Assessment

Portfolio assessment is a term with many meanings, and it is a process that can serve a variety of purposes. A portfolio is a collection of student work that can exhibit a student's efforts, progress, and achievements in various areas of the curriculum. A portfolio assessment can be an examination of student-selected samples of work experiences and documents related to outcomes being assessed, and it can address and support progress toward achieving academic goals, including student efficacy. Portfolio assessments have been used for large-scale assessment and accountability purposes (e.g., the Vermont and Kentucky statewide assessment systems), for purposes of school-to-work transitions, and for purposes of certification. For example, portfolio assessments are used as part of the National Board for Professional Teaching Standards assessment of expert teachers.

The Development of Portfolio Assessment

Portfolio assessments grew in popularity in the United States in the 1990s as part of a widespread interest in alternative assessment. Because of high-stakes accountability, the 1980s saw an increase in norm-referenced, multiple-choice tests designed to measure academic achievement. By the end of the decade, however, there were increased criticisms over the reliance on these tests, which opponents believed assessed only a very limited range of knowledge and encouraged a "drill and kill" multiple-choice curriculum. Advocates of alternative assessment argued that teachers and schools modeled their curriculum to match the limited norm-referenced tests to try to assure that their students did well, "teaching to the test" rather than teaching content relevant to the subject matter. Therefore, it was important that assessments were worth teaching to and modeled the types of significant teaching and learning activities that were worthwhile educational experiences and would prepare students for future, real-world success.

Involving a wide variety of learning products and artifacts, such assessments would also enable teachers and researchers to examine the wide array of complex thinking and problem-solving skills required for subject-matter accomplishment. More likely than traditional assessments to be multidimensional, these assessments also could reveal various aspects of the learning process, including the development of cognitive skills, strategies, and decision-making processes. By providing feedback to schools and districts about the strengths and weaknesses of their performance, and influencing what and how teachers teach, it was thought portfolio assessment could support the goals of school reform. By engaging students more deeply in the instructional and assessment process, furthermore, portfolios could also benefit student learning.

Types of Portfolios

While portfolios have broad potential and can be useful for the assessments of students' performance for a variety of purposes in core curriculum areas, the contents and criteria used to assess portfolios must be designed to serve those purposes. For example, showcase portfolios exhibit the best of student performance, while working portfolios may contain drafts that students and teachers use to reflect on process. Progress portfolios contain multiple examples of the same type of work done over time and are used to assess progress. If cognitive processes are intended for assessment, content and rubrics must be designed to capture those processes.

Portfolio assessments can provide both formative and summative opportunities for monitoring progress toward reaching identified outcomes. By setting criteria for content and outcomes, portfolios can communicate concrete information about what is expected of students in terms of the content and quality of performance in specific curriculum areas, while also providing a way of assessing their progress along the way. Depending on content and criteria, portfolios can provide teachers and researchers with information relevant to the cognitive processes that students use to achieve academic outcomes.

Uses of Portfolios

Much of the literature on portfolio assessment has focused on portfolios as a way to integrate assessment and instruction and to promote meaningful classroom learning. Many advocates of this function believe that a successful portfolio assessment program requires the ongoing involvement of students in the creation and assessment process. Portfolio design should provide students with the opportunities to become more reflective about their own work, while demonstrating their abilities to learn and achieve in academics.

For example, some feel it is important for teachers and students to work together to prioritize the criteria that will be used as a basis for assessing and evaluating student progress. During the instructional process, students and teachers work together to identify significant pieces of work and the processes required for the portfolio. As students develop their portfolio, they are able to receive feedback from peers and teachers about their work. Because of the greater amount of time required for portfolio projects, there is a greater opportunity for introspection and collaborative reflection. This allows students to reflect and report about their own thinking processes as they monitor their own comprehension and observe their emerging understanding of subjects and skills. The portfolio process is dynamic and is affected by the interaction between students and teachers.

Portfolio assessments can also serve summative assessment purposes in the classroom, serving as the basis for letter grades. Student conferences at key points during the year can also be part of the summative process. Such conferences involve the student and teacher (and perhaps the parent) in joint review of the completion of the portfolio components, in querying the cognitive processes related to artifact selection, and in dealing with other relevant issues, such as students' perceptions of individual progress in reaching academic outcomes.

The use of portfolios for large-scale assessment and accountability purposes pose vexing measurement challenges. Portfolios typically require complex production and writing, tasks that can be costly to score and for which reliability problems have occurred. Generalizability and comparability can also be an issue in portfolio assessment, as portfolio tasks are unique and can vary in topic and difficulty from one classroom to the next. For example, Maryl Gearhart and Joan Herman have raised the question of comparability of scores because of differences in the help students may receive from their teachers, parents, and peers within and across classrooms. To the extent student choice is involved, contents may even be different from one student to the next. Conditions of, and opportunities for, performance thus vary from one student to another.

These measurement issues take portfolio assessment outside of the domain of conventional psychometrics. The qualities of the most useful portfolios for instructional purposes–deeply embedded in instruction, involving student choice, and unique to each classroom and student–seem to contradict the requirements of sound psychometrics. However, this does not mean that psychometric methodology should be ignored, but rather that new ways should be created to further develop measurement theory to address reliability, validity, and generalizability.