Performance-based assessments can “provide students with rich, contextualized, and engaging tasks and allow students to choose or design tasks or questions that are meaningful to them” (Lam, 1995, ¶ 9). Such individualization can help produce bias-free scores in a way that standardized tests cannot because the tasks can be used with students of varying learning experiences and backgrounds. Performance-based assessments allow instructors to use methods most appropriate for each student, which minimizes bias by considering extraneous factors in the assessment design that can affect student performance.
Keywords Analytic Rubric; Authentic Assessment; Experiential Learning; High-Stakes Testing; Holistic Rubric; No Child Left Behind Act of 2001 (NCLB); Goals and Objectives; Performance-Based Assessment; Portfolio; Rubrics; Scoring Rubric; Summative Assessment; Test Bias
Performance-based assessments require students to utilize and prove their depth of knowledge and skill-set. This can be accomplished in many ways, including writing exercises, completing mathematical computations, and conducting experiments. Performance-based assessments can range from very basic in nature to more comprehensive collections of work over time, such as an entire school term. Regardless of the format, performance-based assessments tend to have several features in common. They also require direct observation of student behavior instead of grading a paper to determine student competence (Elliott, 1995).
Performance-based assessments can “provide students with rich, contextualized, and engaging tasks and allow students to choose or design tasks or questions that are meaningful to them” (Lam, 1995, ¶ 9). Such individualization can help produce bias-free scores in a way that standardized tests cannot because the tasks can be used with students of varying learning experiences and backgrounds. Performance-based assessments allow instructors to use methods most appropriate for each student, which minimizes bias by considering extraneous factors in the assessment design that can affect student performance. For example, instead of using a paper-and-pencil word problem test in mathematics, an instructor could present the questions orally to ESL students, rewording the sentences or using another language in order for the student to better comprehend the questions. Performance-based assessments also allow students an opportunity to defend and explain to the instructor their project and how they arrived at their conclusions. (Lam, 1995).
Basis in Authentic Assessment
Performance-based assessments are a type of authentic assessment and can be used in conjunction with more traditional types of assessment. For example, if an instructor would like to focus on recall of facts, then a multiple choice or short answer assessment may be an appropriate choice. However, when instructors are focused on more complex learning outcomes such as reasoning, communication, and teamwork, then a performance-based assessment may be the more appropriate choice (Perlman, 2002, as cited in Moskal, 2003a). Performance-based assessments do not have only one right answer. Instead, there are levels of proficiency that may be attained by students. Performance-based assessments can come in different formats, and the activities may be completed in either a group or individually. A major difference between performance-based assessments and other forms of assessment is that performance-based assessments require students to demonstrate their knowledge in a particular context (Brualdi, 1998; Wiggins, 1993, as cited in Moskal, 2003a).
Performance-based assessments gained popularity in the early 1990s when there was a movement toward curriculum reform that promoted a more hands-on, experiential learning. They were conceived to help develop students' higher order thinking and reasoning skills and not just test their ability to memorize facts and details or calculate. Performance-based assessments often consist of a problem to solve or a concrete task to complete, and students may be judged on their ability to investigate, the methods they use, the logic they display, and the conclusions they develop as well as whether or not they correctly solved the problem. Performance-based assessments may also require students to work in a group setting or participate in group discussions to learn real-world skills they will need later in life. Performance-based assessments may not have a single correct approach and tend to better assess students' ability to use the knowledge they have attained, whereas multiple choice, true or false, and fill-in-the blank questions can sometimes measure students' test-taking skills more than their knowledge of the subject matter (Seal, 1993).
Types of Performance-Based Assessments
Performance-based assessments can be broken down into five categories:
• Portfolios: a collection of student work that represents a student's progress and activities and may include drafts of student work to show the evolution of a project.
• On-Demand Tasks: require students to answer an instructor's prompt or respond to a problem within a short period of time with little in interpretation.
• Projects: last longer than on-demand tasks, such as an entire term, and can require working in a group setting.
• Exhibitions: presentations of various kinds of student work and are similar to a portfolio but shared with more than instructors and parents.
• Instructor Observations: should be unobtrusive and occur primarily to rate student performance and also to determine students' strengths and weaknesses (Kane & Khattri, 1995).
Use of Rubrics
Unlike most other forms of assessment, performance-based assessments do not have only one right answer. Instead, there are levels of proficiency which may be attained by students, which means instructors need an instrument that will allow them to rate each student's performance. This can most easily be accomplished by using a rubric. A rubric is a rating sheet that allows instructors to determine the level of competency each student has achieved for each concept being assessed (Brualdi, 1998). Most rubrics list levels of competency based on numbers or impartial phrasing. Four levels tend to be the preferable number, such as:
• "Exceeds Expectations,"
• "Needs Improvement,"
Whereas three levels can limit the rubric to
• "Consistently Successful,"
• "Making Progress,"
• "Needs Improvement."
This can make it difficult to translate into a letter grade or score for summative assessment grading purposes and report cards (Andrade, 2000).
Since performance-based assessments assess more than one skill, rubrics are generally used to score the assessment. Rubrics for portfolios need to be somewhat generic, and rubrics used for projects, on-demand tasks, and exhibitions should be tailored to the specific assignment. To help students understand the connection between science and medicine, an example of a performance-based assessment would be to have students choose a specific topic that interests them, explain the topic's relationship to the human body, and then outline what advances still need to be made regarding the chosen topic. Students would be assessed on the thoroughness and quality of their research, their writing ability, and the visual presentation they produce, all of which would be addressed by the scoring rubric. A performance-based assessment for mathematics could include having students complete three different categories of problems-puzzles, investigations, and applications-with students assessed on their ability to communicate and to apply the concepts to word problems (Kane & Khattri, 1995).
Before a performance-based assessment or scoring rubric is developed, instructors must clearly identify the purpose of the activity, which will help guide the development of both the assessment and rubric.
• Goals are the broad statements of expected student outcomes,...
(The entire section is 3667 words.)