Benefits of portfolios – particularly electronic portfolios – are easy to list. They allow students to reflect on their learning experiences, tie their learning to their other learning experiences and to real world applications, gain practical computer experience, and assemble a compilation of their notable work (Holtzman & Dagavarian, 2007). But as Barrett (2007) points out in a review of existing research on portfolios, the body of research on portfolios for K-12 schools is slim. Most of the research instead focuses its attention on the training of teachers and the need for a professional portfolio (p. 436). A quick search through popular databases confirms the claim. The dearth of research means that a basis for researching portfolios generally has not been present, and Barrett (2007) notes that even as recently as 2004 researchers have been developing frameworks for researching electronic portfolios (p. 437). A further problem with the literature on portfolios is that though there is much of it (and most of that being favorable), little of it documents empirical, comparative research, leaving little evidence on what actual effect portfolios have on students’ performance (Struyven, Dochy, Janssens, Scheifhout, & Gielen, 2006).
Knowing this, the important question to an inquiring educator is fairly obvious: What does research say about the effects of using portfolios in K-12 education? The question carries with it a number of qualifying questions of interest. Can portfolios be used as a valid tool of assessment in place of high stakes tests? Would the process of creating a portfolio foster enough deep learning that students would perform better on high stakes testing than if teachers were to use methods more typically labeled “teaching to the test”? What are the benefits of portfolios upon students’ learning? Are there benefits in building portfolios that are not attached to students’ learning that make them worthwhile anyway?
Types of Portfolios
Though portfolios can be used in a variety of manners and fall under a number of definitions, there are three types of portfolios usually mentioned: assessment portfolios (also called documentation portfolios), learning portfolios (or process portfolios), and showcase portfolios (Barrett, 2004; Hewett, 2004). Since the showcase portfolio is a fairly straight-forward piece, conversations about portfolio implementations tend to sway between learning and assessment portfolios.
Barrett (2007) cites the definition of Zubizarreta to explain learning portfolios. These provide “a structure for students to reflect systematically over time on the learning process and to develop the aptitudes, skills and habits that come from critical reflection” (as cited on p. 438). These learning portfolios are considered a type of constructivist assessment, where the standard is particular to the one constructing the portfolio, and where, generally, the material created does not lend itself to measurement (Paulson & Paulson, 1994, p. 7). Here, the items in the portfolio are selected by the learners for their ability to display what they have learned. Ultimately, the goal of a learning portfolio is to make its creator a better learner – a learner who can produce better work in the future after having reexamined the current work and him or herself as a learner (Zellers & Mudrey, 2007; Hewett, 2004).
In the constructivist classroom, reflection is a key element of the learning process. As Bransford, Brown, and Cocking (1999) state,
Teaching practices congruent with a metacognitive approach to learning include those that focus on sense-making, self-assessment, and reflection on what worked and what needs improving. These practices have been shown to increase the degree to which students transfer their learning to new settings and events. (1)
The opportunities provided for reflection by the portfolio process can aid this goal, as portfolios have been shown to increase student metacognition (Zellers & Mudrey, 2007). For process portfolios, specifically intentional and separate reflection pieces are staples of the package. They are there to force students to review the learning process and the items included (Fernsten & Fernsten, 2005).
Stefanakis explains that the goal of an assessment portfolio is to “construct a comprehensive assessment system . . . to make visible what students do and what teachers teach” (as cited in Barrett, 2007, pp. 438-39). These portfolios might also be defined as a type of positivist assessment, where standards are set externally in advance and portfolios are crafted to measure themselves against those standards. In such an environment the organization of the portfolio and the expectations are typically clear and normed across the institution using them (Paulson & Paulson, 1994, p. 7).
In a sense, the assessment portfolio is the student’s attempt to reconstruct the ideal, whereas the learning portfolio is the students’ attempt to construct a picture of what has occurred. The biggest caution about the types, however, is that
more research is needed on examples of implementation that clearly differentiate between student-owned electronic portfolios and the assessment systems used to record evidence of students’ progress toward meeting standards. (Barrett, 2007, p. 439)
Concerns with Assessment Portfolios
Concerns exist about assessment portfolios. For one, Barrett (2004) expresses concern that the portfolio process often begins as an attempt to add a reflective and “deep learning” element to a student’s experience but is then lost beneath administrators’ needs for data to assess students’ learning (p. 2). For another, assessment portfolios tend to be designed with the institution’s needs in mind above the specific needs of the learner. The resulting portfolio then becomes a type of busy-work students are loathe to complete (Barrett, 2004).
The animosity to the assessment portfolio comes especially from the way assessment changes what the portfolio attempts to do. A primary need for assessment is reliability, which with portfolios is needed especially across evaluators. To attain that reliability, typically each element of a portfolio is evaluated separately and anonymously, which eliminates views of the holistic learning experience as portrayed by the learner. Maintaining the holistic element, however, produces “formidable challenges to reliability” (Paulson & Paulson, 1994, p. 9). Such challenges were publicly visible in Vermont when the RAND Corporation authored a study that called the results of the statewide portfolio assessment program unreliable. Among the issues the authors labeled as likely causes of the unreliability was the lack of standardization involved with portfolios. When students perform “the same tasks under the same conditions” (p. 24), rater reliability in particular becomes possible. But when the standardization is not present, it is difficult for raters to apply consistent criteria and interpretation (Koretz, McCaffrey, Klein, Bell, & Stecher, 1992).
In response, supporters of learning portfolios point out that the difficulty of using portfolios for assessment is part of their appeal. As Paulson and Paulson (1994) observe, they “are by nature complex and holistic pictures of a child’s learning” (p. 1).
Another View on the Confrontation
One possible perspective on using portfolios in the classroom is that the emphasis on assessment versus reflection is slightly misguided. It could be a situation where antithesis should not exist. The idea here is that the portfolio process that emerges out of a learner-centered classroom could prepare students for traditional assessment just as well as other methods of instruction. Struyven et al. (2006) hypothesized just this and though they were not able to show that alternate assessments and learner-centered classrooms were better than traditional lecture based or teacher-based classrooms at preparing students for traditional multiple choice tests, they were able to show that the learner-based methods and assessments drew similar or equal results. It is worth noting, however, that Struyven et al. (2006) cited studies in their literature survey that suggested a “deep learning” approach could prepare students for tests on equal levels with lecture-based instruction and then add the bonus of supplying them with a deeper learning than the lecture-based methods (p. 203). In their discussion they then explain the weaknesses of their own study and call for more research to re-test their hypothesis, to see if the learner and process based instruction could outstrip lecture methods. The possibility remains, then, that the portfolio need not be a high-stakes test substitution but could be the most proper preparation for the high-stakes test.
Implementation of Portfolios
When implementing portfolios, a number of considerations will help educators to utilize the tool well. For starters, going alone is a difficult process, and teachers who were alone in their pursuit of the portfolio method tended to struggle (Barrett, 2007, p. 446). For another, portfolios tend to flourish when the emphasis of instruction is on the learning and reflection rather than peripheral elements like the technology used to create the electronic portfolio. In all this, the teacher’s role is vital, as it is the teacher who communicates the goals of the portfolio’s development, instructs students in the creation of reflective writing, and guides the technological construction of the tool (Barrett, 2007).
One of the great difficulties of using portfolios is that they change the course of the classroom. The most significant and obvious change is the time spent on the portfolio process – time that used to be spent on something else. Zellers and Mudrey (2007) make this concept clear, pointing out that to use portfolios well, teachers will need to devote significant time for teaching the technology involved in hosting the portfolio as well as the reflective process that surrounds its creation. For a learning portfolio in particular to reach its goals, students must learn that reflective process early on (Zellers & Mudrey, 2007).
Paulson and Paulson (1994) point out that portfolios not only change the classroom environment, they change the learning being assessed, due to their emphasis on “integrative and reflective processes” (p. 9). It is not surprising, therefore, that some classes have emerged as better suited for the creation and use of portfolios (Zellers & Mudrey, 2007). Related to the idea that some classrooms are better fits for portfolios than others is the emphasis of many proponents of portfolios, like Fernsten and Fernsten (2005), that any classroom engaging in a reflective portfolio building process must build an atmosphere of trust. That trust can then engender honest and personally relevant reflection, as well as conversations about how students can become better learners.
This changed classroom is actually part of the goal of those seeking to implement portfolios – turning the classroom into more of a learner-centered environment rather than a teacher (or teaching)-centered one. In such a room, portfolios are used to provide students with “opportunities to become active learners as they set goals for learning, engage in self-reflections, review goals periodically and assume responsibility for their own learning” (Hewett, 2004, p. 27). The feel that one attempts to pass on to the learner in this context is that assessment is a process for the learner to engage in, not have imposed upon him (Hewett, 2004).
Supporting these goals further are ideas that have emerged about how to use and craft learning portfolios in a way that could motivate and appeal to students. Barrett (2004) suggests that the power of digital story-telling could involve students in an experience that could motivate them similar to the way games do. The story telling experience that emerges through that experience is often posited as a valid tool for expressing reflection and metacognition. In that sense, the portfolio becomes “a story of learning” (Barrett, 2004, p. 10).
That feel is likely part of why portfolios have become ubiquitous in teacher-training programs, as students claim to gain more knowledge in classes where portfolios are used (Hewett, 2004). As mentioned earlier, however, the positive support expressed by those involved in the creation and collection of portfolios has not been attached to empirical research. Though students like portfolios, little to no data exists to prove it is better than other methods of instruction or assessment (Struyven et al., 2006).
Research Questions Answered
Though many questions remain unaddressed in research, a few can be answered at least partially. Portfolios have been accepted as a strong tool for introducing reflective thinking into the classroom, and thus have been considered a driver of metacognitive learning. It has been shown, however, that when portfolios are used for assessment, much of the metacognitive and reflective dimensions are withdrawn from the process. This is because in that context portfolios are made standardized and are driven by the need for valid assessment data. Learning portfolios retain the metacognitive and reflective elements, but they do not produce valid results on large scale assessment, suggesting so far that they cannot adequately replace traditional testing formats.
The closely related question of whether portfolios can be used as a preparatory tool for high stakes tests has been explored more thoroughly by researchers than others questions posed, but no satisfactory answer has yet emerged. Struyven et al. (2006) have called for additional studies to answer this question, and as an educator who has built reflective practice into my classroom, I would echo that call.
Calls for Research
Researching whether building learning portfolios and engaging in the attached reflective process is a powerful preparation for testing is only a place to start in examining the effect of portfolios on students’ learning. As students give high marks to the use of the portfolio, studies should seek to find whether the effects on those students’ learning correspond with their favorable opinions. One might examine not only whether their learning improved within the course but how well that learning was retained long-term, compared to traditional lecture based or test-preparatory style classes.
The positive influence of reflection upon students learning seems to be an accepted fact in the literature surveyed here. It would be good practice to examine this claim itself to solidify the effect reflection has upon learning compared to other methods. Assuming that the practice of reflection would hold up to that examination, it would also be worthwhile for researchers to examine whether pieces called portfolios are truly the best methods of reflection, or whether other methods, like consistent upkeep of a learning journal, might compete with portfolios as a valid tool for conducting this practice.
As a teacher, I have long hoped to embed the building of a portfolio into my students’ experiences. Looking at the research has forced me to question the assumed outcomes of the process. Realizing that I already have students engage metacognitively through written reflections, I may be forced to admit this truth: the portfolio I would have them build appears to be little more than a well-organized keepsake for the future. For validity of creating that, however, I need no empirical research. Posterity and nostalgia are reasons enough.
Barrett, H. C. (2004). Digital Stories of Deep Learning. Electronic Portfolios. Retrieved April 15, 2008, from http://electronicportfolios.org/digistory/epstory.html.
Barrett, H. C. (2007). Researching electronic portfolios and learner engagement: The REFLECT initiative. Journal of Adolescent & Adult Literacy, 50(6), 436-449.
Bransford, J. D., Brown, A. L., & Cocking, R. R. (1999). How people learn: Brain, mind, experience, and school. . Washington, D.C.: National Academy. Retrieved May 2, 2008, from http://www.nap.edu/html/howpeople1/notice.html.
Fernsten, L., & Fernsten, J. (2005). Portfolio assessment and reflection: Enhancing learning through effective practice. Reflective Practice, 6(2), 303-309.
Hewett, S. M. (2004). Electronic Portfolios: Improving instructional practices. TechTrends, 48(5), 26-30.
Holtzman, D., & Dagavarian, D. A. (2007). The use of electronic portfolios in assessing student learning outcomes. The Journal of Continuing Higher Education, 55(1), 65-69.
Koretz, D., McCaffrey, D., Klein, S., Bell, R., & Stecher, B. (1992, December). The reliability of scores from the 1992 Vermont portfolio assessment. RAND Research. Retrieved May 1, 2008, from http://rand.org/pubs/drafts/DRU159/.
Paulson, F. L., & Paulson, P. R. (1994). Assessing portfolios using the constructivist paradigm. Annual Meeting of the American Educational Research Association. New Orleans.
Struyven, K., Dochy, F., Janssens, S., Schelfhout, W., & Gielen, S. (2006). The overall effects of end-of-course assessment on student performance: A comparison between multiple choice testing, peer assessment, case-based assessment and portfolio assessment. Studies in Educational Evaluation, 32, 202-222.
Zellers, M., & Mudrey, R. R. (2007). Electronic portfolios and metacognition: A phenomenological examination of the implementation of e-portfolios from the instructors’ perspective. International Journal of Instructional Media, 34(4), 419-430.