Feature

Why Add Value in Assessment?

A pilot project in Ohio gains interest in value-added measurements of individual student by James W. Mahoney
An acquaintance recently shared a story with me about a couple that used a friend's cabin in northern Minnesota. While enjoying the serenity and remoteness, the couple kept warm using logs from the woodpile in the fireplace. Upon their return, the cabin's owner inquired as to whether they had replaced the used logs by adding wood to the pile for the next visitor. The owner explained it was an old Scandinavian tradition to leave the woodpile higher than you found it. The man who had used the cabin replied that he had, then quickly drove back to add to the woodpile that night.

It's a simple premise really. When working with something, leave it better than you found it by adding value. Make the investment grow. Now suppose we apply that same concept to the education of children. Reasonable people might disagree with the amount of progress or gain a child should make, but we can all agree on this—the role of educators is to take children from where they find them and add value. We add value by teaching subject areas. We add value by modeling appropriate behavior. We add value by who we are.

The truth is that good teachers add value in a number of measurable and immeasurable ways. While we can measure the gain of a student in mathematics, it is impossible to measure the impact of a teacher who motivated a child to want to do math.

Overall Snapshot

For a moment, let's focus on measuring the value-added component. We know that at any given grade level there are students who already exceed the state's standards for that grade level, just as there are those who fall far below the standards. The reasons for this discrepancy are endless.

Elementary school teachers have children who cannot name all the primary colors and at the same time have students who can identify 5,000 words. Teachers cannot control how kids enter the classroom and should not be held accountable for factors they cannot control. Teachers do have control over the progress children make and we should all expect children to make gains regardless of each child's starting point.

Now take value-added one step further. How much gain should a child make in reading or math in one year? Again, people might disagree, but let's assume our expectation is that a child should make a minimum of one year's worth of gain for one year's worth of schooling. Students who come in at a low level of achievement will need to make more than one year's worth of gain.

Suppose a 5th grader is reading at a 2nd-grade level. If we want that child to be reading on grade level by the end of 8th grade, then the child will need to make slightly more than one year's growth each year. Now suppose another 5th grader is reading at the 8th-grade level. No one, particularly the child’s parents, would be satisfied if the child was reading on an 8th-grade level in three years.

In order to get an overall snapshot of each child’s academic growth, we need to measure the individual progress of children from year to year. The truth is, what gets measured gets improved. While teachers always have tried to take a child from point A to point B during the school year, the inherent challenge has been to measure progress when every child's point A and B are different. Part of the evidence lies in the annual tests to be given in grades 3-8 in math and reading. These tests will help measure points A and B for every child, providing a more accurate view of each child’s performance and progress.

Increasing Interest

Accountability policies that affect large groups of children in a pro-active manner are essential, but most parents want to know, "How is my child doing?” rather than “How is my child's school or district doing?"

In Ohio, Battelle for Kids is leading a pilot project involving 78 school districts that follows the growth of the same individual students over time. This includes the data from approximately 718 buildings, 200,000 students or 20 percent of students in grades 3-8 in Ohio.

Long-time school leaders are enthusiastic about value-added’s use as a management tool. “Value-added research has become an integral component in our efforts to improve individual student achievement,” says Dick Ross, superintendent of the Reynoldsburg, Ohio, City School District. “The utilization of the data provided through value-added analysis has had an impact on our educational practices throughout all grade levels.”

A relative newcomer to the superintendency agrees. Wade Lucas of the Coshocton, Ohio, City Schools says value-added assessment provides his district’s staff “with true measures of the job we are doing with our students. We encourage accountability; in fact we demand it. The approach taken through value-added allows an opportunity to analyze each child in our school system and provide the type of intervention to assure that we leave no child behind.”

Battelle for Kids, a nonprofit affiliate of the Ohio Business Roundtable, was created three years ago with an initial $10 million investment from Battelle Memorial Institute in Columbus, Ohio. The pilot includes districts with urban (Columbus, Cleveland), suburban and rural schools. The growth model being used is the one developed and pioneered by William Sanders, a former University of Tennessee statistician. The success of the pilot project has resulted in state legislation requiring this progress measure be included as a part of Ohio's accountability system in 2007-08.

Sensible Analysis

Creating an accountability system that measures the progress of children is important for several reasons.

  • Value-added analysis is fair, sensible and reasonable.
    People understand that we must improve students' knowledge and skills every year. We need to measure that journey and see whether children are hitting the targets set for them. We need to review the individual progress of each child and the collective progress of all children.

    For example, there may be a unique reason why one student made very little progress in the 4th grade. However, if this one student’s experience is mirrored by nearly every child in the 4th grade, then that would be compelling evidence that something is amiss. Perhaps it's the curriculum. Maybe it's the test given that year. It also could be the primary text isn't aligned to the curriculum or test.

    This is where educators need to use data to ask and answer important questions about learning. Expecting a minimum of one year's growth for every year of schooling is clear. It's sensible, reasonable and understandable. It's the reason why our pilot group has grown from 42 to 78 school districts over two years without any active recruitment on our part.

  • To improve student achievement you must increase progress.
    Think of something you do well. Did you learn to do it overnight? Of course not. The way to ratchet up achievement is to ratchet up progress. Start with attainable goals that will take you where you want to go. Do not overwhelm someone with what they don't know and remove all hope of ever getting there. It is much easier for a child to say "I won't" than to say "I can't" or "I don't know how." If a child is two years behind in any grade-level subject, it is not reasonable to expect that deficit to be made up in one year.

    What's reasonable is to expect the deficit to be made up through more than one year's growth during each of the succeeding years to 8th grade. We have buildings in our pilot that do not presently meet the state standard (75 percent passing rate) but are making huge progress with every group of kids. This program fuels hope to staff and students alike. It causes them to talk about their practices and what is working.

  • Value-added analysis levels the playing field.
    We need to expect growth for all children, not just those who are behind. Again, we might argue about the amount of growth, but we need all children to grow each year and we need to add value to their education. Without measuring each child's progress, it is possible for a child to score well beyond the standard and actually lose ground over a few years, yet appear successful because he scores above the standard. Several affluent buildings with high-performing students clearly make the case that these students also can make progress. Each child's target may be different, but what's important is that each child has an appropriate target.

Parental Needs

  • Parents want to know.
    If the dream of No Child Left Behind is to become true, then we need to know the starting point for each child. Value-added analysis answers for parents the central question “Is my child making progress?” Simply comparing one grade level to another, one school to another or one district to another fails to answer the need for information on the individual child.

    Parents often track the height measurement of their children by marking lines in the closet or on the wall. Both the child and parent can compare the growth over the years. The same can be done for learning and education by annually measuring growth and improvement. As a parent, this measuring stick can be the stick or carrot to improve performance.

  • Value-added analysis can measure program effects.
    Measuring the value added to groups of children over time offers data with which to evaluate programs. What is the impact of early literature programs? Cooperative learning? Phonics? Math using manipulatives? The list of programs that can be measured is endless, yet we have programs being used everywhere that haven't been tested for their efficacy in the learning process.

    School districts now are designing their own internal studies around data. For example, one of our pilot districts had two comparable buildings making entirely different rates of progress with similar students at the same grade level in science. Their analysis suggested the difference was attributable to a departmental strategy in one building that enabled teachers to specialize in certain areas of science standards instruction.

    Value-added analysis offers a way to gauge a program’s effectiveness and provides critical data to practitioners and policymakers to guide instruction, target resources and help children.

  • Value-added analysis offers hope and encouragement.
    Confidence comes from competence. High self-esteem is the result of achievement, not the cause of it. Some schools and students start so far behind they have little hope of meeting, in any fashion, the state’s standard of performance. So why try, especially if there are not any intermediate goals that offer hope and encouragement? When a goal is so far out of reach it could never reasonably be expected to be met, what happens? People quit, give up or blame the system.

    As human beings, we need opportunities to grow, succeed and celebrate. I am not talking about minimum goals or low aspirations. I am talking about setting reasonable expectations and acknowledging progress the way we acknowledge achievement. You can't get through the door of achievement without also going through the door of progress.

Forward Motion

Oliver Wendell Holmes, the great chief justice of the U.S. Supreme Court, once noted, “The great thing in this world is not so much where you stand as in what direction you are moving.”

Value-added analysis is the right direction. In the 2003 annual Phi Delta Kappan/Gallup Poll surveying the public's attitudes about schools, 84 percent of the public stated that the best way to gauge the performance of a school was to measure individual student progress from where the child started. In this age of standards-based education it makes sense and will help every child do his or her best. Is that not the aspiration of all parents for their child?

Jim Mahoney, a former superintendent in Ohio, is executive director of Battelle for Kids, 41 S. High St., Suite 2220, Columbus, OH 43215. E-mail: jmahoney@battelleforkids.com