Saturday, February 08, 2014

Alex Russo Concludes Opt Out Is Irrelevant Because There is No Testing Problem in America

The poodle of corporate education reformers, Alex Russo, has a puff piece in Atlantic aimed to disparage the Opt Out movement. Doing his best to present the growing opposition to psychometric insanity as a momentary flare-up in the steady march toward measuring badly what no longer matters while resegregating America, Alex takes a page from the Jay Mathews playbook of lazy reporting and unsupported opinion fronting as journalism.  
Two issues that deserve attention.  The first is Russo's claim that civil rights groups support the Obama-Gates plan for big data, more tests, and the White Race to the Top.  This is simply a lie.  From WaPo in 2010:
The “Framework for Providing All Students an Opportunity to Learn” is a collaboration of these groups: Lawyers’ Committee for Civil Rights Under Law, Rainbow PUSH Coalition, Schott Foundation for Public Education, National Association for the Advancement of Colored People, National Coalition for Educating Black Children, National Urban League, and the NAACP Legal Defense and Educational Fund, Inc.
 . . . .
“For far too long, communities of color have been testing grounds for unproven methods of educational change while all levels of government have resisted the tough decisions required to expand access to effective educational methods. The federal government currently requires school districts to use evidence-based approaches to receive federal funds in DOE’s Investing in Innovation grant process. So, too, in all reforms impacting low-income and high-minority communities, federal and state governments should meet the same evidence-based requirement as they prescribe specific approaches to school reform and distribute billions of dollars to implement them.
“Rather than addressing inequitable access to research-proven methodologies like high-quality early childhood education and a stable supply of experienced, highly effective teachers, recent education reform proposals have favored “stop gap” quick fixes that may look new on the surface but offer no real long-term strategy for effective systemic change. The absence of these “stop gap” programs in affluent communities speaks to the marginal nature of this approach. We therefore urge an end to the federal push to encourage states to adopt federally prescribed methodologies that have little or no evidentiary support – for primary implementation only in low-income and high-minority communities.
My second objection comes from Russo's attempt to minimize the extent of the testing problem, going so far as to claim that confirmation is lacking on the testing tsunami since NCLB: 
Anecdotal reports suggest that testing has increased in recent years, though it is not known how much—or how widespread—the actual increase may be.
Really, Alex??  You didn't notice??  Or is it you just don't believe what teachers and researchers have been telling you since before NCLB became law?  

Maybe a number would help Alex understand the extent of the problem, but, then, it is hard to quantify the wholesale shift in American education from teaching and learning to test prep and testing.  

And how about the standardized conflagration about to be unleashed with Common Core's burgeoning batteries of exams??  

This is from the Brown Center in 2012, a while before Russo started looking for evidence of increasing tests and the increasing costs emanating from such increases:

. . . .It is unsurprising that testing costs increased in the decade following the passage of NCLB, as states expanded the use of their existing assessments to cover additional grades and developed new tests. The most current and comprehensive publicly available data on state spending on assessments were collected by the Common Core assessment consortia, SBAC and PARCC, through surveys of their member states in 2010.8 The SBAC states reported spending between $7 and $123 per student on their math and ELA assessments.9 For the vast majority of states, these numbers are substantially more than spending almost ten years earlier but still represent a small share of overall expenditures. The state that spent by far the most on testing on a per-student basis (Hawaii) still devoted less than one percent of total spending to this purpose.10 PARCC reported spending per test, as compared to per student, but assuming that the average student takes two to four tests, the overall pattern of assessment costs is similar.
Costs that amount to less than one percent of per-pupil spending may seem trivial, but warrant careful examination for several reasons. First, given rising pressures on schools to “do more with less,” all expenditures should be examined for their cost-effectiveness, and even small amounts of spending or savings add up across our nation’s K-12 education system. Spending in U.S. public schools totaled $658 billion in 2008-09 (the most recent year for which data are available), so even one-half of one percent would add up to more than $3 billion each year.11 And states can make changes to their assessment budgets with relative ease compared to some larger categories of expenditures, such as employee salaries, which are often constrained by collective bargaining agreements. For example, Georgia cancelled the upcoming spring 2013 administration of its state test to first- and second-grade students due to budget constraints.12
Second, there is the risk of multi-million-dollar assessment contracts contributing to a political backlash against testing among parents and taxpayers who oppose the use of standardized testing for accountability purposes or object to public dollars flowing to for-profit companies (as most of the testing contractors are). For example, the Wall Street Journal recently reported that “[a] national coalition of parents and civil-rights groups, including the NAACP Legal Defense and Educational Fund, signed a petition in April [2012] asking Congress to reduce federal testing mandates.”13 These anecdotes aside, Americans remain broadly supportive of testing. In a 2011 survey, only nine percent of all respondents (and 11 percent of parents) said they were opposed to the federal government requiring states to test students in math and reading in grades 3-8 and once in high school (as mandated under NCLB).14 But as education budgets continue to tighten, expenditures on testing may draw increased scrutiny.
Third, change is afoot in testing systems across the country. Cheating scandals have prompted concerns about test security, especially as more districts tie test scores to the evaluations of teachers—who often proctor their own students’ exams. Criticism of multiple-choice “bubble tests” has increased interest in moving toward exams with other types of items, such as performance tasks that are designed to assess students’ analytical reasoning skills more effectively than a question with a single correct answer. Some states have started to move towards computerized assessments, and both of the tests being developed by the Common Core consortia are computer-based. All of these proposed upgrades of assessments systems have implications for costs, and will be scrutinized in light of competing demands on state budgets.
And finally, though the federal government financed the initial development costs of the new assessments being created by the Common Core consortia, states will likely have to fund the maintenance and enhancement of these assessments after the federal grant ends in September of 2014, a full six or more months before the first operational tests are administered. It is not yet clear how participating states will share the cost of sustaining the consortia assessments, or whether the federal government will provide additional support for this effort. What is clear is that states that currently have inexpensive assessments will be under pressure to spend more to pay for the ongoing costs of the consortia assessments.
Still unconvinced, Alex?  Here are some other "anecdotal" reminders of what is happening in the school world that you know nothing about.  From FairTest:
Myth: Common Core tests will be much better than current exams, with many items measuring higher-order skills.Reality: New tests will largely consist of the same old, multiple-choice questions.Proponents initially hyped new assessments that they said would measure – and help teachers promote – critical thinking. In fact, the exams will remain predominantly multiple choice. Heavy reliance on such items continues to promote rote teaching and learning. Assessments will generally include just one session of short performance tasks per subject. Some short-answer and “essay” questions will appear, just as on many current state tests. Common Core math items are often simple computation tasks buried in complex and sometimes confusing “word problems” (PARCC, 2012; SBAC, 2012). The prominent Gordon Commission of measurement and education experts concluded Common Core tests are currently “far from what is ultimately needed for either accountability or classroom instructional improvement purposes” (Gordon Commission, 2013).
Myth: Adoption of Common Core exams will end NCLB testing overkill.Reality: Under Common Core, there will be many more tests and the same misuses.NCLB triggered a testing tsunami (Guisbond, et al., 2012); the Common Core will flood classrooms with even more tests. Both consortia keep mandatory annual English/language arts (ELA) and math testing in grades 3-8 and once in high school, as with NCLB. However, the tests will be longer than current state exams. PARCC will test reading and math in three high school grades instead of one; SBAC moves reading and math tests from 10th grade to 11th. In PARCC states, high schoolers will also take a speaking and listening test. PARCC also offers “formative” tests for kindergarten through second grade. Both consortia produce and encourage additional interim testing two to three times a year (PARCC, 2012; SBAC, 2012). As with NCLB, Common Core tests will be used improperly to make high-stakes decisions, including high school graduation (Gewertz, 2012), teacher evaluation, and school accountability.  
Myth: New multi-state assessments will save taxpayers money.Reality: Test costs will increase for most states. Schools will spend even more for computer infrastructure upgrades.Costs have been a big concern, especially for the five states that dropped out of a testing consortium as of August 2013. PARCC acknowledges that half its member states will spend more than they do for current tests. Georgia pulled out when PARCC announced costs of new, computer-delivered summative math and ELA tests alone totaled $2.5 million more than its existing state assessment budget. States lack resources to upgrade equipment, bandwidth and provide technical support, a cost likely to exceed that of the tests themselves (Herbert, 2012). One analysis indicates that Race to the Top would provide districts with less than ten cents on the dollar to defray these expenses plusmandated teacher evaluations (Mitchell, 2012).
Myth: New assessment consortia will replace error-prone test manufacturers.Reality: The same, incompetent, profit-driven companies will make new exams and prep materials.The same old firms, including Pearson, Educational Testing Service and CTB/McGraw-Hill, are producing the tests. These firms have long histories of mistakes and incompetence. The multi-national Pearson, for example, has been responsible for poor-quality items, scoring errors, computer system crashes and missed deadlines (Strauss, 2013). Despite these failures, Pearson shared $23 million in contracts to design the first 18,000 PARCC test items (Gewertz, 2012).
Myth: More rigor means more, or better, learning.Reality: Harder tests do not make kids smarter.In New York, teachers witnessed students brought to tears (Hernandez & Baker, 2013), faced with confusing instructions and unfamiliar material on Common Core tests. New York tests gave fifth graders questions written at an 8th grade level (Ravitch, 2013). New York and Kentucky showed dramatic drops in proficiency and wider achievement gaps. Poor results hammer students’ self-confidence and disengage them from learning. They also bolster misperceptions about public school failure, place urban schools in the cross hairs and lend ammunition to privatization schemes. If a child struggles to clear the high bar at five feet, she will not become a "world class" jumper because someone raised the bar to six feet and yelled "jump higher," or if her “poor” performance is used to punish her coach.
Myth: Common Core assessments are designed to meet the needs of all students.Reality: The new tests put students with disabilities and English language learners at risk.Advocates for English language learners (Maxwell, 2013) have raised concerns about a lack of appropriate accommodations. A U.S. Education Department’s technical review assessed the consortia’s efforts in July 2013 and issued a stern warning, saying that attempts to accommodate students with disabilities and ELLs need more attention(Gewertz, 2013).
Myth: Common Core "proficiency" is an objective measure of college- and career-readiness.Reality: Proficiency levels on Common Core tests are subjective, like all performance levels.Recent disclosures demonstrate that New York State set passing scores arbitrarily (Burris, 2013). There is no evidence that these standards or tests are linked to the skills and knowledge students need for their wide range of college and career choices (Ravitch, 2013). In addition, school officials have often yielded to the temptation to cheat and manipulate test results to bolster the credibility of their favored reforms. Examples include Atlanta, New York, Washington, DC, Indiana, Florida, and more (FairTest, 2012).
Myth: States have to implement the Common Core assessments; they have no other choice.Reality: Yes they do. Activists should call for an indefinite moratorium on Common Core tests to allow time for implementation of truly better assessments.High-quality assessment improves teaching and learning and provides useful information about schools. Examples of better assessments include well-designed formative assessments (FairTest, 2006), performance assessments that are part of the curriculum (New York Performance Standards Consortium), and portfolios or Learning Records(FairTest, 2007) of actual student work. Schools can be evaluated using multiple sources of evidence that includes limited, low-stakes testing, school quality reviews, and samples of ongoing student work (Neill, 2010). It’s time to step back and reconsider what kinds of assessments will help our students and teachers succeed in school and life.

References
  • FairTest. 2007. “The Learning Record.” http://www.fairtest.org/learning-record\

No comments:

Post a Comment