The question "Why use authentic assessment?" is not meant to suggest that you have to choose between traditional assessments such as tests and more authentic or performance assessments. Often, teachers use a mix of traditional and authentic assessments to serve different purposes. This section, then, attempts to explain why teachers might choose authentic assessments for certain types of judgments and why authentic assessments have become more popular in recent years.

Authentic Assessments are Direct Measures

We do not just want students to know the content of the disciplines when they graduate. We, of course, want them to be able to use the acquired knowledge and skills in the real world. So, our assessments have to also tell us if students can apply what they have learned in authentic situations. If a student does well on a test of knowledge we might infer that the student could also apply that knowledge. But that is rather indirect evidence. I could more directly check for the ability to apply by asking the student to use what they have learned in some meaningful way. To return to an example I have used elsewhere, if I taught someone to play golf I would not check what they have learned with just a written test. I would want to see more direct, authentic evidence. I would put my student out on a golf course to play. Similarly, if we want to know if our students can interpret literature, calculate potential savings on sale items, test a hypothesis, develop a fitness plan, converse in a foreign language, or apply other knowledge and skills they have learned, then authentic assessments will provide the most direct evidence.

Can you think of professions which require some direct demonstration of relevant skills before someone can be employed in that field? Doctors, electricians, teachers, actors and others must all provide direct evidence of competence to be hired. Completing a written or oral test or interview is usually not sufficient. Shouldn't we ask the same of our students before we say they are ready to graduate? Or pass a course? Or move on to the next grade?

Authentic Assessments Capture Constructive Nature of Learning

A considerable body of research on learning has found that we cannot simply be fed knowledge. We need to construct our own meaning of the world, using information we have gathered and were taught and our own experiences with the world (e.g., Bransford & Vye, 1989; Forman & Kuschner, 1977; Neisser, 1967; Steffe & Gale, 1995; Wittrock, 1991). Thus, assessments cannot just ask students to repeat back information they have received. Students must also be asked to demonstrate that they have accurately constructed meaning about what they have been taught. Furthermore, students must be given the opportunity to engage in the construction of meaning. Authentic tasks not only serve as assessments but also as vehicles for such learning.

Authentic Assessments Integrate Teaching, Learning and Assessment

Authentic assessment, in contrast to more traditional assessment, encourages the integration of teaching, learning and assessing.  In the "traditional assessment" model, teaching and learning are often separated from assessment, i.e., a test is administered after knowledge or skills have (hopefully) been acquired.  In the authentic assessment model, the same authentic task used to measure the students' ability to apply the knowledge or skills is used as a vehicle for student learning.   For example, when presented with a real-world problem to solve, students are learning in the process of developing a solution, teachers are facilitating the process, and the students' solutions to the problem becomes an assessment of how well the students can meaningfully apply the concepts.

Authentic Assessments Provide Multiple Paths to Demonstration

We all have different strengths and weaknesses in how we learn. Similarly, we are different in how we can best demonstrate what we have learned. Regarding the traditional assessment model, answering multiple-choice questions does not allow for much variability in how students demonstrate the knowledge and skills they have acquired. On the one hand, that is a strength of tests because it makes sure everyone is being compared on the same domains in the same manner which increases the consistency and comparability of the measure. On the other hand, testing favors those who are better test-takers and does not give students any choice in how they believe they can best demonstrate what they have learned.

Thus, it is recommended (e.g., Wiggins, 1998) that multiple and varied assessments be used so that 1) a sufficient number of samples are obtained (multiple), and 2) a sufficient variety of measures are used (varied). Variety of measurement can be accomplished by assessing the students through different measures that allows you to see them apply what they have learned in different ways and from different perspectives. Typically, you will be more confident in the students' grasp of the material if they can do so. But some variety of assessment can also be accomplished within a single measure. Authentic tasks tend to give the students more freedom in how they will demonstrate what they have learned. By carefully identifying the criteria of good performance on the authentic task ahead of time, the teacher can still make comparable judgments of student performance even though student performance might be expressed quite differently from student to student. For example, the products students create to demonstrate authentic learning on the same task might take different forms (e.g., posters, oral presentations, videos, websites). Or, even though students might be required to produce the same authentic product, there can be room within the product for different modes of expression. For example, writing a good persuasive essay requires a common set of skills from students, but there is still room for variation in how that essay is constructed.

Copyright 2018, Jon Mueller. Professor of Psychology, North Central College, Naperville, IL. Comments, questions or suggestions about this website should be sent to the author, Jon Mueller, at jfmueller@noctrl.edu.