By John H. Horst
It’s an annual ritual I usually do not look forward to. I cannot necessarily say I looked forward to turning in my letter exempting my high school junior from standardized tests, but at least I had the sense this year that I had an ally in San Diego Unified School District Superintendent Cindy Marten.
She recently sent home a letter explaining her position on these tests, their relationship to Common Core, and the procedures for opting out. This was met with controversy, mainly from groups concerned that the school district be held accountable for the progress of minority and lower income students. They look to test scores to indicate this progress and equality.
Teacher performance, of course, is the other side to the testing debate, with conservatives generally calling for teachers to be evaluated in part of the basis of student scores. Conservative agreement, however, seems a bit muted due to larger concerns with Common Core.
As a parent, I’ll articulate here why I support the superintendent and believe San Diego parents should uniformly opt out of these tests.
The Statistical Norm vs. the Cultural Norm
Back when my high school junior was in third grade we took him to the public library. He loved reading and was eager to see what they had to offer. We asked for the third grade reading list.
We were shocked.
To the last one, the books on that list were books our son had read in first grade. We had to go up to the fifth grade list before we started seeing books we felt were appropriate. Our son is not an academic prodigy. Like most kids he does well in some classes and struggles in others. We would expect his reading skills to be slightly above average. But if that reading list was any indication, he would have been somewhere in the 95th percentile or higher.
A ‘standardized’ test is a test which has been designed to produce a “score spread.” If you were to map the scores as points on a grid and roughly draw a line through those points the lines would form a bell curve — or the “standard deviation.” If you draw a vertical line through the very top of the curve you have the statistical “norm.” Then each point, or score, is in a certain percentile compared to that norm. The problem starts when that statistical norm does not correspond to what we might call our cultural norm.
My reading list story above is an example. If that list reflected a statistical norm for third grade reading skills then it bears no resemblance to what our family might otherwise feel to be the cultural norm for reading skills. So at that point, if you were to show me test scores for my son back then which showed him to be the 95th percentile I would just shrug and say “So what?” He would have been a third grader in the 95th percentile against a norm more appropriate for a first grader.
Or to quote Hillary Clinton: “At this point, what difference does it make?”
Evaluating Quality in Education
The tyranny of the statistical norm also affects what we believe about quality in education. Here it is important to understand why these tests were designed to begin with. In World War I the Army was taking in recruits for the war effort by the thousands. They needed to decide which recruits would go off to Officer Candidate School (OCS). So they commissioned a test — Army Alpha — to establish a statistically valid norm so they could rank each recruit against the norm and send the highest scoring recruits to OCS. The rest would fill out the enlisted corps.
Colleges and universities realized immediately that they had an identical problem: Say the school had 100 seats for incoming freshmen, but had 1,000 applicants. They would need an objective way to identify the top 10 percent — or the 90th percentile in test-speak — for admission. Standardized testing was born.
It is crucial to recognize that identifying a percentile against a statistical norm is not the same (indeed, it is not even close) as evaluating how effective a teacher, school or district is in teaching to state standards. The former is called “norm referenced” testing, the latter “criterion referenced” testing. Or to put this another way — to point at where our kids are scoring against a statistical norm and then to make claims about how well our schools are doing on that basis is a logical square peg in a round hole.
Test Scores vs. ‘Thought Leadership’
In 2006 Newsweek’s Fareed Zakariah wrote a compelling article which noted that the same students (in this case from Singapore) who score the highest on international standardized tests are nowhere to be found later on in life when you look for “thought leadership” in vaunted fields like science, technology, engineering, and math. Singapore’s Minister of Education Tharman Shanmugaratnam explained:
“We both have meritocracies. Yours is a talent meritocracy, ours is an exam meritocracy. There are some parts of the intellect that we are not able to test well — like creativity, curiosity, a sense of adventure, ambition. Most of all, America has a culture of learning that challenges conventional wisdom, even if it means challenging authority. These are the areas where Singapore must learn from America.”
This should force us to ask some tough questions. Ultimately, when pressed to justify Common Core, its advocates will point to the need to remain competitive in the world economy and will point to test scores as the indication that we have a problem. The question, though, is why Common Core advocates more broadly — and testing advocates in particular — believe these tests tell us anything useful about being competitive. If they do, why is it that the highest scoring students are noticeably absent later on among thought leaders?
I have some personal experience that helps me understand this. I received both my bachelors and masters degrees in the Philippines. The schools were both great schools, but the culture of the Asian classroom is not one that fosters open debate. To disagree with the professor is especially thought to be a sign of boastfulness. Shanmugaratnam rightly sees this as fundamental — we have an educational culture that rewards challenging the conventional wisdom and even disagreeing with the “authorities.” Asian education (and I realize this is a broad brush) is driven more toward scores on high-stakes tests (which have a huge impact on future opportunities) rather than developing a love for learning and a willingness to try things everyone else says cannot be done.
Standardized tests — and Common Core more broadly because it is all driven by test scores — will only push us away from this educational culture which produces thought leaders in favor of a culture which produces — well — test scores.
Ms. Marten is right about our schools: Their product is not a test score. Let’s keep it that way by opting out of standardized testing.