“Yikes,” said Peter Cunningham.

ea7402bc90f01a9ed744ea6088ea944c_400x400

Peter Cunningham.

I don’t normally get into Twitter exchanges. I read ’em. I post ’em. I leave ’em be.

The other day I came across this Peter Cunningham tweet:

I was genuinely puzzled by the “Yikes,” because I couldn’t figure out what the relationship was between parents having the right to opt their kids out from stupid tests like PARCC and low graduation rates or remediation rates.

It was a timely “Yikes,” because the results of Illinois student PARCC tests were just released last week. These are scores from tests given last year. It is now the second trimester for most students in the state. I cannot imagine, nor can anyone tell me what use these results offer other than to be used as punishment.

Which is what I tweeted to Peter.

Peter Cunningham is a former Arne Duncan advisor from Chicago and then the Department of Education and now runs an organization and blog that promotes the standard corporate reform agenda of phony accountability and charter schools.

I was puzzled again. If, as Peter says, Oregon shows low graduation rates and high college remediation rates, what are we tracking by annual testing that we don’t already know?

I think Peter just likes tests. Alas.

As someone involved in education for many years – and by “involved,” I don’t mean Peter was engaged in teaching or spending time in a classroom in front of students – he must surely know that a test designed to understand what students know is not designed to tell us how teachers teach.

That’s not how it works.

Because a student performs poorly on a test does not necessarily suggest that it was the teacher who is struggling. Unless Peter thinks the PARCC is a test of teachers instead of students.

That is exactly the fallacy of value added measures.

7 thoughts on ““Yikes,” said Peter Cunningham.

  1. I happened on your blog and found this post interesting. As a parent I struggle with the notion of teachers and accountability. I agree with your statement “a test designed to understand what students know is not designed to tell us how teachers teach”, but when I step back and think of what I want for my daughters, what is the appropriate level of accountability that a teacher should have to create the learning environment for their students? If the “test” essentially evaluates what my daughter knows, isn’t that directly linked to (not implying completely inclusive of) the learning environment that the teacher created for her? Your statement “student performs poorly on a test does not necessarily suggest that it was the teacher who is struggling” seems accurate on a student by student case, but taken in aggregate (over multiple tests / over multiple years) it would seem to suggest that the teacher is struggling to create a positive learning environment.

    Very interested in your honest thoughts and open dialogue around this…

    • Should we evaluate a teacher on Chicago’s north shore where some of the wealthiest folks live on test results and evaluate a teacher on Chicago’s west side where some of the poorest folks live based on the same test results?

      What if the teachers switched locations? Would the results change? Would the excellent teacher on the north shore still be an excellent teacher on the west side?

      Would a struggling teacher on the west side suddenly become an excellent one?

      Are factors out of the domain of the classroom something worth considering?

      Is the performance on a test by a third grade student in a specific subject area the result of the teaching skills of that third grade teacher alone? What about the teachers who taught that student in earlier grades?

      The teacher is absolutely accountable for the teaching environment in your daughters’ classroom. The problem is in how to assess and measure it. And even there, to what degree should we take into account factors like funding, support staff, physical plant, school leadership, the teacher in the next room, the other students in the class, the age of the building or the class size?

      Look. I don’t believe a single high stakes test given once a year, the results of which are not known until six months into the next year, offers much in evaluating a student’s knowledge.

      Nothing supports using it to accurately hold a teacher accountable.

      Which is not to say the teacher has no accountability.

  2. Fred –

    Thanks for the response. You definitely raise some good points, let me respond and ask a few other questions:

    “Should we evaluate a teacher on Chicago’s north shore where some of the wealthiest folks live on test results and evaluate a teacher on Chicago’s west side where some of the poorest folks live based on the same test results?”

    My un-nuanced answer is yes because the student on the north shore and the student on Chicago’s west side should both have teachers that are impactful. To the point I believe you are inferring, the results in absolute terms should not be compared but I believe the differential growth could be determined and applied to each teacher. My daughter will be taking the MAP test at the beginning of each school year (this is her first year since she is in first grade so my understanding is in its early stages). Based on this year we get the results within 5 weeks. If my daughter shows minimal to no growth, does it matter where I live? To your point, if I lived on the north shore I may be able to afford private tutoring and Mathnasium classes, etc.. which benefits the student in the north shore (and potentially masks the teacher’s ineptness) but shouldn’t affect the evaluation of the growth of students on the west side.

    “Are factors out of the domain of the classroom something worth considering?”

    Absolutely, and factors you mentioned like “funding, support staff, physical plant, school leadership, the teacher in the next room, the other students in the class, the age of the building or the class size” should be accounted for in the evaluation. I’m not sure if your comment is that it is not considered today or it is too complex to consider. I agree it should be considered and disagree that it is too complex to try to consider.

    “Is the performance on a test by a third grade student in a specific subject area the result of the teaching skills of that third grade teacher alone? What about the teachers who taught that student in earlier grades?”

    Excellent point. To my earlier point about differential growth above, wouldn’t this be accounted for? I understand that has you progress in higher grades an assumption of foundational knowledge is required, but to throw your hands up and point upsteam that it was your colleagues fault seems more like lack of accountability rather than ownership of the challenge.

    Look. I don’t believe a single high stakes test given once a year, the results of which are not known until six months into the next year, offers much in evaluating a student’s knowledge.

    OK, but what offers more in evaluating a students growth over the school year? and why are the test “high-stakes”? In my opinion, they seem to be an consistent evaluation of a student’s mastery on specific subjects at the appropriate grade level. I wouldn’t imagine they high stakes for teachers who know they have created a learning environment for their students. Maybe I am being naive here, but if a teacher is being impactful wouldn’t the test show that?

    Which is not to say the teacher has no accountability.

    I guess I still don’t understand what teachers are accountable for? When I have parent teacher conferences, what should be within the realm of discussion (i.e. what are they accountable for) and what should be outside of it? How do I contextually think about my daughter’s classroom to have a collaborative conversation with her teacher in the scope of his/her accountability?

    • There are generally speaking two kinds of tests. One is a formative assessment. If, as a teacher, I am doing instruction on one skill, say one-point perspective, I come up with an assessment measure which I can use to see if my students are getting it. Or should I go back and go over it some more. Then there are summative assessments. These are assessments done to see whether my students grasped the central essential topic we have been covering. Or a years worth of work.

      In order for either kind of assessment to be useful they must have both validity and reliability. Validity means did the test measure what it was supposed to measure. Reliability means did the test measure that consistently over time and among students.

      A test that is intended to measure a student’s understanding, whether formative (like a spelling test) or as a summative measure, like the PARCC, should not be used to measure something else, like the effectiveness of the teacher. It wouldn’t be valid since it was not designed for that purpose.

      Evaluating student growth over a period of time is not only – or mainly – measured by a metric or a score. If we reduce teaching and learning to points on a scale, we reduce both its complexity and its value.

      Do we look at our children as if they were measuring cups? Yours is holding half a cup. Mine is holding three quarters of a cup. Then we measure again at the end of the year.

      My children, who are grown and gone (I have two grandchildren now) and I am proud to say are teachers themselves, were way more complex than that. As are yours. And I never would have wanted them to be seen as growth scores.

      At parent/teacher conferences I would ask the teacher, “What is my child accountable for? What are you accountable for? What am I accountable for?

      It is the power of the relationship between you, the teacher and your child that is the ultimate basis for everyone in that relationship’s accountability.

      As to your first point. I agree absolutely. Whether a teacher teachers poor kids or rich kids, we should expect the highest level of skill and dedication. But those skills must be the best fit for the children and community in which they teach. And since those communities can be very different, with different needs, one size can never fit all.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s