

For the past century, standardized testing in the United States came in the form of hours-long multiple choice and essay exams, which could be just as arduous (and with acceptance to college primarily riding on the scores, perhaps just as gruesome). We have now come to recognize that these high-stakes evaluations have fundamental flaws: gender, wealth, and racial biases and the subversion of results through expensive prep courses that un-level the playing field; even parent-funded cheating.

Newcomers to the country tend to start conversing in English before they learn to write in the language, but standardized testing for the most part does not come in a voice response format except for the language exams themselves. The inability to provide equitable access to high schoolers with special needs during the pandemic led to court cases and fast-tracked a movement away from College Board examination that was clearly coming and in many cases clearly needed. Many colleges and universities, including the University of California system, are jettisoning both the SAT and the ACT as admission requirements.

But this evolution leaves us with some unavoidable questions: What lies ahead for data collection to track state and national progress toward a better-educated workforce? Should all standardized testing be discarded in the fight against unhealthy overtesting and high-stakes pitfalls that we’ve come to understand? There are projections of a shortage in America of millions of highly educated and skilled workers, and global economic competition is likely to continue its steady rise. It would be foolish to deprive ourselves of a consistent means to gauge how our education system is serving our young people — and by extension us all.

The answer to student overtesting is not to throw out all standardized tests, but rather, to streamline and enhance the assessment process.

The onsite Smarter Balanced (or an equivalent) should be able to stand alone as an instrument to meaningfully measure both capacity for higher education and future employability, whether via college attendance or vocational training. This action also serves to eliminate ancillary exam costs for students and parents who cannot afford them, thereby closing the divide between families who have and those who have not. Transportation and access would be much less of an obstacle when the test site is the school site and testing is conducted during school hours.

Fortunately, advanced education technology such as classroom robots can now make standardized assessments more feasible for both teachers and homeschooling parents. In addition to data-driven machine learning that leads to greater precision and accuracy, artificial intelligence is not burdened by emotions. Interactive assistive technologies have unlimited patience as they don’t experience frustration like human teachers and parents do, such as when children either don’t listen or don’t understand, and may need the same directions repeated many times over; or when grading becomes too tedious and time-consuming.

Machines can be coded to not base scoring on the test taker’s gender, race, ethnicity, physical appearance, spoken accent, etc. They can be built to speak and comprehend multiple world languages, display visuals and glossaries, go from text to speech and back.

But the most important key here is to design test questions and formats as a component of multicultural teaching that celebrates diversity, whether for use on a school campus or at home. The aim should be scaled up test item banks and differentiated, equitable rating protocols to evaluate performance, both of which appropriately reflect the student’s cultural and linguistic background. And that, too, is more doable with the help of artificial intelligence. Indeed, such personalization will pave the way for every child to benefit from an individualized education program, not just kids identified for special education.

Not that anyone would or should want TikTok to dictate their future, but before you scoff, think about it like this: if TikTok can accurately predict which images, songs, videos you will like the most, why can’t algorithms similar to those that power TikTok potentially be relied on to forecast which job tasks and therefore which jobs you might enjoy the most and be most motivated to excel at, and then suggest, but leave it up to you, to select a corresponding course of study.

•••

Janie Dam, is a math teacher and for 13 years has been the school wide testing and data coordinator at Granada Hills Charter in Los Angeles. She has been a professional educator since 2000 and is a member of United Teachers of Los Angeles.

The opinions in this commentary are those of the author. If you would like to submit a commentary, please review our guidelines and contact us.

Do you count on EdSource’s reporting daily? Make your donation today to our year end fundraising campaign by Dec. 31st to keep us going without a paywall or ads.

Share Article

Comments (1)

Leave a Comment

Your email address will not be published. Required fields are marked * *

Comments Policy

We welcome your comments. All comments are moderated for civility, relevance and other considerations. Click here for EdSource's Comments Policy.

  1. Dr. Bill Conrad 1 week ago1 week ago

    Of course, the bigger issue is the widespread assessment illiteracy that exists within the educational community. All assessments have different purposes and different audiences. Much of the rage against the thermometers derives from educators expecting that summative assessments do formative assessment work and that formative assessments do summative work. We cannot, for example, use teacher-made formative assessments as measures for grade level work or college readiness as the teachers may be teaching several grades below level. … Read More

    Of course, the bigger issue is the widespread assessment illiteracy that exists within the educational community. All assessments have different purposes and different audiences.

    Much of the rage against the thermometers derives from educators expecting that summative assessments do formative assessment work and that formative assessments do summative work.

    We cannot, for example, use teacher-made formative assessments as measures for grade level work or college readiness as the teachers may be teaching several grades below level. This is especially true for our children of color.

    We need a grade-level standardized tests to provide us with grade-level performance!

    We should also not expect to use state level standardized tests to diagnose individual student needs as these tests do not use enough items to sample content topics. These tests sample topics across many students allowing the results to be used to gauge how well larger populations are faring relative to learning topics.

    Additionally, assessments no longer gauge student performance based on the number of items correct but rather on the quality and complexity of the items that the student answers correctly. These computer adaptive assessments present items within the zone of proximal development for the child reducing the stress of having the child trying to answer questions well beyond their capabilities!

    Standardized assessments have gotten a bad rap in recent years! The bottom line is that we really do need assessments to help us understand whether our students are on a grade-level trajectory to be successful in college and career.

    Read The Fog of Education!