corollary for these students is that there is little point in trying or hoping for better.

The way in which information is provided is therefore a delicate matter. Grades, and even undue praise, can reinforce expectations of failure and lead to reluctance to invest effort. Yet this culture is deeply embedded in American schools and is hard to change. This fact highlights the importance of the nature and form of the information provided to students. Thus, priority should be given to providing students with information that they can use to reach desired learning goals (Ames, 1992; Butler, 1988; Dweck, 1986).

In helping teachers and students establish where students stand in relation to learning goals, assessment activities are not only useful during and at the end of a unit of teaching, they also can be valuable at the start of a piece of work. Suitably open and nontechnical questions or activities can stimulate students to express how much they already know and understand about a topic. This may be particularly important when the students come from a variety of backgrounds, with some having studied aspects of the topic before, either independently or with other teachers in different schools. Such assessment can both stimulate the thinking of the students and inform the teacher of the existing ideas and vocabularies from which the teaching has to start and on which it has to build.

The following example from the Lawrence Hall of Science assessment handbook (Barber et al., 1995) demonstrates how assessment mechanisms can enrich science investigations and provide the teacher with useful information. In this illustration, students are challenged to design and conduct two experiments to determine which of three reactants —baking soda, calcium chloride, and a phenol red solution (phenol red and water)—when mixed together produces heat. The students already have completed an activity in which they mixed all three substances. The students are expected to refer to their observations and the results of that first activity. Box 3-4 illustrates a data sheet used by the students for the assessment activity, which provides prompts to record their experimental design and observations. Through this investigation, the teacher would be able to assess students' abilities to do the following:

• Design a controlled experiment in which only one ingredient is omit-