Why test: To Predict How People Will Perform
Conventional multiple-choice testing measures
the examinee’s ability to articulate knowledge – specifically
what can be called declarative knowledge. Declarative knowledge
is largely the recitation of facts and information. Performance
Testing is designed to see whether an examinee can put that
knowledge to use.
Some examples:
- Working through a solution is the trial-and-error process that most of us engage in while doing conventional tasks.
- Most of us don’t memorize the steps for doing a problem.
- In doing a task on a Mac or Windows computer, we try one menu – only to determine the necessary feature is on another menu. We open a tab in a dialog box, then find the parameter we want on another tab. These probes and restarts are the natural part of working through a solution – and they are specifically what can’t be done in a multiple-choice environment.
Here's a more detailed example, that of solving a problem by looking up reference information. Using reference materials requires multiple stages to assess:
- What trigger event or association makes the connection between the problem and reference materials?
- How can the solution to the problem be accessed?
- How can the reference materials be translated into action and applied to resolve the issue?
- How can the applied solution be evaluated?
By breaking a reference problem down into its components, multiple-choice testing does the most complex part of the problem for the examinee. Problem-solving is a complex process of using a repertoire of tools and techniques to solve a problem that doesn't have a defined, procedural solution. Because the person approaching the problem can't be expected to know in advance which tool or technique will solve the problem, a multiple-choice question cannot ask the person which technique to use (without cueing the answer). A complex problem requires multiple stages to pose and evaluate solutions – a situation which multiple-choice questions cannot address.
Why Performance Test?
To Measure Intellectual Capital
To Measure the Need for Training
To Assess the Effectiveness of Training
