"No one test or assessment should be asked to serve all the assessment purposes. We need, at this point, a system made up of articulated components, glued together by their adherence to content standards and serving explicit purposes for assessment."
— National Council for Education Standards and Testing, 1992

As far back as 1992, when the standards movement was launched, the National Council for Education Standards and Testing, a team of experts in education, assessment, and policy, reminded our nation of the different purposes for assessment—from public accountability to creating individualizing instructional plans for children. They also cautioned us to use multiple measures to fit these purposes including standardized tests, running records, informal reading inventories, classroom projects, portfolios, writing samples, debates, literature circle discussions, and more. Now we are hearing the same reminder (Brennan, Kim, Wenz-Gross, Siperstein 2001; Herman 2001; International Reading Association 1999). In fact, in July 2001 members of the National Education Association, the nation's largest teacher's union, endorsed a policy calling for a combination of standardized tests and other assessment tools such as teacher designed assessments when making important educational decisions (Blair and Archer 2001).

Assessment has always been a part of the educational landscape. However, because assessment can serve so many different purposes and can come in so many different forms, it has been confusing and, sometimes, it has been the subject of contentious debate. Unfortunately, as a result, many of us have come to view assessment as a necessary evil, a requirement rather than a helpful part of instruction. But assessment is a critical part of instruction and it can be useful if we understand the pieces of the puzzle.

A Balanced Approach to Assessment
A balanced assessment system consists of three parts: standard assessment, classroom-based assessment, and student self-assessment. Each of these parts serves a different purpose and each is a different and important piece of the puzzle.

Standard Assessment
Standard assessment is the term we use here to refer to assessments that are given to all students in a state, school district, or school. We used to think of these simply as norm-referenced tests, such as the Stanford Achievement Test or the Gates-MacGinitie Reading Test, that compare student performance to the performance of a national sample of students at the same grade. But these days, many states have constructed their own tests, such as the ISAT in Illinois, the FCAT in Florida, or the direct writing assessment in California, that are not norm-referenced. Instead, they have used criteria for student performance set by the state. Regardless of whether a state uses a norm-referenced test or a state-developed test, these standard assessments are designed to evaluate students in a uniform, systematic way against some established standard. All fifty states now administer some type of standard assessment to students at targeted grade levels (Orlofsky and Olson 2001).

There are cautions to keep in mind related to standard assessments. First, although familiarizing students with the test format is important, studies indicate that if students only practice with the format of a test, they are less likely to actually learn (Linn 2000; Popham 1999). Preparing students to do well involves more than test preparation; it involves helping students learn to apply important reading and writing strategies.

A second caution relates to interpreting test scores. Standard assessments, by their very nature, are not precise, but rather rough approximations of student performance (Popham 1999). Furthermore, they are not good measures of students who are performing substantially below or above their grade placement. For example, most of the reading selections on a standard reading assessment designed for fourth-grade students would be at the third, fourth, or fifth-grade level. It would be unlikely for such a test to include passages at first or second grade level or at eighth or ninth grade level. As a result, students who are reading at these levels will be unable to demonstrate their abilities on the standard grade-level test. What's more, even if these students were to make gains from September to June, it would be very difficult to show gains using these tests—they simply don't have enough items at the lower and higher levels.

Finally, recent surveys have documented that only ten of fifty states provide teachers or students with feedback on how individual students perform on particular test items found on standard assessments (Education Week 2001). They simply provide overall scores, and often they provide those scores after students have moved on to another grade and another teacher. Boser (2001) concludes that "states rarely provide feedback needed for teachers and students to learn from their mistakes." So, the lack of specific and timely feedback makes it unlikely that teachers or students could use the results to direct future learning. In summary, standard assessments provide important systematic information about student learning in relation to other students or to a pre-established standard of performance. This information is particularly useful to people outside of the classroom such as legislators and administrators. Standard assessments work much like a thermometer, taking students' temperature to evaluate their academic health or abilities. However, thermometers don’t help us know exactly what is causing our illness or how to get better. For that we need finer-grained assessments and well-trained physicians. That's where classroom-based assessment comes into play.

Classroom-based Assessment
In the past several years, classroom-based assessment has enjoyed renewed support from policymakers (National Research Council 1999), assessment experts (Shepard 1999), and teachers (International Reading Association 1999) alike, giving it a central position in all assessment discussions. Recent studies suggest that teachers, themselves, are the most important assessment tool. This makes perfect sense when you realize that teachers spend a third to half of their classroom time in assessment-related activities (Stiggins and Conklin 1992) and that they make decisions about what and how to teach approximately every two to three minutes (Shavelson and Stern, 1981).

Teachers must be able to develop assessment strategies, gather evidence, analyze what they see, and ultimately, make instructional adjustments to respond to student needs. This is precisely why classroom assessment is so powerful. Classroom-based assessment is conducted close to actual learning and to children; as a result, it is most likely to be aligned with instruction, provide immediate feedback to teachers and students, engage students in assessment of their learning, and influence instructional decisions. Classroom assessment also occurs more frequently than standard or norm-referenced testing, and it can be more precisely tailored to individual children and to instruction. With classroom-based assessment, assessment and instruction are melded. Both teachers and students become learners. Teachers become more focused on what and how to teach, and students become more self-directed, motivated, and focused on learning (Graue 1993; Wolf 1989).

Classroom-based assessment includes a wide range of tools and strategies. Because the assessments grow out of actual classroom activities, they are more likely to resemble authentic reading, writing, and standard assessments (Hiebert, Valencia, and Afflerbach 1994; Wiggins 1993). For example, students might demonstrate their literacy abilities by conducting research and writing a report, developing a character analysis, debating a character's motives, dramatizing a favorite story, drawing and writing about a nonfiction piece, or reading aloud and discussing a portion of text with the teacher. These assessments can range from relatively short assessments to long-term projects. They often require students to apply their skills and strategies to new reading and writing tasks, and they often value the thinking behind work—the process—as much as the finished product (Pearson and Valencia 1987; Wiggins 1989; Wolf 1989).

Just because assessment is conducted in the classroom, doesn't make it good assessment. Research suggests that classroom assessment must have three critical features. First, it must be aligned with instruction. Although this seems obvious, teachers sometimes inadvertently hold students accountable for things they haven't adequately taught or students haven't adequately practiced (Valencia 1998). At the same time, teachers must make decisions about the most important things to assess rather than treating all learning as equally valuable (Wiggins 1989). There is nothing worse than collecting lots of information that you don’t use or that targets unimportant learning. Classroom time is too precious to waste. So, teachers must be strategic and focused as they implement classroom assessment. State and district curriculum guides, published instructional materials, national standards documents, and professional colleagues are good resources for determining important learning outcomes for students (Education Week 2001; Valencia and Place 1994).

Classroom assessment also needs to be ongoing. This implies that teachers must continually re-evaluate student learning and then use that information to adjust instruction.

Finally, good classroom assessment must rely on a variety of forms of assessment. For some students, written work is difficult, so too much reliance on written work will put them at a disadvantage (Jenkins, Johnson, and Hileman 2000). Similarly, particular activities or topics will inspire excellent performance in some students and frustrate others. And, work supported by teachers or completed collaboratively with peers may give a different impression of students' capabilities than work completed independently. Including a variety of types of assessments over time will ensure that students are provided with ample opportunities to demonstrate their abilities and that teachers' conclusions are well-founded.

Student Self-assessment
Student self-assessment may seem like an extravagant addition to the assessment system, however, both scholars and classroom experience suggest that it is an important piece of the puzzle. Students who are engaged in self-assessment do not become dependent on teachers to determine how well they are doing or where they need more work (Reif 1990). They see learning as within their control and gain a sense of responsibility and ownership. They move from passive learners (Johnston and Winograd 1985), unengaged and uninspired, to active learners. As a result, these students become more focused on their work. They learn the qualities of good work, how to judge their work against those qualities, and how to assess their own efforts and feelings of accomplishment (Reif 1990; Wolf 1989). They are more likely to set goals and to accomplish them, and consequently, their learning improves (Andrade 2000; Stiggins 1997). These students are also more likely to share common goals and expectations with their teachers (Valencia 1998). That means that teachers and students can work together, rather than at cross-purposes, because they have a shared understanding of what they want to accomplish.

Self-assessment can sometimes overemphasize superficial aspects of students' work (i.e. handwriting, drawings), efforts (i.e. I worked hard), or unexamined feelings (i.e. I like it, it's good). In fact, studies suggest that without support to go beyond the superficial, students tend not to develop a more reflective and analytic stance toward their learning (Valencia, 1998). A related caution is that self-assessment can easily become routine and uninspired if it is overused or used in the same way regardless of the kind of work. Students can grow as weary of self-assessment as any mundane activity. The antidote for such problems is to provide instruction in self-assessment (modeling, guidance, practice), time (self-assessment cannot be rushed), and many opportunities for students to discuss insights about their own learning. Like any skill or strategy, self-assessment needs support to develop.

How Can Teachers Become More Effective at Balanced Assessment?
Putting together the pieces of the assessment puzzle is difficult. On one hand, we must struggle continually to overcome tradition and the current inclination to rely on a single, standard score. As educators, we need to counter the illusion of a simple score and the almost exclusive confidence those outside education ascribe to standard assessments. On the other hand, we must learn to deal with multiple indicators (i.e. indicators from standard, classroom-based, and student self-assessment as well as multiple indicators within each type of assessment). Sometimes information from multiple sources will converge, providing a consistent evaluation of student performance; other times the information may be discrepant because of differences in assessment formats, the skills and strategies tested, or simply inconsistencies in student learning. But, it is important to value all the information and to remember that the more samples of student learning we collect, the more trustworthy and informative our results.

Suggestions for Implementing a Balanced Assessment System

• When making important educational decisions or sharing information with parents, be sure to use information from all three pieces—standard assessment, classroom assessment, and student self-assessment. Be sure you understand the purpose and focus of each assessment as well as the strengths and limitations of each. Help parents understand them as well. Use the information to describe students’ strengths and needs, rather than to label them with a grade or a number, and to plan for instruction.
   
• Focus assessment on the most important outcomes in the curriculum. Although teachers informally assess every time they interact with students and every time students work on an activity, you do not have to document every interaction or every lesson. Daily lessons and activities are often building blocks to more complex goals. Determine the most important goals you have for each unit. Then select a couple of artifacts or focal points for your anecdotal notes or checklists. Use these assessments judiciously depending on the situation, your goals, and the particular students. Collecting too much information is as problematic as not collecting enough.
   
• "Front-load" instruction. This simply means that you should be clear about the goals of instruction and make those explicit to the students. For example, if students are going to read about environmental issues and be asked to take a position, they will need to learn how to distinguish fact from opinion, synthesize information, and draw conclusions. Both you and the students have a better chance of achieving your goals if you make clear to them the relationship between the skills they are learning and the task they are completing.
   
• Help students understand what good reading and writing look like by providing them with examples, examining work together, and discussing criteria. For example, help the class develop criteria for a good research report or book talk and then have children evaluate their work according to the criteria. Use criteria and scoring rubrics provided with instructional materials with the children instead of using them just for grading.
   
• Make self-assessment a dependable, integral part of your classroom. Begin with non-academic activities such as judging how well the class is working in groups, or how sharing time is working, or discussing favorite artwork. Some of these activities require students to consider qualities of good performance; others require judgments based on personal criteria. Both, however, require students to step back from their work or their behavior to think reflectively about it. You will need to develop these abilities over time with your students.
   
• When assessing growth over time, be sure to consider both the assessment task and the individual student. For example, if you want to assess students' ability to read and summarize, the difficulty of the text and the type of text will be important to consider. A student’s summary of a second-grade text at the beginning of the year may be better than his summary of a fifth-grade text at the end of the year, but the change in difficulty level would signal growth. Similarly, the topic or text type (narrative vs. information) will influence the quality of students' summaries. Multiple measures are especially important when assessing growth.
   
• Create a system to help you keep track of the assessment information. Some teachers use a three-ring notebook with a section for each student, others use a computerized system, and others use a combination of work folders/ portfolios and teacher records. Whatever system you use, be sure to keep samples of student work and to document your assessments. These samples will help you communicate with both parents and students, providing the evidentiary trail of students' learning. If collections of work are set up collaboratively with students, they provide an excellent vehicle for conversation about students' strengths, needs, and future goals. In addition, by looking across students’ folders, you will be able to analyze your instruction. It will become obvious, for example, the kinds of activities on which students are spending most of their time and areas in which they need more support.
   
• Use classroom assessments to help with grading. You do not need to grade every piece of work or every assessment. The evidence you collect will provide the basis for the grades you assign. Some of the more formal assessments such as tests, performance activities, and projects are easier to grade. Other assessments such as oral discussions, response journals, and rough drafts of writing are more difficult to grade but still provide useful information. Together, these graded and ungraded artifacts provide strong evidence for your grading decisions.
   
• Begin classroom assessment slowly. Make good use of assessments that come with your instructional program or assessments you already have in place. You don't need to develop everything from scratch. Begin with several important outcomes, take time to review assessment results, and then use those results to shape your instruction. Classroom assessment is the piece of the puzzle that will most influence your teaching and student achievement. It takes time, but it is time well spent.

Final Thoughts
The goal of assessment is to improve learning and teaching. In truth, we cannot be good teachers nor can students be effective learners unless we use evidence to guide instruction and learning. By understanding the different types of assessments and using them wisely, we are more likely to make those good decisions. When the pieces of the puzzle are fit together, they provide a clear picture of learning and a road map to success.

Works Cited
Andrade, H.G. 2000. Using rubrics to promote thinking and learning. Educational Leadership 57: 13-18.

Blair, J. and Archer, J. 2001. NEA members denounce high-stakes testing. Education Week (web-only) 20, no. 42. http://www.edweek.org/ew/ewstory.cfm?slug=42neatest_web.h20&keywords=National%20Education%20Association.

Brennan, R.T., Kim, J., Wenz-Gross, M., and Siperstein, G.N. 2001. The relative equitability of high-stakes testing versus teacher-assigned grades: An analysis of the Massachusetts Comprehensive Assessment System (MCAS). Harvard Educational Review 71: 173–216.

Education Week. 2001. A better balance: Standards, tests, and the tools to succeed. 20, no.17.

Graue, M. E. 1993. Integrating theory and practice through instructional assessment. Educational Assessment 1, no. 4: 283–310.

Hiebert, E. H., and Calfee, R. C. 1989. Advancing academic literacy through teachers' assessments. Educational Leadership 46, no. 7: 50–54.

Hiebert, E. H., Valencia, S. W., and Afflerbach, P. P. 1994. Understand authentic reading assessment: Definitions and perspectives. In S. W. Valencia, E. H. Hiebert, and P. P. Afflerbach, eds. Authentic reading assessment: Practices and possibilities. Newark, DE: International Reading Association. International Reading Association. 1999.

International Reading Association. 1982. Misuse of grade equivalents. The Reading Teacher 35, no. 4: 464.

Jenkins, J.R., Johnson, E., and Hileman, J. 2000. Sources of individual differences on the new reading performance assessments. Unpublished manuscript.

Johnston, P.H. and Winograd, P.N. 1985. Passive failure in reading. Journal of Reading Behavior 17, no. 4: 279–301.

Linn, R. L. 2000. Assessments and accountability. Educational Researcher 29, no. 2: 4–16.

National Research Council, Committee on Title I Testing and Assessment, R.F. Elmore and R. Rothstein, eds., Board on Testing and Assessment, Commission on Behavioral and Social Sciences and Education. 1999. Testing, teaching, and learning: A guide for states and school districts. Washington, DC: National Academy Press.

Orlofsky, G.F. and Olson, L. 2001. The state of the states. Education Week 20, no. 17: 86–108.

Pearson, P. D., and Valencia, S. W. 1987. Assessment, accountability, and professional prerogative. In J. E. Readence and R. S. Baldwin, eds. Research in literacy: Merging perspectives. Rochester, NY: National Reading Conference.

Popham, J.W. 1999. Why standardized test scores don't measure educational quality. Educational Leadership 56, no. 6: 8–15.

Rief, L. 1990. Finding the value in evaluation: Self-evaluation in a middle school classroom. Educational Leadership 47, no. 6: 24–29.

Shavelson, R. J. and Stern, P. 1981. Research on teachers pedagogical thoughts, judgments, decisions and behavior. Review of Educational Research 51, no. 4: 455–498.

Shepard, L.A. 1999. The role of assessment in a learning culture. Educational Researcher 29, no. 7: 4–14.

Stiggins, R.J. 1997. Student-centered classroom assessment, 2nd ed. Upper Saddle River, NJ: Merrill.

Stiggins, R.J. and Conklin, N.F. 1992. In teachers' hands: Investigating the practices of classroom assessment. Albany, NY: State University of New York Press.

Valencia, S.W. 1998. Literacy portfolios in action. Belmont, CA: Wadsworth.

Valencia, S. W. and Place, N. 1994. Literacy portfolios for teaching, learning, and accountability: The Bellevue literacy assessment project. In S. W. Valencia, E. H. Hiebert, and P. P. Afflerbach, eds. Authentic reading assessment; Practices
and possibilities. Newark, DE: International Reading Association.

Wiggins, G. 1989. A true test: Toward more authentic and equitable assessment. Phi Delta Kappan 79, no. 7: 703–713.

Wiggins, G. 1998. Educative assessment: Designing assessments to inform and improve student performance. San Francisco, CA: Jossey-Bass.