Tom McMorran was named Connecticut’s principal of the year in 2012. Here he offers a lesson to our nation’s politicians about the Common Core standards and high-stakes testing. Send this to your state legislators and your member of Congress and the Senate.
Tom sent the following comment:
It is time to school our politicians about CCSS and High-Stakes testing.
Here is a day in course level 101.
2012 High School Principal of the Year NASSP
In order for an argument to carry weight and cause one not only (1) to believe it, but also (2) to take action based on that belief, the argument must have warrant. There is nothing subtle here. The weakest form of argument is some version of “I am in power and I say so…” Or, in any teen’s mother’s words: “Because I am the parent!”
When the person presenting the argument relies on some authority to shore up his/her argument, then we have a duty to test the reliability of the authority. In philosophy or rhetoric or simply argumentation this is known as an appeal to authority.
Last week Gina, Mary Ann, and I attended another workshop at the Connecticut Association of Schools (CAS). This is the body that is, in theory, an institution that is independent from the State Department of Education. The presentation was made by Dr. Diane Ulman, who is the Chief Talent Officer at the DOE. She was appointed by Commissioner Pryor.
As part of her presentation, Dr. Ulman reminded us that the Governor’s Council, The Gates Foundation, a range of other foundations and 46 states have signed on to CCSS. In other words, she offered an appeal to authority. Now, for an appeal to authority to work, credentials must be established. And any group that has a personal, financial interest in public policy must make their bias known. So, let’s ask a very basic question: Where’s the money? For Pearson, Houghton Mifflin, and other publishing companies the prospects are enormous. Smarter Balance, the private, for-profit company received half a billion federal dollars to develop the next generation of assessments, which will replace the CMT and CAPT and be administered in about 26 states. You may recall the President’s State of the Union Address; he all but bragged about the 4.3 billion for Race to the Top (RTTT) funding, and how it was amazingly inexpensive for the Federal government to get these 46 cash-strapped states to sign on.
So, when you hear the proponents of the Common Core State Standards and High-Stakes Testing appeal to authority, you have a duty to weigh the degree to which the authority has sufficient warrant to be believed. Here, let me try it: Elvis is still alive. Evidence? 50 million Elvis fans cannot be wrong.
Before meaningful inferences can be drawn from any data set, the researcher has a duty to ensure that the social phenomenon under consideration has not been conflated with other factors. In other words, if you want to give a test that measures the contributions of a teacher to a student’s growth, you must account for and guard against any other factor that might conflate with the primary inquiry. It works like this:
1. We want to know if the teacher’s skill as a reading teacher leads to observable reading skills in her/his students.
2. Therefore, if we give all students the same reading assessment, we should be able to conduct a comparison between teacher A’s students and teacher B’s students.
3. From that comparison we can tell if one teacher is better than another at teaching reading.
So, what’s wrong with that?
A. If the assessment was designed to measure student performance, it can only be used for teacher evaluation by an act of hopeful extension. If the assessment had been designed to measure teacher performance, then it could only be used to measure student performance indirectly.
B. In order for teacher A to be compared with teacher B, the context for all potentially confounding factors for the experiment must be the same. In other words, the only factor that can be measured is, in this case, reading.
But wait, Tienken, Lynch, Turnanian, and Tramaglini have something to say about this in “Use of Community Wealth Demographics to Predict Statewide Test Results in Grades 6 & 7.”
Here’s the very short version: If you tell these researchers three out-of-school demographic variables, then they can tell you a New Jersey school system’s 6th Language Arts scores on the New Jersey Assessment of Knowledge for grade 6 (NJASK6). Tell them (a) the percentage of lone parent households in the community, (b) the percentage of people with advance degrees, and (c) the percentage of people without a high school diploma, and they can plug those data points into a formula that will predict the scores within an acceptable range.
If confounding factors such as a town’s wealth are predictors of performance, then how can we use a reading assessment designed to measure a student’s performance in order to decide whether or not a teacher has effectively taught the skills or knowledge measured by the test?
Here is another wee complication: In New York the APPR rating system that is a year ahead of Connecticut’s uses a growth over time model, which sounds great. But, if you are the unlucky teacher who earned the highest rating in your first year and then for some reason you “slipped” to proficient in your second year, you have not shown growth over time, have you?
The foundation of the CCSS argument has been negative comparisons between international assessments of 15 year olds in which Americans appear to come out near the middle of the testing range. The argument runs like this: The future economy needs 21st Century Skills. Other countries are out-scoring us, therefore the strength of our economy is threatened over the next few decades.
But, if we recall our faculty reading of Yong Zhao’s Catching Up, or Leading the Way, we recall that there is an inverse relationship between performance on a standardized international assessment and productivity over time. Yes, that’s right. The same group of 14 yr olds who came in dead last in the First International Math Study (TIMS) is now a group of the 60-somethings who control the American economy, which is still rated among the top three most productive economies according to the World Economic Forum.
So, to make the international comparisons look bad, the proponents of this argument have to place the USA into a comparison with the 58 countries for which there is competitive data. Yikes, it looks like the mid-21st century will be dominated by Bulgaria; didn’t see that coming, but that’s what the tests show. If, on the other hand, one compares the US to the G-20 or G-7 Economies, the negative comparisons cease to be statistically valid.
Also, let’s just pause for a minute here and consider the PISA study of 15 year olds. You have to be 15 to take the test. So, if an American kid averages 170 days of school attendance a year, and among those days are mid-years, finals, and field trips, then let’s say there is a good chance for 140 days of instruction. But Asian countries regularly offer up to 240 days of school, so let’s knock off twenty and call it 220. Should an American student be able to compete with his/her counterparts in math? Well, actually, even on the much-vaunted PISA fully one out of four students performing at level five, the highest level, is an American.
So, if we follow the scores-to-economics argument, we would be likely to engage in behaviors that promote success on a test, but this will lead to lower creativity and productivity in the adult world!
Campbell’s Law: 1975 “The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social process it was intended to monitor.”
Here is what Nichols and Berliner have to say at the end of a comprehensive examination of NCLB and high-stakes testing: “We are going to do something unheard of in the history of academic research. In this concluding chapter, we are not going to call for more research. There is absolutely no need for new research on high-stakes testing! Sufficient evidence to declare that high-stakes testing does not work already exists.” (2006, Collateral Damage, p. 175).
1. I am NOT saying that we should have no standards. I am not saying that a standards-based curriculum is a bad thing; in fact, I am in favor of it.
2. I am NOT saying that we shouldn’t desire excellence for all students. I am not saying that all students should be able to have meaningful adult lives.
3. I am NOT saying that teachers shouldn’t link their performance to student achievement. I am not saying that we should avoid standardized assessments.
I AM SAYING that the worn out application of so-called hard-nosed business practices (which I do not believe business men or women apply to their own concerns) have any place in a school. I AM SAYING that there is a better way, and it is for all of us educators to embrace our responsibilities as professionals and act from Informed Professional Judgment. I AM SAYING that we can either define ourselves or accept the so-called reform that is happening to us.
It might be that we have to acknowledge and optimistically embrace the following proposition: The High School Structure that has served us so well is not broken; it is obsolete, and it is time for us to transform it!