Sunday, January 17, 2016

New Jersey Study Commission on Testing: Keep Weighing the Pig

New Jersey's Study Commission on the Use of Student Assessments issued its report this week after a year and a half of deliberations. The commission was ostensibly formed to look into over-testing in New Jersey schools, but anyone who follows New Jersey politics at all will not be surprised that this commission told Governor Christie exactly what he wanted to hear. That is what study commissions in New Jersey are for.

The report is  a densely written 34 pages long and contains 49 recommendations, so to save you some reading time, Russ on Reading has dug into the report to summarize the findings here.

1. Keep weighing the pig every year

In the introduction the commission acknowledges the "anxiety and fear" surrounding the testing issue, but concludes that this is just so much smoke and mirrors. The report says:

... one point must be abundantly clear: the Study Commission firmly believes all students in New Jersey’s public schools who are eligible should be required to take the State standardized assessment (i.e., PARCC)

And why must New Jersey students in grades 3-8 be required to take this test every year?

Doing so will ensure all students are progressing well in their educational endeavors and all public schools are effective for all students. High-quality assessments such as PARCC will hold schools accountable for serving all of their students, including those from disadvantaged backgrounds. The Study Commission believes it will be impossible to effectively close achievement gaps between and among students without accurate and actionable information.

So there we have it. The same old education reform mantra. We need to keep testing because we need to know how students are progressing and to hold schools accountable for serving all students. In other words the Study Commission has bought into the rhetoric of No Child Left Behind, which the federal government left behind a few weeks ago.

This leads me to ask a few questions.

Are standardized tests an effective way to hold schools accountable?

All the evidence would indicate no. According to testing watchdog FairTest, multiple choice, standardized tests "cannot provide meaningful accountability" information.

Are PARCC tests "high quality?"

This is highly debatable. In New Jersey, less than half of the students who took the test in 2015 scored proficient or above. This would indicate that the test is not appropriately rigorous, but just too damn hard. New Jersey typically ranks second or third in educational attainment across the states, yet half of New Jersey's student couldn't pass the test. You can find my full report on the test results here. I have also done an analysis of readability issues with the PARCC here, here and here.

But don't take my word for it. The Gordon Commission report, produced under the sponsorship of ETS of all people, concluded that Common Core tests are currently "far from what is ultimately needed either for accountability or classroom instructional improvement.

Are standardized tests effective in narrowing the achievement gap?

Of course not. We have twelve years of evidence through NCLB's test and punish policies that the achievement gap is not narrowed by weighing it. Narrowing the achievement gap requires a full frontal attack on inequity. Standardized tests can show us that our society and our educational opportunity are inequitable, but all we really need to do to know that is to look at the neighborhoods in leafy suburbs where kids score well and the long deteriorating neighborhoods in the inner-city where kids struggle to score well.

I have an idea. Instead of yearly testing, let's all just stipulate that the achievement gap exists and that it is in reality an opportunity gap. Then we can do away with all the tests that keep telling us what we already know and focus our attention on things that are likely to reduce the opportunity gap like cleaner, safer schools, wrap around health and wellness programs, and attracting high quality teachers.

2. What we have here is a failure to communicate

The commission seems to believe that the real problem with the testing in New Jersey is a failure to communicate just how wonderful all this testing is and those insidious grumpy pants on social media (including, I suppose, me and other bloggers) who keep spreading nastiness about testing and who fail to grasp the wonderfulness of it all. To that end they make several recommendations for putting new lipstick on the old pig.

The recommendations include creating a shared vision, a proactive communication campaign, a communication team in the NJ Department of Education and a professional email database for the quick dissemination of information. Maybe the NJDOE could enlist the services of a Hollywood public relations firm to make this all baloney look good.

3.  If there is too much testing, it must be the school districts' fault.

The Study Commission seems to not understand what all the fuss is about over-testing, since the Statewide assessments are only done once a year and are mandated by State and federal law. So, if there is over-testing, as many people who testified before the commission complained there was, then it must be the fault of the local districts. To that end the Study Commission recommends that school districts "conduct a thorough inventory of their own assessment systems." School districts are encouraged to make sure their own testing regimes are efficient and used effectively to improve student learning. 

I am sure the districts are grateful for the encouragement and waiting with bated breath for the monies that will support said inventory.

4. Using standardized tests to measure teacher accountability is honky dory.

In the area of teacher accountability and standardized tests, the Commission finds that "the positive and encouraging results of the educator evaluation system thus far, indicate that the system is working." The Study Commission dismisses educator concerns about using test scores for teacher evaluation by saying that much of the concern should be mitigated by the fact that the vast majority of teachers in the first year of the evaluation design were rated as "effective" or "highly effective" and that the ratings were pretty much the same whether student growth scores were used or not.

Once again the Study Commission thinks the biggest problem is a failure to communicate well. If only the NJDOE could get all the good news out about teacher evaluation, people would be embracing it. The Commission calls for professional learning efforts and greater transparency about how teacher evaluations are calculated. They further recommend that the NJDOE "encourage" school districts to use the information for teacher improvement, "particularly of novice and struggling teachers."

The Study Commission might have spent some time looking at all the evidence that shows how the use of standardized test scores in any proportion as a part of a teacher's evaluation is simply unsupportable. They might have read the work of Audrey Amrein- Beardsley (Vamboozled and Rethinking Value Added Models in Education) or looked at the white paper from the American Statistical Association calling into question any significant use of standardized tests as a measure of teacher effectiveness. They might have then said that these measures should be removed entirely from formal teacher evaluation, but that is not the message the Governor wanted to hear.

5. Lots of professional development will help fix the problem

Again, all this noise about standardized testing is simply a matter of misunderstanding. Professional development for teachers, developed under the NJDOE in conjunction with other statewide associations and advocacy groups, will raise the assessment literacy of the teachers in the state and help them to see how useful these standardized tests are. I can see the teachers lining up for these sessions now.

No amount of professional development will change the fact that standardized tests are not very useful for making instructional decisions about individual children because the information is not timely nor does it provide enough detail on individual students. Standardized tests have their place in long range school district curriculum planning, but we don't need yearly tests to get this information.

There is some good news for parents and educators in the report.

1. Special populations of students need special treatment in testing situations.

The Study Commission clearly heard the concerns of parents and teachers of students with special needs and ELL students. They recognize that these students need special accommodations when it comes to standardized testing. They make six recommendations in this area that boil down to acknowledging there is an issue, suggesting the NJDOE do something about it in collaboration with school districts and calling on the NJDOE to talk to the federal Department of Education who insists all these special populations take these tests, to see if more flexibility can be provided in testing these special populations.

These recommendations lack specificity, but the acknowledgment of the issue in the face of stupid federal rules is welcome. It is truly cruel and unusual punishment to insist that all students, no matter what their learning challenges take these tests.

2. Varied Assessments

The report acknowledges that a broad array of assessments is necessary for every school district including formative assessments, end of unit assessments, interim assessments and state assessments. The effective use of the first three is a best practice. Teachers use formative assessment daily through their observation of student learning to adjust teaching and provide small group and individual support when needed. End of unit assessments are designed by teachers and actually test what has been taught to see if it has been learned. Interim assessments may be used to provide a sense of student reading level (benchmark) or student performance against a level of expected performance as measured on a rubric.

All three of these types of assessments provide teachers and administrators with real actionable feedback to help them work with real actual kids. The results of these assessments are timely and they inform instruction. The fourth type of assessment, standardized tests, simply does not provide this feedback and should be used sparingly.

So, there is my quick take on the Study Commission report. Blogger, parent and teacher Marie Corfield has launched a more detailed report here. I suggest you check out what Marie has to say on each of the recommendations. Meanwhile, I know we all can't wait for the glossy, "YEAH, PARCC". public relations campaign that will be coming from the NJDOE. The best way for parents to respond to the publicity barrage will be to simply have their children opt out of the tests. Opting out brings the whole scheme crashing down on itself and impacts the quality of a child's education not at all.

Wednesday, January 6, 2016

The Homework Problem

                                Homework! Oh homework!
                                I hate you! You stink!
                                I wish I could wash you away in the sink.
from Homework, Oh Homework, by Jack Prelutsky)

Sunday’s New York Times carried an article by Vicki Abeles, producer director of the documentary Race to Nowhere, entitled Is the Drive for Success Making Our Children Sick? The article discusses the negative impact on students, both privileged and poor, of the drive for academic success. This opinion piece followed on the heels of an article earlier in the week that reported on the efforts of the West Windsor-Plainsboro School District in New Jersey to deal with the issue of student stress. The Superintendent of the West Windsor Schools, David Aderhold, reported to parents that the schools were in crisis due to the overwhelming feelings of stress students were experiencing. Among many other causes, both articles cited homework as one of the major stressors and both articles suggested that controlling the amount of homework could help in reducing that stress.

Homework is as much a part of the American school culture as the three-ring binder, the text book and the football team. Since seemingly the beginning of time, teachers have assigned homework and students have complained about it. Watch any Our Gang comedy from the 1930’s and you will see that children have always gone to extraordinary lengths to avoid homework. But whether the concern is over-stressed kids or homework resistant kids, homework keeps being assigned, keeps being completed or avoided or copied by kids and keeps being expected by parents. Should it be?

Teachers I have spoken to have argued that homework is necessary for reinforcement of information taught in school, or to prepare for a lesson that is to be given the next day, or to extend learning that has taken place in class or because there is not enough time to cover all the material during class. Some teachers have told me that it is important to assign homework to elementary students so they are prepared for the homework they will get in middle and high school. Other teachers have told me they assign homework because parents expect it.

I must say that I don’t find any of those reasons to be compelling ones for assigning homework, but what does the research say? Do students really benefit from homework? And if they do, what kind do they need?

Harris Cooper, professor of neuroscience and psychology at Duke University and author of The Battle Over Homework, has done the most extensive and compelling research in homework. His findings can be summarized as follows:

·         When homework and in-class study were compared, in-class study proved superior.
·         Homework had no academic impact on the achievement of children in the elementary grades.
·         Homework had some positive impact on academic achievement (as measured by tests) of children in middle school as long as the homework was no more than 90 minutes a night.
·         Homework had the most impact on academic achievement (as measured by tests) in the high school.
·         Homework probably works best when the material is not too complex or unfamiliar.
·         It is better to distribute homework material across several assignments rather than have homework concentrate only on the material covered on that day.

Author and lecturer Alfie Kohn is perhaps America’s chief critic of homework. He finds Harris’ findings on the academic benefits of homework less than compelling. In his 2006 book, The Homework Myth, Kohn looks at Cooper’s research and concludes, 

Taken as a whole, current research might be characterized as inconclusive… a careful examination of the data raises serious doubts about whether meaningful learning is enhanced by homework for most students.

Kohn spends an entire chapter on the idea of homework as reinforcement, the number one reason teachers say they assign homework. Reinforcement assignments are not effective for students who do not already understand the concept, so they may end up reinforcing incorrect understandings. Reinforcement assignments are also ineffective for students who have already mastered the material. In other words, for a large number of students in any class, a reinforcement assignment is either wasting time or reinforcing bad habits. Kohn says it would be better to have the reinforcement work take place in the classroom where the teacher can clear up misconceptions and provide new challenges for students who have mastered the material.

With the work of Cooper and Kohn in mind, I think most of what is assigned in today’s schools as homework should happen in the classroom. I think of the art class as an appropriate model. When students attend an art class, they receive some instruction in technique and then they try those techniques out, right there in class in front of the teacher. In that art class the teacher sees the evidence of the student’s understanding and application of the skill and provides immediate formative feedback. I believe this should be the model for all classrooms. The classroom should be a workshop where students learn, apply and receive timely feedback on their work. This model works for all disciplines and virtually eliminates the need for reinforcement homework.

And eliminating the need for homework is a valuable thing because homework has so many negative impacts on children. As the articles cited at the beginning of this post highlight, one negative aspect is stress, stress on children and stress on families. Another negative aspect of homework is the impact on student perceptions of school and learning. Kids see homework as an intrusion on their “free-time” and I do not think that judgment is unreasonable. Homework may also lead to cheating. When I did a study of cheating at the high school where I worked 10 years ago, I found that one major cause of cheating was students feeling over-burdened by homework that they did not find particularly useful or engaging.

Should we do away with homework entirely? Probably not. But as teachers we need to think closely about our reasons for assigning homework and what the likely benefits will be of any given assignment. We need to decide if the time spent on assigning, completing, grading and reviewing homework, is worth the educational gains made. We need to decide if those gains might better be achieved by having the kids do the work under our watchful eye. If we assign homework because there is too much curriculum to cover, we need to reflect on whether we are trying to cover too much. If we are assigning homework so that students are prepared for the next class, we need to examine if that is an effective practice or if it is better to provide the background information in class. 

If we do assign homework, we should be sure assignments are focused on integrating concepts across more than one class, that the homework is readily doable for the students to whom it is assigned and that students get helpful feedback on their work. It should also be brief and assigned perhaps once or twice a week and never on weekends and holidays. For me, the best homework assignments would be those where students get the opportunity to explore a topic of their own interest or read a book or magazine of their own choice.

Maybe the homework that would be best for children is not homework assigned from school, but real “home” work. Time at home to spend interacting with family and friends and time to explore personal passions.