Common Types of Performance Tasks Used in Credentialing

While some taxonomies exist to classify performance tasks (see, e.g., Bennett, Ward, Rock & LaHart, 1990; Reid, Raymond & Zara, 2009), we propose our own way to classify performance tasks for credentialing, and that is by the degree of their fidelity (in the neutral sense of “verisimilitude”) to a valued role performance (Linn, Baker & Dunbar, 1991; Messick, 1994). Fidelity is what makes performance tasks inherently attractive for credentialing tests, as these tasks allow the examinee to directly demonstrate the skills of interest in ways that are faithful to the practice setting (Fredricksen & Collins, 1989; Lane & Stone, 2006). Fidelity, which tends to vary inversely with standardization, may be visualized as a continuum; at the most restrictive end, a credentialing test could contain performance tasks that are in a very standardized environment, where each candidate is guided to perform the same set of narrowly circumscribed tasks. At the most naturalistic end, the required performance is a faithful representation of what a candidate would do in the role, under naturalistic conditions that are allowed to vary in substantial and meaningful ways across the tasks.

Table 15.1 lists performance tasks ordered by fidelity and standardization using three general categories, where these two factors can manifest in either the scenario in which the task is administered

Table 15.1 Task Types Organized by Level of Fidelity for the Role of the Physician

High fidelity/low standardization

Moderate fidelity/moderate standardization

Low fidelity/high standardization

  • • On-the-job observation of performance
  • • Operation of medical device in an unstructured, genuine environment
  • • Interaction with patient in unstandardized, genuine format
  • • Operation of a medical device in a structured environment
  • • Oral examination questions about simulated patient cases
  • • Interaction with high- quality simulations of patients or medical scenarios
  • • Interaction with standardized patients
  • • Written open-response or essay examination questions related to medical knowledge
  • • Interaction with software program to perform specific tasks
  • • Interpretation of a single source of patient data or information

or the task itself. A single role—that of the physician—is used, because the fidelity/standardization level of a given type of task can vary according to the professional role for which that task is used in assessment. Tasks are grouped as high, medium and low on the fidelity/standardization spectrum here purely for the sake of simplicity. For more information related to the practical aspects of constructing such tasks, readers are referred to Johnson, Penny and Gordon (2009), who cover the basic steps and important factors related to the production of a performance task in any setting, or the chapter by Lane (this volume).

< Prev   CONTENTS   Source   Next >