STEM PALS: Moving Away from Standardized Tests
From Richard Larson
“There exists no magic bullet!” This cliché applies to so many situations, but especially to the complex task of educating our young people. For years now we have been hearing about the need for increased accountability in our education system. And with regard to STEM and k-12, we hear all sorts of ways to increase student performance and interest in STEM careers. We also hear of various ways to evaluate teachers. The notion of evaluation is good. But bad evaluation can be terrible.

In speaking with many high school STEM teachers over recent months, I am convinced that the current “Magic Bullet” trend to evaluate our students based on a single number, a score from a standardized multiple choice test, is a huge mistake. How many of us in careers would feel that we could accurately be evaluated by such a single number? Not many, I presume.

What would you propose to evaluate you and your career accomplishments, demonstrating strong competence in your chosen field? My answer (and yours, too, I guess): a representative portfolio of your accomplishments. That’s right, a portfolio of accomplishments, not a single number. That’s what we would use for ourselves, but why is this not good enough for our children? Is it simply that the ‘cost’ would be too high? That getting a computer-scanned number on a test taken by many thousands is much less expensive for the ‘system’ to process? Friends of STEM, whatever cost savings can be gained here will be swamped by the negative unintended consequences of applying such a Magic Bullet approach.

Teachers have come up to me and said things almost like this, “Dick, we would love to bring into our classrooms compelling STEM lessons to develop critical thinking skills, lessons not in textbooks and not on the mandated test-focused curriculum, but we can’t. I fear that at the next parents’ night, a parent might come up to me and threaten a lawsuit because I spend a class session on such a mind-expanding STEM lesson, not focusing on the next standardized test, and as a result ‘Little Johnnie’ might score 2 points less on the standardized test.” And so the focus is on the test. And this focus can be all encompassing, as many teachers work only on the test days or even weeks before the students take it. How is that for developing critical thinking skills, the type required of STEM professionals?
Then there is cheating. When a single number evaluates a human, there is great temptation to cheat. Some students cheat. Some teachers cheat. We have all read recent news stories – from urban school districts where some teachers erase and correct wrong answers to tony suburbs, where a test-savvy imposter will take an SAT test for someone for the price of $2,500. Magic Bullet solutions almost always lead to cheating, because there is so much at stake and it is so easy to cheat.

There is precedent. Portfolios are currently used to evaluate students. At my own school, MIT, the Department of Architecture requires each student to develop a portfolio of his or her works. It is the contents of such portfolios, not a single number that is used to evaluate each architecture student. In k-12 settings, it is not unusual to evaluate with portfolios a student’s growing abilities in art, music and other ‘creative’ endeavors. But why not math and science, both very creative? Suppose a student becomes fascinated with some of the more obscure derivations of the Pythagorean theorem, researches the subject and writes a paper about it. Doesn’t that deserve to be her portfolio? What if another student does a project on evolving methods to deflect meteorites headed to Planet Earth, allowing humanity to use science and engineering to avoid Doomsday? Doesn’t that deserve to be in a portfolio? What about science fair projects? Student-designed devices such as robots made for STEM competitions? Don’t’ these deserve to be in portfolios? Is a single-number standardized test score a more accurate assessment of such students?

Diversity. Portfolios celebrate diversity. A one-size-fits-all assessment scheme such as the single number test score ignores diversity in students’ interests, capabilities and accomplishments. Which is better: a straight B student or a student with 50% C’s and 50% A’s? They have the same grade-point average. If the A’s of the second student were related to the student’s passion, interest and accomplishment in a subject (say math), then a portfolio showing his math skills and research would demonstrate that. We should celebrate a student’s excellence in a subject. Our letter grade example relates to standardized test scores. Suppose two students get the same aggregate score, say 80% correct. But the first student gets 80% on each of the five sections of the test. The second student gets zero percent correct on one section and 100% on the other four. Do you feel the same about each student? Each got the same aggregate score. The single-number Magic Bullet scheme ignores diversity in interests and abilities within each student. In terms of under-represented groups, the acknowledgement of diverse interests would allow a supportive teacher to note relative strengths and interests in a student and encourage that student to go far, to reach for the stars, even though grades on some other state-mandated parts of the curriculum may be sub-standard. Portfolios demonstrate what a student can do; they do not focus on what she cannot do.

Imagine if we had such an evaluation system for our k-12 students, STEM and otherwise! Imagine if we would celebrate learning, allowing time for self-discovery, for finding one’s own excellence, and moving that forward. Imagine the replacement of test stress with discovery excitement. Every student finds his/her own excellence.

I fear that we may be losing a generation of students to the standardized-test, rote-memorization and regurgitation system. That does not bode well for them or for our future STEM workforce.

Richard Larson is the Mitsui Professor of Engineering Systems at MIT. He is also the Director of MIT LINC and the Principle Investigator of MIT BLOSSOMS.

Back to newsletter