Creating Tests Worth Taking.

T he teaching of higher-level thinking operations is a topic that interests many of today's educators. These operations include comprehension and interpretation of text, scientific processes, and mathe matical problem solving. While much has been written on the need for students to perform higher-level thinking operations in all subject areas, the teaching of these operations often fails, not because the idea is poor, but because the instruction is inadequate. How does one help students perform higher-level operations? One solution that researchers have devel oped is to teach students cognitive strategies (Pressley et al. 1990: Perkinsetal. 1989; Weinstein 1979). A strategy is not a direct procedure; it is not an algorithm. Rather a strategy is a heuristic that supports or facili tates the learner as he or she leams to perform the higher-level operations. For example, to facilitate reading comprehension, students may be taught to use cognition strategies such as generating questions about their reading. To generate questions, students need to search the text and combine information, which in turn helps them comprehend what they read. To help students in the writing process, they may be taught how to organize their writing and how to use self-talk prompts to facilitate the revi sion process. These cognitive strate gies are more like supports or sugges tions than actual step-by-step directives. But how does one teach cognitive strategies? Our review of about 50 studies in which students ranging from 3rd grade through college were taught cognitive strategies showed that successful teachers of such strategies frequently used instructional proce dures called scaffolds ( Palincsar and Brown 1984; Paris et al. 1986; Wood et al. 1976). Scaffolds are forms of support provided by the teacher (or another student) to help students bridge the gap between their current abilities and the intended goal. Scaf folds may be tools, such as cue cards, or techniques, such as teacher modeling. Although scaffolds can be applied to the teaching of all skills, they are particularly useful, and often indispensable, for teaching higherlevel cognitive strategies, where many of the steps or procedures necessary to carry out these strategies cannot be specified. Instead of providing explicit steps, one supports, or scaffolds, the students as they learn the skill. The support that scaffolds provide is both temporary (Tobias 1982) and adjustable, allowing learners "to participate at an ever-increasing level of competence" (Palincsar and Brown 1984, p. 122). Scaffolding gradually decreases as the learning process unfolds and students become profi cient. Before using scaffolds, it is impor tant to determine whether students have sufficient background ability to leam a new cognitive strategy. Researchers (particularly Palincsar and Brown 1984) note that scaffolds are only useful within the student's "zone of proximal development" (Vygotsky 1978), that is. the area where the student cannot proceed alone, but can proceed when guided by a teacher using scaffolds. When Palincsar and Brown (1984) taught strategies designed to foster reading comprehension, they selected students whose decoding skills were near grade level, but whose comprehension was below grade level. They did not select students with poor decoding skills, because such students did not have sufficient background skills to profit from this instruction. Similarly, scaf folds cannot help students read a physics text or history text for which they do not have the necessary back ground knowledge.