CALPER Language Assessment

Center for Advanced Language Proficiency Education and Research at The Pennsylvania State University

Performance Assessment

Overview

According to the North Central Regional Educational Laboratory (NCREL) (2001), performance assessment involves “the direct, systematic observation of an actual pupil performance…and rating of that performance according to pre-established performance criteria.” Typically, performance assessments consider both the process of learner work as well as the end result. Unlike more traditional paper-and-pencil tests that target discreet features of language, performance assessments generally strive to engage learners in ‘real world’ tasks, sometimes referred to as authentic tasks, that require higher-order thinking skills such as analysis, synthesis, and evaluation.

Performance tasks vary widely, and may culminate in either a final product or a performance. In addition, performance tasks differ regarding their degree of ‘open-endedness,’ ranging from essay questions that might have a limited number of acceptable responses to tasks that stress heavily learner creativity and originality, such as song writing, games, and radio or television broadcasts.

Background

Performance assessment arose largely in response to concerns that the kinds of abilities displayed during traditional assessments such as written quizzes and tests are not good indicators of how well learners can be expected to perform in non-testing situations. In the case of language assessment, this means that researchers and practitioners began to question the degree to which scores on a language test reveal learners’ ability to use the language for various purposes in other contexts. For example, how could individuals score well on tests of language proficiency or achievement and yet struggle to successfully perform various everyday tasks in non-assessment settings? Since at least the 1990s, task-based approaches to language teaching have been very popular and have endeavored precisely to prepare learners not for traditional tests of language ability but for the kinds of activities and contexts they will likely encounter outside the classroom. In part owing to this enthusiasm for using authentic tasks in the classroom, performance-based assessments have received a good deal of attention.

Scoring performance assessments can be very complicated because they target not only learners’ knowledge of vocabulary, grammatical structures, pronunciation, and other features of language but also their ability to bring this knowledge to bear on the task at hand. In addition, learners often rely on other abilities and domains of knowledge to complete such tasks. When designing performance assessments, it is important to take account of the task objectives, the expected products, the resources students will need, whether the activity will be done in an individual, pair, or group format, and the amount of time to be devoted to each stage of the task. It is also recommended that students be provided with explicit guidelines and templates at every stage as well as a copy of the rubric or criteria that will be employed to evaluate their performance.

Application

Given the almost unending number of tasks that might be used as a performance assessment, the examples below are merely a sampling of some common approaches.

Spence-Brown (2001) developed a performance assessment for learners of L2 Japanese that required the learners to seek out Japanese-speaking immigrants and to conduct interviews with them in the target language. In Spence-Brown’s model, the task was not only an assessment but also a learning activity because students needed to prepare over time the topics to be covered in the interview and the linguistic and cultural resources and knowledge needed to successfully carry out the interview.

Douglas (2001) explains that performance assessments are increasingly common in language education for specific purposes. Take, for instance, a Spanish language course designed to prepare students for professional work in the medical field. Learners are asked to engage in a role play activity that simulates the demands and requirements they are likely to encounter in hospital settings. One learner is assigned the role of a patient and directed to describe symptoms related to an illness while the other student attempts to diagnose the illness and explain treatment options. The students are given time to research the linguistic resources needed for the task and to rehearse prior to performing the role play for an evaluator.

In a content-based class combining history and language, advanced learners of French are studying the European exploration and settlement of the New World. They are asked to write a letter in French to family members in the Old World who are considering immigrating. In this case, successful performance depends upon a number of factors, including learners’ knowledge of the historical period, ability to write an expository (and perhaps persuasive) text, and their appropriate use of French to achieve these goals. A teacher might consider developing specific rubrics for each of these facets of performance.

 


Suggested Readings and References: 

    • Bachman, L. F. (2002). Some reflections on task-based language performance assessment. Language Testing, 19, 453 – 476. This article provides a critical review of issues and trends in language performance assessment research and practice.
    • Douglas, D. (2001). Language for specific purposes assessment criteria: Where do they come from? Language Testing, 18, 171-185. The author suggests an analysis of the target language use context as a starting point for developing valid criteria to evaluate performance assessments.
    • Norris, J. M., Brown, J. D., Hudson, T., & Yoshioka, J. (1998). Designing second language performance tasks.Honolulu, HI: University of Hawaii Press.
    • McNamara, T. F. (2001). Language assessment as social practice: Challenges for research. Language Testing, 18, 333 – 349. This article explores the problem of interpreting performance during assessments and considers the influence of task specifications and the assessor on the resulting performance.
    • Spence-Brown, R. (2001). The eye of the beholder: Authenticity in an embedded assessment task. Language Testing, 18, 463-481. This paper describes how a performance assessment for learners of L2 Japanese is interpreted and carried out differently by individual students.

There are two archived webinars from the LARC/CALPER Webinars on Assessment series, which might be of interest to you.

  • John Norris on “How Do We Assess Task-based Performance.” Access here
  • Meg Malone on ‘Performance Assessment.” Access here

Next section: Peer Assessment

Skip to toolbar