Sunday, November 29, 2015

$2.00 a Day - How Does Poverty Enter Into the Equation?

I just finished reading an excellent book -- $2.00 a Day: Living on Almost Nothing in America by Kathryn J. Edin & H. Luke Shaefer.  This eye-opening book provides a picture of the lives of those Americans living on "almost nothing" -- and records the tremendous increase in their numbers since welfare "reform" in 1996.  The descriptions of the lives of the children impacted by poverty are particularly gripping.  It's impossible to read this book and think that all the problems these children have will be solved if we simply "raise the bar" with standardized tests and increase the stakes associated with test results.  How can "raising the bar" help children living in shelters, or who go hungry most nights, or who are abused, or who live in continual chaos? There is a wonderful example in the book of a teacher (a dedicated TFA intern!) who makes a difference in the lives of one student and her family -- not by demanding more, "raising the bar," imposing more "rigorous" requirements on homework, but by helping the family with medical and dental care, clothing, and the like.

Welfare "reform" in 1996 appears to have succeeded in providing more assistance to low-income working people through the means of the "earned income tax credit" but ended up missing the other half of the plan -- providing access to jobs for people who need them.  (It could also be that the EITC actually functions as a subsidy for employers, allowing the continuation of very low wages rather than requiring a livable minimum wage -- the authors point out that there is no state in the country in which a person working full-time at the minimum wage can afford a market rate two-bedroom apartment.)

If we truly want to make a difference for this generation of children, and as a corollary improve the United States' performance on international standardized tests (whether or not that's an important goal, it's clear that many think it is), we need to change their life circumstances and make it possible for them to focus on learning. In comparing the U.S. with other developed countries, the worst statistic is that according to the OECD the U.S. poverty rate is the "highest in the developed world." We should be determined to change that, and invest in doing so. Rather than arguing about such matters as whether 2% of instructional time is the "right" amount of testing, or how to measure that, our representatives could be spending their time working on improving the life situations of children living in poverty.

Sunday, November 22, 2015

How Much Testing Is Too Much?? (Hint: It's More Than the Actual Test-Taking Time)

Arrrgggh -- it's been a long time since I've posted -- too many other projects going on, I guess.  Meanwhile, limiting the time allotted to standardized testing has become a popular topic. Even the politicians are supporting test-taking limits, at least if those limits don't impact whatever "reforms" they support. (See the federal government's "Testing Action Plan," which proposes a 2% limit on classroom time spent on testing, and agrees to provide flexibility -- but in areas other than those that are federally mandated.)

I agree that there is too much testing, and way too much emphasis on standardized tests as a measure of how students are doing, how teachers are doing, and how schools are doing.  But -- will simply enacting a law, as New York has done, limiting testing to 2% of instructional time, solve the problem if everything else stays intact, or will it simply add yet another piece of paperwork as school administrators have to calculate testing time and undoubtedly submit a report to prove compliance with the new law?

First of all, how is this going to be calculated? And which tests are going to be included? (And what research supports the 2% figure as the correct one, anyway?  But that's another topic.) Reading the federal "Testing Action Plan," it appears that the limit applies only to the actual time students spend taking the tests and only to state and local "standardized" tests.  The plan includes an offer to schools of flexibility in areas other than federally mandated testing, and nothing that in any way indicates that less weight is going to be placed on the results of testing in evaluating teachers and schools.

Let's take the time issue first.  How will the actual classroom test-taking time be measured?  In Massachusetts, there are at least three different ways of measuring actual test-taking time -- (1) the Department of Elementary and Secondary Education's estimate of how long each test session takes, (2) the DESE's recommendation for principals in scheduling the test-taking time, and (3) the time it actually takes students to take the tests.  Not surprisingly, these three measurements are quite different.  For example, the MCAS 4th grade "long composition" test consists of two test sessions.  The DESE characterizes these as 45-minute sessions, but in its 150-page instruction manual for principals recommends scheduling 2 hours for each session.  Furthermore, in practice, these two sessions generally take 4th graders most or all of the school day to complete.  In our school, starting at 8:15, most of the 4th graders were finished by the time a late lunch was served at 1:15 or 1:30, but a significant percentage worked through their lunch and finished only at the end of the school day at 2:25.  Thus, the two sessions actually took close to 5 hours for most students, and 6 for many.  (In fact, the DESE seems to anticipate this length -- the instruction manual includes directions for principals about how to maintain security for students who need to work through lunch, as well as a recently-added permission for students to continue working for a short period of time after the end of the school day as long as arrangements are made for them to be picked up and taken home!)

My guess is that in calculating the 2%, the number that will be included is the stated 45 minutes per test session. 

The second issue, of course, is that the impact on instruction is far more than the testing time for those students taking the tests.  In addition to those students actually testing, the other students in the school are also affected by the testing.  First, students who have finished testing cannot be immediately involved in classroom instruction as long as their teachers are still proctoring testing.  Second, because many students require testing accommodations, most specialist teachers and many teachers in the other grades are involved in proctoring the tests, thus impacting the non-tested students.  (There is no provision for any funding for schools for extra staffing in order to complete the testing and, in fact, there is a prohibition on having anyone other than school department employees do the testing.  It would be possible to hire substitute teachers to work with students who are not testing in order to have their teachers help with proctoring, but aside from the extra expense that's not the same as regular classroom instruction.) Finally, in most cases, the school schedule is significantly affected by scheduling time for testing, in order to schedule the testing at the best time of day and give students the best opportunity to do well on the tests.  That affects the quality of instruction for students in the whole school, both tested and non-tested students. In our school, with only two grades (4th and 5th) testing generally took 10 school days per year, not including time for make-up testing or any test preparation or any district or school level testing.  (Ten school days per year amounts to 5.5% of the school year. )  The problem is worse in schools with more grade levels involved in testing.  Last year, I looked at the spring testing schedule in another local middle school and counted 29 days on which the school schedule and instruction were impacted by testing. (29 days is 16% of the school year.)

As far as I can tell, the current proposal for a 2% limit on testing time will simply create another paperwork headache for principals, who will be required to calculate testing time, undoubtedly either simply using the stated time of 45 minutes/test session or with another 20 pages or so of instructions on how to calculate the actual time when students take anywhere from 1-3 hours per test session, and then will be required to notify parents and file an "action plan" if testing exceeds the limit.

What a great idea (NOT)!!!  Has anybody considered the alternative of allowing schools and districts to determine the amount and type of testing needed to provide sufficient information on student progress and achievement and report out that information to their communities???? Or, rather than enacting yet another supposed "quick fix" that will cause more problems that it solves, taking the time to review the impact of the legislation and regulations that already exist and developing a more appropriate long-term solution?