Definition:  Measurement using a series of tasks related to a domain of interest.

          Description.  Here participants demonstrate their abilities through doing the tasks of interests rather than reporting on their ability or perceptions about the task.  Often participants also provide an explanation of the strategy they employed to perform the task (Gredler, 1996).  In school settings, performance assessment focuses on content such as writing, mathematics, and science.  In contrast to multiple-choice tests, testers evaluate the resulting performance by scoring such components as the solution strategies employed as well as the type of errors.  Such information is particularly useful for designing instructional and intervention programs to improve performance.

Test developers consider the amount of structure present with a performance task, such as how extensive the instructions should be and whether any clues should be provided.  If no structure is provided, the task may be too ambiguous for the respondent; the resulting performance may provide no relevant information about the test-taker’s abilities.  If too structured, respondents may be forced to employ just one method of solving the task.  Test developers must also decide the extent to which they will focus on the product of the performance as well as the respondent’s process.  This decision is important because one cannot assume that process and product are linked in the same way for each individual.

Similar alternative methods include portfolio assessments, simulations, and work assessment centers (Gredler, 1996).  Portfolios are collections of work that represent actual performance in school or work settings.  Simulations attempt to present aspects of actual situations and then observe respondents’ behavior.  Work assessment centers assess individuals in small groups through such tasks as situational tests (e.g., leaderless group discussion) and interviews.  Read more here and watch this video about performance assessment in education.