William L. Sanders, the architect of Tennessee's widely-known, growth-based "value-added" model of measuring school and teacher effectiveness, responds to this paper. Read his response here.

Tennessee is one of 15 states participating in a pilot program to explore new ways to measure school performance under the federal No Child Left Behind Act. Under NCLB, states are held accountable for ensuring that sufficient numbers of schools' students are meeting state proficiency standards and improving schools that fail to measure up. In 2005, then-U.S. Secretary of Education Margaret Spellings launched a pilot program to study ways of measuring this "adequate yearly progress" that would reward schools for improving student achievement over the course of a school year.

Originally, NCLB judged schools mostly on whether sufficient numbers of their students met state standards each year. The law also included a "safe harbor" provision, which gives schools credit for improving student achievement even if they do not meet the specific performance targets. Some critics of the law charged that these provisions do not sufficiently account for the fact that some schools have students who are more challenging to educate than others. In response, Spellings sought to experiment with ways to measure annual student progress, known as growth models.

Tennessee was one of the first seven states that Spellings approved to implement a new school-rating system under a pilot program, which permits schools to comply with NCLB's adequate yearly progress requirement by having their students make enough progress each year to ensure that they meet a state-defined definition of proficiency within three or four years-a concept known in education circles as "growth-to-proficiency." Since Tennessee was approved, their method, a "projected" or "expected" score approach, has been adopted by two other states—Ohio and, most recently, Pennsylvania. It is also under consideration by several other states and districts. Yet while growth models can help develop better school accountability systems, a close analysis of Tennessee's approach illustrates important issues that state and federal policymakers should be aware of as they consider expanding the use of growth-to-proficiency models.

The Tennessee approach is emerging as a model that other states may follow; therefore, this paper examines the advantages and disadvantages of that scheme. The analysis examines two broad themes. It considers the implications of the Tennessee growth model with regard to the amount and pace of progress students are expected to make on Tennessee's state tests, each year and over multiple years, in order for such progress to be deemed "adequate." It also discusses the broader implications for drawing conclusions about the achievement levels Tennessee students are expected to meet relative to national standards and other states as measured by the National Assessment of Educational Progress (NAEP), a national test regularly given to a sample of students in every state.

The Tennessee growth model has some distinct advantages over the current NCLB "status" model, under which schools are judged based on how cohorts of students (i.e., all students in a single grade or group of grades) compare to a similar cohort of students from the previous year. As such, school ratings under NCLB are currently based on changes in the performance of different students from year to year, rather than growth in the performance of the same students over time. At the broadest level, the Tennessee growth model may more fairly credit those schools and districts that have made significant progress with low-achieving students that would not be reflected in the percentage of students who have met or exceeded the proficiency benchmark. This, in turn, may have instructional benefits for individual students as schools focus additional attention on a broader group of students and track the progress of students over a multi-year period.

The Tennessee growth model system, however, has some significant potential downsides:

• By setting an interim goal short of proficiency, in a state judged by the U.S. Department of Education to have among the lowest standards of any state, it may be setting the bar so low as to evoke fairly small gains in student achievement.
• While the "expected score" system estimates a student's path to proficiency in three years, in fact, many students will not make it to proficient in three years or ever because of a statistical phenomenon known as Zeno's paradox.
• Finally, because this model relies on multiple regression analysis, one must be a statistician to understand it. Although complexity may be a necessary trade-off for more accuracy, there is a loss of simplicity and transparency for parents and the general public.

These downsides could have a significant negative impact on student learning in Tennessee and show some of the risks associated with growth models if they are not implemented with a keen eye toward ensuring results for the lowest-achieving students. Slowing down the pace at which students are expected to learn academic skills in elementary school may create long-term problems for students and create larger and more difficult burdens for public education in junior high, high school, and beyond. A wide body of research suggests, for example, that children who do not acquire language skills in the early grades have an increasingly difficult time catching up to their peers as they progress. This parallels neuropsychological research that shows critical periods for brain development in language and other areas of cognitive functioning.

The Tennessee growth model will also reduce the number of schools identified by NCLB as falling short academically. This could be a positive change if it allows the state to focus more intensely on the lowest-performing schools. However, it will also mean that some schools may not be able to avail themselves of resources that could help address student learning problems early enough to prevent future academic failure. These and many other trade-offs are inherent in accountability systems, and for the foreseeable future each state will have to decide for itself how to manage competing considerations within federal rules.

The growth models currently being used also vary greatly in their methods and take on unique characteristics based on the types of standards and assessment systems that their respective states employ. These specifics matter a great deal in judging what making "adequate yearly progress" or AYP means in any particular state using a growth model.

The purpose of this paper is not to recommend for or against the Tennessee growth model, an approach that reasonable people can disagree over. Rather, because the NCLB accountability system is essentially federalist in that it allows states great leeway in determining their standards, assessments, benchmarks, and definitions of adequate progress, it is important to take a state-specific focus to help consumers of these systems-teachers, parents, administrators, elected officials, and advocacy groups-explain their implications for measuring student achievement and to inform future revisions of both the NCLB law and state plans under the act. The intense and organized resistance to NCLB and the ensuing national debate about the law have to some extent obscured the real trade-offs between different approaches to school accountability. This paper is intended to help those involved in this process make more thoughtful and informed decisions by examining the approach of this important state.

Download "Are We There Yet? What Policymakers Can Learn From Tennessee's Growth Model."

This report was made possible by a grant from the Carnegie Corporation of New York. The statements made and views expressed are solely the responsibility of the author.