Ever since the No Child Left Behind act was passed in 2001, district administrators, teachers, policymakers and testing companies have been working to implement the law's requirements to improve achievement so that all students are proficient in state standards by 2014. The law requires the use of standardized tests to measure which students have attained profi ciency and which students have been "left behind." This particular requirement has created a lively debate over the role of tests and their proper use and misuse.
What have been the effects of NCLB on the testing industry?
NCLB has certainly increased the amount of testing done at the state level. In many cases, this state testing replaced district testing, so it was not an additive to the industry. One big change was that off-the-shelf tests, such as norm-referenced tests, were replaced by custom-developed tests designed to measure the state standards. This is a very expensive undertaking and stretched the budgets of smaller states.
Some have claimed that the increase in custom-built state assessments has placed a strain on the testing industry and that we are taxing the capacity of the industry to score all of those tests. This is not true. Testing companies have added capacity, and some of the smaller companies that were niche players have now become larger and have vertically integrated to full-service providers. The result is more competition for state contracts, which keeps costs to taxpayers lower. Prior to NCLB, for example, Educational Testing Service did not provide state assessments for elementary and secondary education. Since passage of NCLB, we have provided assessments for ten states to date.
How are testing companies addressing concerns about standardized testing?
The testing industry provides information to increase the public's knowledge and understanding of standardized testing through its industry associations such as testingfacts.org and as individual companies. Through various mediums, we provide answers to people with questions about what test scores mean and what they tell us about students.
For example, many people mistakenly believe that score differences between groups are an indication of an unfair or biased test. If a test has been developed properly by a reputable testing company, techniques are applied to minimize test bias, and score differences between groups are not the result of test bias. Maryland is a good example. The recent Center on Education Policy report found the achievement gap in reading narrowed between black and white students and between low-income and middle-income students, but math scores were mixed. Overall, in many areas black and Hispanic students are still struggling to catch up to white and Asian-American students. Poverty, inadequate language skills, unsatisfactory home conditions, poor attendance-these are all barriers to student success and contributors to score differences. It's easier to blame tests than it is to fix the enduring socioeconomic and educational disparities that produce score differences.
Another common misconception is that teachers today must "teach to the test," by teaching only material that will appear on the state NCLB tests. This practice does not help students learn content standards, and students suffer the consequences later on in their education. On the other hand, if a teacher is teaching a curriculum based upon the state standards, and the test accurately measures the standards, then the students will get higher test scores.
The system may never be perfect, but the consequences of eliminating tests come with substantial risks. These include graduating underprepared and underskilled students, having to increase the number of costly collegiate remedial programs to compensate for academic deficiencies, and having a lack of reliable, unbiased information on student performance.
What does the future hold for the testing industry?
Ultimately, a lot depends on what happens with NCLB. I believe it will be reauthorized in some form after this election cycle and that most states will continue to administer about the same number of assessments as they do today. The methods used might change or be less restricted than NCLB allows in its present form. For example, we might see more constructed-response or performance assessment as states seek to have richer measurement of some of their standards. We are also likely to see more states create balanced systems of assessment for learning given throughout the year in addition to the assessment of learning that takes place in the spring.
Other likely changes are more rigorous assessment of 21st-century skills, such as information and communication technology (ICT) literacy. Our research shows that only half of the students we tested for ICT skills could correctly judge the objectivity of a Web site. Educators nationwide have expressed concern regarding students' inability-regardless of district or demographic-to recognize quality information sources through electronic mediums. As American students continue to do poorly in national and international comparisons, more and more educational policymakers are calling for these kinds of skills to be part of the state standards and included in both state and local assessments. By uniquely defining ICT literacy, we're helping students analyze and skillfully use the vast amount of information on the Internet.
As it has in the past, the industry will respond to meet the changing needs of K12 educators and policymakers. But the overall goal-advancing learning and ensuring that the United States has a competitive, competent workforce-will remain unaltered.
John Oswald is senior vice president and general manager of K12 Assessment Programs for Educational Testing Service (www.ets.org).