1998 CRESST Conference "Reinventing Assessment and Accountability to Help All Children Learn"
Introductory Remarks
Robert Glaser, University of Pittsburgh

The title of this session "Reinventing Assessment for All Children to
Learn" is clearly a pervasive theme of this conference and a goal of the efforts of all of us. We have had a stream of thoughtful reports on work
that addresses this issue. So, in the context of this sessions and
stimulated by the past two days, let me react by saying what I believe to
be some fundamental issues or first principles in reinventing assessment. I do this briefly before turning to our real speakers who are in the main act.

I mention five issues: (1) maximizing access to education, (2) assessing high levels of achievement and competence, (3) new psychometric tactics, and methods for the (principled) design of assessment situations, (4) assisting instruction along with assessing teaching quality, and (5) the reporting of assessment outcomes. A brief comment on each of these:

1. Access to education. In order to provide access to educational opportunities and to lessen the possible exclusionary aspects of evaluation, assessment practices must be designed primarily to survey the possibilities for student growth. Assessment and instruction can be integrated so that teachers can recognize and support childrens' strengths and provide learning environments that lead to developing knowledge and opportunities for acquiring learning skills.

Briefly put, improving access to educational opportunity can be built on
at least three individual and social affordances that can be assessed.
These are: First, community-based knowledge, and competencies
available to our students that result from the demands of problem solving and reasoning in everyday life. This available competence can be related to school practices and content. Second, abilities for self-regulation of learning that enable students to monitor and exercise appropriate control over their performance. (We can call this increasing self-agency). And, third, a redirection of beliefs about the value of effort in contrast to fixed ability. So that students are aware that making the effort to develop abilities that enable themselves to learn is equally important to showing the work they have done.

2. Assessing achievement and attained competence. I can do little better here than restating what we are all striving for in assessing school achievement. We must assess not only the content of what is learned, but also the way in which this information is used, and the forms of knowing that give this knowledge intellectual and creative power, and generative competence for further learning.

A fundamental point here is taking on the problem of integrating the
findings of cognitive psychology and psychometrics. In this regard, I can refer to the foresight of a 1987 conference here at UCLA that resulted in a book edited by Merle Wittrock and Eva Baker called "Testing and Cognition." In a chapter in this book, it was pointed out that: " The assessment systems we derive depend intimately on our knowledge about the nature of human performance, and how humans learn and acquire knowledge and skill. In the present state of the art, the study of performance must be given at least equal status with measurement technique for effective approaches to subject-matter assessment. A scientific base for instruction and the assessment of outcomes cannot prosper if we have only minimal understanding of the characteristics of students' acquired performances. These characteristics must be described so that the changes that take place with learning can be specified. Thus, the study of human competence and expertise is an essential focus for determining dimensions of assessment."

Thus, we must rely on the emerging picture of the properties of acquired proficiency in school subjects and in work situations to make tests responsive to the structures and processes that develop as individuals move from beginning to advanced learners.

3. New tactics and methods. As the interaction of cognition and assessment comes about, it will predictability result in new psychometric methodologies for analyzing effective measurement properties, and also in significant changes in techniques for the design and formative construction of assessment situations. The design of assessment will take on some of the characteristics of cognitive task analysis and the experimental study of assessment features and their setting that can elicit target performance.

4. Assisting instruction and teaching quality. Of course, assessment instruments can be degraded or enhanced by the environments in which they are used and the skill of the professionals using them. Thus, a fundamental issue is the integration of teaching and assessment in instructional settings, together with the coordinate use of assessment to inform teachers about their own effective practice and the quality of their expertise.

5. The reporting and display of assessment outcomes. Much effort is being devoted to this problem, and I believe that we need to put our most inventive minds to it. National and international assessments do come up with interesting displays and reports of test results, but we need more concentrated effort on new forms of public reporting of assessment outcomes to students, parents, and the community. We need displays that not only refer to standards, but also patterns of growth and developing achievement, and profile presentations of components of performance for reflection and discussion (feedback and appropriate action). We should enlist various media and forms of information technology in this task.

Of the issues that I have just mentioned at least two or more will be considered by Bob Mislevy and Lloyd Bond. Their reputations honor us and they certainly need no introductions.