Intelligent tutors and instructional programs such as Facets (described in Chapter 5) and CGI share an emphasis on providing clearer benchmarks of student thinking so that teachers can understand precursors and successors to the performances they are observing in real time. Thus these programs provide a “space” of student development in which teachers can work, a space that emphasizes ongoing formative assessment as an integral part of teaching practice. Yet these approaches remain under specified in important senses. Having good formative benchmarks in mind directs attention to important components and landmarks of thinking, yet teachers’ flexible and sensitive repertoires of assistance are still essential to achieving these goals. In general, these programs leave to teachers the task of generating and testing these repertoires. Thus, as noted earlier, the effectiveness of formative assessment rests on a bedrock of informed professional practice. Models of learning flesh out components and systems of reasoning, but they derive their purpose and character from the practices within which they are embedded. Similarly, descriptions of typical practices make little sense in the absence of careful consideration of the forms of knowledge representation and reasoning they entail (Cobb, 1998).
Complex cognitively based measurement models can be embedded in intelligent tutoring systems and diagnostic assessment programs and put to good use without the teacher’s having to participate in their construction. Many of the examples of assessments described in this report, such as Facets, intelligent tutoring systems, and BEAR (see Chapter 4), use statistical models and analysis techniques to handle some of the operational challenges. Providing teachers with carefully designed tools for classroom assessment can increase the utility of the information obtained. A goal for the future is to develop tools that make high-quality assessment more feasible for teachers. The topic of technology’s impact on the implementation of classroom assessment is one to which we return in Chapter 7.
As described in Chapter 3, learning is a process of continuously modifying knowledge and skills. Sometimes new inputs call for additions and extensions to existing knowledge structures; at other times they call for radical reconstruction. In all cases, feedback is essential to guide, test, challenge, or redirect the learner’s thinking.
Simply giving students frequent feedback in the classroom may or may not be helpful. For example, highly atomized drill-and-practice software can provide frequent feedback, but in so doing can foster rote learning and context dependency in students. A further concern is whether such software is being used appropriately given a student’s level of skill development. For