The final goal of the mini-sabbatical is to help teachers develop assessments that reflect diverse learning goals and provide meaningful feedback to teachers and students. Popular forms of assessment test students' knowledge of facts, concepts, and processes in a domain. They rarely assess students' ability to solve problems, reason, cooperate with others, or demonstrate other skills and capabilities attained in situated learning environments. They also are not always able to help teachers and students diagnose learning successes and failures in ways that help modify and improve on the learning process (Stasz et al., 1992).
In addition to a general need for valid, reliable, and affordable methods for assessing skills, teachers need a systematic approach for choosing among assessment methods for their particular needs. One such approach, which we adopted for the mini-sabbatical, guides teachers through the following steps:[4]
Different types of assessment strategies are used for different purposes, and these purposes determine how to measure knowledge and skills. A paper and pencil test is most appropriate, for example, for assessing students' knowledge of mathematical or history facts. In contrast, a performance event, such as a group-led experiment or problem-solving exercise, is a more fitting strategy for assessing a student's ability to think and solve problems as a member of a team.
Written tests are popular types of assessments, including multiple-choice or open-ended items, essays, and problem- or scenario-based items. A second type, performance tasks, may consist of one or a set of multiple physical tasks, such as giving a speech or changing the oil in a car engine.
To ensure that an assessment strategy will provide accurate information, the technical quality of the measures must be considered. Three aspects of quality are of particular concern: (1) reliability: How accurate is the information? (2) validity: Does the assessment measure what it is intended to measure? (3) fairness: Is the assessment free of biases against any group of students? Higher reliability, or degree of accuracy of an assessment, enhances fairness.
Finally, a teacher must consider a number of practical issues. Is the assessment feasible in terms of the cost and time required to administer and score? Alternative assessments, like performances or portfolios, are more expensive to develop, administer, and score than selected-response tests. Complexity is also an issue; alternative assessments are often more complex than traditional tests because they can require special materials (e.g., manipulatives) or special training for administration and scoring. During the mini-sabbatical, teachers were introduced to the assessment concepts. They then discussed ways to assess student work in their individual curriculum units.