POLICY RESEARCH: AUTHENTIC ASSESSMENT

credit: www.ioxassessment.com/index.php?cPath=37

Education Resources Information Center (ERIC) has used the term “performance test” since 1966, and education journals have devoted full issues to the subject of alternative assessment since 1989 (Rudner & Boston, 1992). Today, alternatives to traditional standardized testing are most commonly referred to as “authentic assessment”, though also known as “performance assessment,” “alternative assessment,” and “direct assessment” (Mueller, 2008).

What is Authentic Assessment?

Authentic assessment involves teaching and learning in which the assessment, product, and process of student work are one and the same, and the task at hand involves significant implementation of knowledge and skills in a real-world and intellectual context, with explicit and concrete criteria for success.

What Does Authentic Assessment Look Like?

Authentic assessment can be roughly categorized into six types, though it must be stressed that any type of authentic assessment can be done more or less “authentically.”  These categories alone do not make a certain kind of assessment relevant and applicable to real-world situations; instead, the teacher must ensure that the task given is closely related to an experience that might be encountered by a professional in that particular discipline.

The six broad categories of authentic assessment are:

• performance
• portfolio
• group learning
• open-ended/constructed
• experimental
• self assessment (CTER WikEd, 2009; Sweet, 1993; Bowers, 1989).

Why Use Authentic Assessment?

• Products of authentic assessment have the potential to demonstrate to student examples of excellent work in a discipline (J. Mehta, personal communication, November 13, 2009).

• According to Wiggins (1990), students benefit from authentic assessment because it provides “greater clarity about their obligations.” Teachers benefit, too, as they “come to believe that assessment results are both meaningful and useful for improving instruction.” Proponents of authentic assessment agree that its merits include:

                    • student choice, which leads to greater motivation and engagement;

                    • multiple levels of work as well as multiple opportunities to demonstrate understanding; collaboration in process and sharing of product;

                    • application and transfer of skills and knowledge;

                    • and relevance to real life, with worthwhile activities.

• Authentic assessment is a more valid indicator of student capabilities that traditional testing because it is a direct measure of student understanding, rather than a representative model (Coalition of Essential Schools, 2002; Eduplace, 1997; Sweet, 1993; Mueller, 2008; Mandernach, 2003; Bowers, 1989).

Who Uses Authentic Assessment?

Several alternative education programs, independent schools, and charter schools use authentic assessment. Waldorf schools use portfolios, while Essential Schools (for example, Central Park West) use exhibitions. Big Picture schools use an array of authentic assessments, and several states, including Maryland, California, Arizona, New York, Connecticut, Vermont, and Kentucky have been working to implementing authentic assessments that are in line with their state standards (Sweet, 1993).

Criticisms of Authentic Assessment

There are two main criticisms of authentic assessment: it is too costly and too unreliable. If the costs of evaluating authentic assessments are restrictive at scale, sampling may be an effective solution: either sampling a small number of students or sampling a small amount of all students’ work could ease any potential financial burden (Wiggins, 1990).

Bowers (1989) argues that the question is not one of deciding whether or not to worry about reliability, but to weigh reliability versus validity.  Traditional standardized tests are effective in their ability to “sort large numbers of students in as efficient a manner as possible,” making them exceedingly reliable. Authentic assessment, however,  “actually test[s] what the educational system is presumably responsible for teaching, namely, the skills prerequisite for performing in the real world”.

Bias is a concern in any evaluative endeavor, and one that should be taken seriously. Even in traditional testing, however, as Wiggins (1990) points out, while the scoring is hypothetically unbiased, the question creation is performed by humans who go unchecked by the public.  Nonetheless, several measures can be taken to ensure that bias is held to a minimum: training, blind readings, and audits are all examples of monitoring evaluation. Initiating the widespread use of rubrics that align tightly with standards, regardless of authentic assessment product form (portfolio, oral defense, etc.) is a potential method of controlling inter-rater reliability.

Read the full report

Back home