It Really IS All About The Tests Alice
What does a good education look like? The answer often depends on who you are in the debate. If you’re a politician or education reformer you will tend to focus on defining what the outcome of the public education system is supposed to be. If you’re an educator you will focus on the way you teach and the amount of freedom you have to teach that way. If you’re a parent you will focus on whatever answers get your child to take in the lessons as school and generally be happy about going to school. That’s why there is often little consensus on the answer to the question about a good education. Because we are becoming a data-driven society, there is a great push to try to use data to answer the question. Finding and interpreting that data, however, is very difficult. There is one element of the debate where there is a coalescing consensus. When it comes to the issue of using standardized testing to obtain this data, there seems to be general agreement that less testing is better. People fall along a continuum in their responses but they seem to be trending toward to zero end of the scale.
Alice Walton wrote in Forbes magazine this month about the developmental inappropriateness of Common Core. She included an examination of the trend toward testing younger and younger children. The experts seem to be claiming that, if we are falling behind other countries, then teaching content and skills sooner and measuring if our teaching is working will catch us up. Unfortunately there is no research to support this claim and plenty of research to support the benefits of play and less structured teaching for younger children instead.
Carol Burris, named 2013 NY High School Principal of the year had this to say in Forbes, “Kids are supposed to enjoy elementary school. There’s never been a time where we’ve had the need for psychological and social support services as we do now, and the Common Core is only going to exacerbate the crisis of over-stressed students, who struggle emotionally day to day.”
When it comes to standardized testing there is a growing trend for teachers and parents to want less of it. Parents have been pushing the right to opt out of testing. Teachers have been resisting testing because of the amount of teaching time that is lost to it, but also because of the high stakes consequences that have been attached to it. Whatever the reason, there is agreement on this issue – less is better.
Psychometrician (a fancy word for test developer) Robert Linn understands the limitations of standardized testing better than most. He has studied the impact of the increased use of such testing since the 80’s and sees a predictable trend in the results. Whenever a new set of standards and associated tests are introduced, student scores start out lower than they had been on the last set of tests. As everyone gets more comfortable with the new tests, scores go up. People generally assume, because school administrators push this idea when they happily claim responsibility for these increases, that the higher scores indicate an increased student understanding of the content or skills. Typically, scores on tests of the same content but not directly linked to the new standards do not show a similar increase.
Follow-up studies (e.g., Linn, Graue & Sanders, 1990) noted that that although states and districts had substantial gains in student test scores in the first few years following the adoption of a test leading to the Lake Wobegon* effect, there was almost always a large drop in the average test score the first year that a new norm- referenced test was adopted. This pattern provided evidence that the gains were spurious and due to familiarity of teachers with the specific test. The test score gains did not generalize to the broader achievement domains the tests were supposed to represent. An important implication of the Lake Wobegon experience is that the same test cannot be used several years in a row in an accountability system without having a serious score inflation problem.
* Ref Garrison Keiler’s program whose famous phrase is that in Lake Wobegon everyone is above average.
This is the teaching to the test that many people decry.
However, I have heard some people say “So what? If that’s what we want kids to know why wouldn’t we teach to the test?”
Linn points out the problem with that line of logic.
“State tests tap only a fraction of the content specified in state content standards. Some content standards may be too difficult or too expensive to measure on a state test. In addition, scores on tests that are administered by states each year may be inflated due to narrow teaching to the test content rather than the broader domain of the content standards (Haut & Elliott, 2011). Several studies (e.g., Chadowsky & Chadowsky, 2010; Jacob, 2005, 2007; Koretz, 2002; Koretz & Barron, 1998) have found that state increases on low-stakes tests are smaller on than the increases reported on high-stakes tests used for holding schools accountable. These results raise serious questions about the validity of the interpretations of educational test results from tests used for purposes of school accountability (see, for example, Baker & Linn, 2004).”
Basically, the tests only show the content knowledge or skills that are easiest/cheapest to measure and are not a good measure of a student’s total understanding. In addition, once these specific limitations become known to teachers, they learn to focus only on the content/skills that they know will be measured by high stakes tests, at the expense of additional content knowledge not measured or measured on low stakes tests. Their students’ scores will rise on the high stakes tests which gives districts bragging rights and makes uninformed parents happy, but no one will recognize what has been lost in teaching.
This is another argument against a common set of national standards. As Patrick Deneen Constitutional Law Professor at Notre Dame University said,
“In our modern push to standardize and make uniform and to equalize, we necessarily must end up discarding any higher aspirations of education‘s end in an embrace of what can only be widely secured agreements about lower and debased ends, an education based upon the lowest common denominator… So given this strong effort at every level to link education with career readiness measurable by money making we see clearly how a basic utilitarian mindset now dominates the definition and understanding of education and how thereby it constrains, limits and narrows the scope of education’s purposes solely to the debased end of work.”
If we want kids to be life long learners, we should constantly reiterate the point that there is so much more out there to learn. What standardized test scores tell them is, “you already know everything we expect you to know.” What high school and college diplomas and standardized tests teach them is that the only learning that matters is that which someone else can measure and certify that you have mastered. And for that privilege you will have to pay. None of these processes will produce kids who grow to be life long learners.
Walton quotes Mindy Kornhaber, a professor at Pennsylvania State University who has been studying Common Core and who understands, as the teachers in many school districts like mine do not, that the standards and testing go hand-in-hand.
“The Common Core standards are in fact supposed to be tested with the Common Core tests that are being produced by the testing consortia… A perceived threat to the Common Core reform (and I have this from a Common Core insider) is the willingness of a number of states to abandon those testing consortia tests.”
We will see the predictable saw tooth pattern of lowered results followed by rising scores until a new test aligned to modified standards is introduced, and then scores will drop again. What we won’t see, because the tests aren’t looking for it, is what content knowledge and skills we have stopped teaching and how that loss will affect the lives of students beyond the years of formal schooling.