Lessons on standardized testing

This post is a response to the article Help Teach Redford a lesson about standard testing that was written by Licia Corbella who is a columnist and editorial page editor for the Calgary Herald. I'm sending this to the Calgary Herald. I'll let you know what they do with it.

Licia Corbella begins her article with a story where she used her school's grade 3 Provincial Achievement standardized test scores to discover that her children's grade 3 teacher was missing an exorbitant amount of work. At first glance, one might say, "good for her", but when you stop and think about it, you realize that the test scores played far less of a role than Corbella would like you to believe.

Corbella claims that she used standardized test scores to identify an important problem at her children's school, but the truth is she found out by calling a friend who lives down the street, who Corbella admits "always seemed to know exactly what was going on." What are the odds that this neighbour is so enlightened because she is synthesizing stanines, quantifying quartiles and analyzing averages by pouring over test score data? Is it not far more likely that this plugged in parent actually visits her children's school to see and hear what is going on?

Is it possible that if Corbella spent more time at her children's school, talking with teachers, asking questions of administration and volunteering her time in some small way that she might not need to spend so much time looking at test scores? And if a parent needs a standardized test to tell them about their child's learning, maybe the problem is not the teacher?

Standardized test scores are only needed to judge a school if you never intend on actually visiting the school, and if you don't ever intend on taking the time and effort to visit the school, then you have no business judging it.

While I agree we should do everything in our power to improve bad teachers or remove them, assessment is a tool for students and teachers, not for governments; this is true for many reasons but perhaps most importantly because standardized tests were never designed to reflect the educational quality of a school. If this surprises you, then quite frankly, you don't understand these kinds of tests enough to be using them in an argument.

The Fraser Institute are test score data mining suits who sit in their faraway offices claiming to be distant authorities. Corbella has to know that the Fraser Institute's work is built on the premise that everything that goes on in a school can be reduced to a single score from a single test on a single day.

Anyone who calls the process of reducing the messiness of teaching and learning to a tidy test score a "public service" is neglectfully ignorant, willfully blind or outright lying.

The Fraser Institute and Licia Corbella's collaborative shrug at the burden of anxiety that our children suffer from under these tests tells us less about the state of our classrooms and more about the Fraser Institutes' ignorance toward the cancerous effects standardized tests have on real people.

Test anxiety has grown into an entire subfield of educational psychology and its prevalence means that the tests that produce this reaction are not giving us a good picture of what many students really know and can do. But to be fair to the likes of Peter Cowley from the Fraser Institute, he is too busy "number crunching" to properly empathize with children.

While it is true the provinces that have standardized testing outperform those provinces that do not, it's important to understand that just because two things are associated that does not mean that one caused the other. Correlation doesn't prove causation -- in other words, the amount of mosquito spray purchased at the corner store will be highly correlated with the probability of mosquitos, but the presence of mosquito spray didn't make the mosquitos. The same goes for the presence of standardized testing and learning.

To understand how successful education policies can exclude standardized testing, we would all be better off if we understood how Finland can have a world renowned education system with almost no standardized tests (they have one that takes place before post-secondary and not every student even writes it).

The old saying "if it matters, measure it" is an interesting one. As an educator, I often ask parents what are their ultimate goals for their children, and I almost always hear that they want their children to be happy, creative, responsible, honest, caring, ethical, humorous, flexible, enthusiastic, courageous and resilient.

If we truly measure what matters, I would like to know where are the standardized tests that measure these things.

The bane of reducing learning to a test score is that it inevitably overvalues whatever can be quantified and undervalues what cannot because a correct answer on a test does not necessarily signal understanding and a wrong answer doesn't necessarily signal its absence.

In life we are often satisfied by a great number myths. One of those myths is that standardized tests act as a window into the classroom by telling us how good our teachers teach and how effective our children learn.

Many people may not know that researchers have come to see that high scores on standardized tests have almost everything to do with parental socioeconomic backgrounds and less to do with teachers, curricula, or what the children learned in the classroom.

Standardized testing victimizes the very people it claims to serve by narrowing curriculum in a way that  often encourages the worse kinds of teaching (test preparation) and shallowest forms of learning (memorization). Because standardized testing cannot tell us what we need to know about our schools, they simply don't make the grade, which provides Alison Redford with all the mandate she needs to rightfully do away with the grade 3 and 6 Provincial Achievement Tests.