Printer Friendly

Performance assessment: the realities that will influence the rewards.

As educators continually strive to improve evaluation methods, performance assessment has grown in popularity and use. Performance assessment is an authentic way to acquire accurate information about students' performance and comprehension (Perrone, 1991). Its appeal may be related to the growing need for local decision-making about student progress and instructional programs, as well to the interest in outcome-based education. Extensive debate about this type of assessment is carried on in education journals, professional meetings and education policy discussions. "Expert" perceptions of the "movement" range from skepticism to a belief that it is the solution to all of education's ills (Arter, 1991; Cizek, 1991).

At best, performance assessment could be the key to restructuring schools for higher standards and improved accountability. At the very least, it adds an expanded dimension to education assessment. At worst, performance assessment could become another promising idea tossed on the junk heap of discarded innovations if care is not taken to really understand its nature and limitations. As more schools move toward performance assessment for student evaluations, teachers and particularly administrators must become knowledgeable users.

Performance assessment is not a totally new idea. It is a common form of assessment in fields such as medicine and law. Industry uses performance assessment to make promotion decisions. Performance assessment was once the most common form of assessment in education, prior to the widespread use of standardized tests (Perrone, 1991). And it has been used regularly in such subjects as physical education, music and art.

Language arts teachers, in particular, have embraced the concept of performance assessment, perhaps because they have been using performance testing to varying degrees for many years (i.e., to evaluate essays, speeches, book reports and various forms of oral reading). Their cry to legitimize teacher observations of students engaged in actual literacy tasks is being heard and supported by other teachers, administrators, governors and legislators. Performance assessment of reading and writing is becoming a reality in many classrooms across the United States.

As the "movement" generates increased enthusiasm and optimism, more and more administrators are seeking help to establish performance assessment procedures not only for reading and English programs, but also for other content areas, such as math and science. The authors, as assessment consultants, have discovered that enthusiasm for and acceptance of performance assessment often exceed knowledge about its nature and complexity. As they wrestle with the task of helping administrators develop environments and structures for moving toward performance assessment, the authors have identified three basic and critical questions that often go unanswered when schools rush to jump on the performance assessment bandwagon:

* What exactly is performance assessment?

* What advice is available from experts and the research?

* What issues and concerns must be addressed?

What Exactly Is Performance Assessment?

Few discussions of performance assessment clearly define the phrase. The authors found Berk's (1986) operational definition to be a fair and concise representation of current beliefs about performance assessment. He defines performance assessment as "the process of gathering data by systematic observation for making decisions about an individual". Stiggins and Bridgeford (1986) establish three criteria for a performance assessment. First, students must apply knowledge they have acquired. Second, students must complete a clearly specified task within the context of either a real or simulated exercise. Third, the task or completed product must be observed and rated with respect to specified criteria in accordance with specified procedures, requiring students to actually demonstrate proficiency.

Stiggins and Bridgeford make a very important distinction between spontaneous and structured performance assessments. Spontaneous assessment grows out of teachers' day-to-day intuitive observations and judgments in the classroom. On the other hand, structured performance assessment must meet standards for reliability and validity and is systematically designed and planned for specific purposes. It uses clearly designed and specified scoring criteria and assesses very well-defined behaviors. Structured performance assessment is the type most performance assessment experts want us to understand.

An example of the critical distinction between structured and unstructured assessments can be found in portfolio assessment. Many articles about portfolio assessment of writing, in particular, do not suggest rigorous criteria for the selection and analysis of portfolio entries. Gathering a collection of student products is certainly a legitimate task for a teacher or school. It must be understood, however, that an unstructured collection provides no data for analysis or comparison. If performance assessment is to become a viable alternative to conventional testing, it must be rigorous. Standards for selection and analysis must be a part of every performance assessment.

Cizek (1991) makes a distinction between indirect and direct measures of student assessment. He suggests that conventional paper-and-pencil tests are indirect measures of what students know about a topic, whereas performance tests are direct measures of students' ability to employ that knowledge during an actual task. Performance assessment goes beyond measuring what students know to measuring what students can do or apply.

What Advice Is Available from Experts and the Research?

Structured performance assessment has been used for decades in business, industry, fine arts and sports. The actual sales record, product, play and game are recognized as legitimate means of assessing knowledge and ability to perform. Managing, supervising, directing and coaching focus on end products or performances. Structured performance assessment is not new to academic subjects, but it has not always been considered as "legitimate" a form of assessment as so-called objective or norm-referenced tests. Performance assessment is more likely to develop into a long-lived innovation if we pay heed to the following five suggestions offered by assessment experts and researchers.

* First, the use of performance assessment in other fields alerts us to the necessity of carefully and clearly describing and analyzing the tasks students will be expected to perform. Commenting on the business sector, Nathan and Cascio (1986) state, "A job analysis is necessary for showing the job-relatedness of all performance appraisal methods and is the basis for the performance standards fed back to employees". This same process must be applied to education. Analysis of the task that the student is to perform is a critical factor in making performance assessment a viable measurement. We cannot construct a list of criteria or a means of rating that criteria if we have not first determined what we expect students to produce.

* Second, all reliability, validity, administration and scoring standards relevant to conventional forms of assessment must also be applied to performance assessment (Brandt, 1992). Unless performance assessment is rigorous, its usefulness will be questioned. We must be wary of the misconception that any alternative assessment will automatically be a better assessment (Arter, 1991). Articles advocating portfolio assessment, for example, often do not include detailed criteria for selecting and evaluating portfolio entries (Rief, 1990) and may lead practitioners to believe that evaluation can be accomplished simply by collecting and describing samples of student work.

* Third, Grant Wiggins suggests that in order for structured performance assessment to be useful, models must be created and criteria and standards must be set (Brandt, 1992). Education does not have to invent this wheel. Established models in art, music, drama, speech and athletics may help educators develop new models appropriate for academic subjects. A significant step was taken in 1985 when a joint committee composed of representatives from American Educational Research Association, American Psychological Association and National Council on Measurement in Education established a set of standards for educational and psychological measurement that included performance assessment standards (AERA/APA/NCME Joint Committee, 1985). These standards are an excellent place for schools to begin their search for appropriate models. States such as Colorado have taken the lead in establishing models that can be tested and revised to meet local needs (Gilbert, 1990).

* Fourth, education professionals need to recognize and respond to factors that may impede further development and use of performance assessment. Developing strategies and structures for dealing with these factors may be as important as developing the new assessment instruments themselves. Speed and low cost were two attractive features of conventional tests.

Performance tests, however, are much more time-consuming to construct and administer (Maeroff, 1991). Developing, administering and scoring performance assessments are labor-intensive tasks, and the teachers who develop these new instruments will require released time. The cost of performance assessments will necessarily be weighed against their usefulness. Widespread support, as well as funding, will be the essentials that keep performance assessment from being discarded as a nonviable innovation.

* Finally, we must be aware that the performance assessment movement in education is really still in its infancy and the body of definitive research is very small. Nor do we have clear research evidence to determine if students who do poorly on conventional tests will do appreciably better on performance tests. Furthermore, there is insufficient research evidence to establish a clear link between results of performance assessment and effective functioning in college or on the job. We can, however, use evidence gathered from other professions that indicates performance assessment is much more transferable to real world tasks. Educators must continue to share what they know about the state of performance assessment if this form of evaluation is to become increasingly useful.

What Issues and Concerns Must Be Addressed?

Issues and concerns specific to the unique applications of performance assessment in education must be addressed carefully and systematically. As more and more districts move toward performance assessment, more and more issues and concerns about the design and use of the new tests will be raised. The questions generated by the following issues do not yet have "right" answers, or even general agreement on possible answers. Performance assessment cannot move ahead, however, until education planners at least come up with answers that are appropriate to their local situations.

The authors have identified four major issues and related questions that are most frequently discussed in the literature and must be answered before performance assessment can be implemented.

ISSUE 1: Establishing Standards for Performance Testing

* Is the assigned task worthy of being assessed?

* Is the end product clearly defined through models of acceptable products?

* Are the criteria for administering and scoring the test precise and clear?

* Do the scoring rubrics, or guidelines and procedures, accurately represent the agreed-upon criteria?

* Who will decide on the criteria? Will parents and students be a part of these decisions?

* How will criteria and standards be conveyed to all stakeholders (teachers, administrators, students, parents, the political community, others)?

ISSUE 2: Assuring Precision of Performance Assessment Instruments

* Is a performance test the best way to assess the behavior under consideration?

* Are uniform operational definitions available for all terms used? For example, the term "composite portfolio" has several definitions in the literature and in practice. One definition must be agreed upon if this term is to be used in a local proposal.

* How carefully structured is the relationship between the performance assessment and other tests in use (predictive validity)? For example, if college admission is a concern for a school district, will its high school writing test predict success on SATs?

* How do the criteria to be judged match the rating items, the prompt and the end product (content validity for observable behavior, construct validity for abstract concepts)? For example, if the end product and the criteria of a speech test were focused on persuasive ability while the prompt required the student to give an informational speech, the content validity of the assessment would be questionable.

* Are the performance assessment tasks related to and transferable to real world tasks (ecological validity)? For example, reading assessment tasks that involve narrative text may not be related to technical types of reading demanded in the workplace (Quellmalz, 1986).

* Will all raters who administer the performance tests achieve a reasonable degree of concurrence or inter-rater reliability?

* Will multiple samples be assessed to achieve a more reliable view of performance?

* Will the performance instrument be systematically critiqued and revised as needed?

ISSUE 3: Using Performance Assessment Scores and Results

* How will performance assessment results be translated into grades for report cards, etc. (Jongsma, 1991)?

* How will performance assessment results be used to affect classroom decisions and instructional practices?

* Will performance assessment results be used to describe individual achievement, group performance or both?

* Will performance assessment results be used in concurrence with conventional test results?

* Will the performance assessment become the new high-stakes test?

ISSUE 4: Training Personnel To Use Performance Assessments

* Are personnel who will design, administer and interpret the performance instruments knowledgeable about assessment--are they assessment literate (Stiggins, 1991)?

* Who will train personnel to administer the assessments?

* How will quality training programs be guaranteed?

* Who will train students to interpret performance assessment feedback or participate in self-assessment?

Some Final Thoughts

As knowledge and understanding of learning continue to grow, the need for more flexible and diverse measurements will only increase. Performance assessment is a promising addition to the traditional tools. If this new tool is to fulfill its promise, performance assessment must be understood and used responsibly. Educators' professional judgments may finally achieve the legitimacy and worth they deserve, provided their assessment expertise grows as well. It is the authors' hope that educators will not leave the development of this new assessment tool solely to testing companies or so-called experts, but will strive to become experts themselves.


AERA/APA/NCME Joint Committee. (1985). Standards for educational and psychological testing. Washington, DC: American Psychological Association.

Arter, J. (1991). Performance assessment: What's out there and how useful is it really? Portland, OR: Northwest Regional Educational Lab. (ERIC Document Reproduction Service No. ED 333 051)

Berk, R. A. (Ed.). (1986). Performance assessment. Baltimore, MD: The Johns Hopkins University Press.

Brandt, R. (1992). On performance assessment: A conversation with Grant Wiggins. Educational Leadership, 49, 35-41.

Cizek, G. J. (1991). Innovation or enervation? Performance assessment in perspective. Phi Delta Kappan, 72, 695-699.

Gilbert, J. C. (1990). Performance-based assessment resource guide. Denver, CO: Colorado Department of Education. (ERIC Document Reproduction Service No. ED 327 304)

Jongsma, K. S. (1991). Rethinking grading practices. The Reading Teacher, 45, 318-320.

Maeroff, G. I. (1991). Assessing alternative assessment. Phi Delta Kappan, 73, 272-281.

Nathan, B. R., & Cascio, W. F. (1986). Technical and legal standards. In R. A. Berk (Ed.), Performance assessment (pp. 1-50). Baltimore, MD: The John Hopkins University Press.

Perrone, V. (Ed.). (1991). Expanding student assessment. Alexandria, VA: Association for Supervision and Curriculum Development.

Quellmalz, E. S. (1986). Writing skills assessment. In R. A. Berk (Ed.), Performance assessment (pp. 492-509). Baltimore, MD: The John Hopkins University Press.

Reif, L. (1990). Finding the value in evaluation: Self-assessment in a middle school classroom. Educational Leadership, 47, 24-29.

Stiggins, R. J. (1991). Assessment literacy. Phi Delta Kappan, 72, 534-539.

Stiggins, R. J., & Bridgeford, N. J. (1986). In R. A. Berk (Ed.), Performance assessment (pp. 469-492). Baltimore, MD: The Johns Hopkins University Press.

Carol Anne Pierson is Associate Professor and Assistant Dean of Education, University of Central Arkansas, Conway. Shirley S. Beck is Assistant Professor, Department of Curriculum and Instruction, Southwest Texas State University, San Marcos.
COPYRIGHT 1993 Association for Childhood Education International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Beck, Shirley S.
Publication:Childhood Education
Date:Sep 22, 1993
Previous Article:What's so new about the project approach?
Next Article:Preparation of elementary teachers.

Related Articles
Use graphic metaphors to communicate organizational change.
Leading through persuasion. (Review).
Sizing up test scores. (Forum).
The business model: value-added analysis is a crucial tool in the accountability toolbox--despite its flaws. (Forum).
The levers of influence. (Positively Influencing Physicians).
Reward systems and self-managed team success.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters