While we often think of tests as a relatively recent creation, perhaps the result of the industrial revolution, the notion of testing dates back nearly to the dawn of civilization. Those early tests measured practical skills and abilities and were often used as, what we in the 21st century would consider, pre-employment and placement exams. With the advent of a more diversified workforce came the introduction of guilds (associations) and apprenticeships that tightly coupled training and assessment into a structured delivery model. Role delineation in the military, public education and the explosion of specialized professions early in the last century all contributed to the rapid adoption of highly standardized multiple-choice exams. Multiple choice tests proved to be highly reliable, easily delivered and nearly universal in their applicability. And they remain so.
Fast forward to today…an increasingly global economy where people sometimes misrepresent themselves on resumes and in job interviews, falsifying or exaggerating their skills and abilities to gain a competitive edge. Couple this with the high degree of variability that exists in the quality of our education system and you have an environment that exposes new vulnerabilities for credentialing agencies and society at large. It is little wonder then, that test sponsors are seeking greater confidence in their certification and licensure decisions. That search has focused on the concept of performance-based testing – measurement of an individual's ability to demonstrate specific skills and/or to perform a series of assigned tasks. While hands-on practical exams, in which candidates are evaluated in the workplace, have been administered for centuries, today's test sponsors need a delivery model that leverages the consistency, security and automation made possible with several emerging technologies.
Early adopters of performance-based testing have learned several important lessons that are worth consideration. Performance-based items are far more expensive and time-consuming to develop than their knowledge-based counterparts and while a single multiple-choice item might contribute to several test objectives, the individual tasks within a performance-based activity are typically much more narrowly defined. As a consequence, performance-based items are not as efficient from an item bank perspective. Also, because these activities are generally more time-consuming, fewer are administered within a test and they are therefore more memorable than traditional item types – raising item exposure and performance drift concerns.
Does all of this mean that performance-based testing is impractical for all but a few large, well-funded testing programs? Not at all! The answer lies in taking a holistic view toward what needs to be accomplished. The solution for many organizations is hybrid testing that leverages both performance-based scenarios and traditional objective test content to provide a more comprehensive gauge of a candidate's true knowledge, skills and abilities. Not only does this approach preserve your organization's existing item bank investment, it also substantially reduces the cost and level of effort associated with a move toward performance measurement.
The future of testing has arrived and a combination of innovative technology and a holistic view of the knowledge, skills and abilities required to measure competency will take us into the next generation of assessments. It's reassuring to know that the fundamentals that we've all come to rely on are truly universal and continue to stand the test of time.