Human-scored performance items offer measurement options beyond those of standard selected-response item types, but require careful planning. This session will identify the most common mistakes made during the development and administration of performance items, as well as real-world strategies for how these mistakes can be avoided. The session will also feature a case study and an interactive exercise. The session will be structured as follows: -Each presenter will provide personal background information, then the topic will introduced to the attendees. (15 minutes) -The presenters will survey the top five mistakes in performance testing. Scott Russell and EW Looney will provide conceptual information behind each mistake, while Jack Terry will provide real-world examples for dealing with these mistakes. (45 minutes) -Presenters will engage in a question and answer period. (10–15 minutes) -An interactive exercise will be conducted in which attendees will be given the opportunity to evaluate and discuss the practices of a fictitious testing organization.
See Peak in action
The authoring tool your
psychometricians will thank you for.
Bring your program. We'll show you Peak against your real blueprints, with your real content.