skip navigational links
NW Laboratory Home

Northwest Education: The Science of Quality: Education Research in School Reform

Evaluation in the Classroom

In Oregon, teachers are learning to use the tools of evaluation to find out if their teaching strategies are really working.

By Jennifer Stepanek

Summer 2004

SALEM, Oregon—In the last five minutes of just about any workshop, teachers usually scramble to fill out the brightly colored evaluation form: "How effective was the training?" "How useful was the information presented?" "Rate your overall satisfaction with the materials."

While these questions are not necessarily trivial, these familiar surveys do not go very deep into assessing the impact of professional development. And in this era of the No Child Left Behind legislation, they certainly do not provide the type of evidence that meets the law's provisions. The act requires that professional development paid for with federal funds be based on scientific research that provides evidence of a positive impact on student achievement.

One way that the U.S. Department of Education is trying to establish scientifically proven professional development models is through the Mathematics and Science Partnerships (MSP) program, which provides funding for state-level projects that focus on improving teacher quality. Each state has designed its MSP program differently, but all the programs support partnerships between school districts and institutions of higher education. In the partnerships, mathematicians and scientists are responsible for designing and providing learning experiences for teachers.

The MSP projects also invest a lot of dollars and effort in evaluating models of professional development. In Oregon, Ginger Redlinger is helping schools to take advantage of this investment. As an education program specialist with the Oregon Department of Education, she is responsible for administering the state MSP program. The program provides an important opportunity for teachers and administrators to be more involved in evaluation. This makes the program challenging, but Redlinger points out that it is also full of opportunities for professional growth—"with every challenge, there's a hidden gem."

She knows these challenges firsthand: Before joining the department three years ago, Redlinger taught mathematics and science for 10 years. "I felt I knew a lot more when I came to the department than I feel like I know now," she says with a laugh. "You step into a new role with new challenges, and it causes you to grow in ways that you've never imagined."

Administering the MSP program has meant that Redlinger is continually learning about evaluation. "It's an evolving process. I find that as I engage in the dialogue around this program, my own understanding is furthered."

Because program evaluation is an unfamiliar area for most teachers and administrators, Redlinger made sure that Oregon's request for proposals provided them with guidelines for developing good evaluation plans. "I wanted people to develop their proposals knowing exactly what the components of a good evaluation are, as well as how they should be using evaluation and where to get information about it," she says. "And it's interesting, because I can see that people are getting it—it's like light bulbs are going on all over the place."

The guidelines include developing clear objectives, explaining how teacher and student outcomes will be measured, and identifying strategies for measuring progress on an ongoing basis. In addition, extra points are awarded for bringing in an external evaluator.

To evaluate the impact of the MSP professional development, teachers will be required to demonstrate changes in their knowledge. That's a tall order. "Asking teachers to use pre- and posttests to reflect on their own learning gains can be a little threatening," Redlinger says. "It puts teachers back in the role of being a student, which I think of as positive, but it's a challenge if you haven't done that in a long time. And then relating their own personal learning gains to the learning gains of their students takes it to another level."

Redlinger is excited to see how this will play out in the classroom. "Teachers generally teach in isolation, and if they do get together, they usually talk about everything up front: What am I going to teach? How am I going to plan the lesson? What are my activities? What are my assessments? Then they go back to their rooms and they teach, and they never talk about what happened."

Redlinger is hoping that, through the MSP projects, teachers will look more closely at evidence of student learning. "Teachers often do some ongoing monitoring and adjustment, but they usually don't stop and say, 'How effective was I for all students? Did that activity really give me the results that I intended?'"

Taking a broader perspective, teachers may begin to look beyond their own classrooms for evidence. Redlinger says that when teachers collaborate across grade levels, they can get a better understanding of long-term outcomes. "For example, if I'm a fourth-grade teacher, I will talk to the seventh- and eighth-grade teachers to find out if the instructional strategies and materials I use to teach patterns are effective in preparing students for algebra," she explains. "That's a whole new way of thinking. Usually as a teacher, if I'm looking at data, I only look at how well my students do while they are in my classroom."

Redlinger believes that the MSP program will help teachers and district leaders to use data and evidence more precisely. "This program requires teachers to use data and to think about things in the larger view," she says. "They'll have to do so in ways that they haven't done before."

For example, Redlinger points out that while more schools are taking time to look at their state assessment data, these data are only one aspect of a much bigger picture of student learning. "The MSP program is an opportunity for people from different parts of the educational system to think and talk about how we evaluate student learning—which scores or which assignments are better indicators of student growth than others? I hope that everyone will come away with a better understanding of the process we can use to determine the effectiveness of practices and programs relative to student performance."

CONSIDER THIS

Base conclusions on evaluative evidence. One way of expressing the difference between research and evaluation is that research is a process for testing a hypothesis, with conclusions based on the results. Evaluation uses research methods, but includes processes for identifying standards and analyzing results against those standards to make evaluative conclusions. Evaluations should include strategies for identifying side effects or unintended outcomes, which are rarely a concern in research studies.

Start by identifying outcomes. The first step in planning any evaluation is identifying the intended outcomes of the program. Along with the outcomes, the evaluation will also be guided by indicators (the means of measuring the outcomes) and criteria (the standards for success).

Use a logic model. A logic model is a graphic representation that illustrates the theory and assumptions behind a program. It is very useful for making sure that the outcomes and criteria make sense and for building a common understanding of a program. Logic models usually include program inputs (the necessary resources), activities (processes, strategies, products, and tools), the direct results of the program (for example, the number of people served), and the outcomes (changes that result from the program).

Combine aspects of formative and summative evaluation. Formative evaluation is used to assess ongoing activities to provide information to improve the program. Summative evaluation is used to assess the program's success in fulfilling its purpose and goals and producing positive outcomes. Each type of evaluation contributes to the overall understanding of a program's success.

Form partnerships. It is becoming more common for evaluation guidelines to specify the use of experimental designs, in which participants are randomly assigned to treatment and control groups. Such designs require a great deal of technical expertise. Bringing in an external evaluator may be necessary to ensure an effective and sound evaluation.

Use mixed methods when appropriate. A combination of qualitative and quantitative methods is often the most effective way to evaluate a program. Mixed-method evaluation designs build on the strengths of each method and minimize the weaknesses. Using more than one type of data also provides more information and builds a better case for the findings.

Keep evaluation standards in mind. Evaluations should be useful (timely, informative, and influential); feasible (as easy as possible to implement, efficient in using time and resources); appropriate (conducted legally and ethically); and accurate (comprehensive, technically sound, based on logical judgments).

Mathematics and Science Partnership Programs:

Alaska
Beverly Smith
(907) 465-2826
beverly_smith@eed.state.ak.us

Idaho
Pat White
(208) 332-6998
pwhite@sde.state.id.us

Montana
Linda Vrooman Peterson
(406) 444-5726
lvpeterson@state.mt.us

Dennie Munro
(406) 444-3114
dmunro@state.mt.us

Oregon
Ginger Redlinger
(503) 378-3600 ext. 4419
ginger.redlinger@state.or.us

Washington
Eric Wuersten
(360) 725-6311
ewuersten@ospi.wednet.edu

Resources

Using Data/Getting Results: A Practical Guide for School Improvement in Mathematics and Science by Nancy Love (Christopher-Gordon Publishers, 2001)
http://ra.terc.edu/publications/using_data/using_data.html

The 2002 User-Friendly Handbook for Project Evaluation (National Science Foundation, 2002)
www.nsf.gov/pubs/2002/nsf02057/start.htm

W.K. Kellogg Foundation Logic Model Development Guide
www.wkkf.org/Programming/Resources.aspx?CID=281

Respond to this article

| The Science of Quality: Education Research in School Reform |