Principles of Effective Literacy Assessment
Cooper (1997) discusses eight principles of effective literacy assessment, based in part on the work of Farr and Tone (1994), Harp (1991), Valencia, (1990a, 1990b), and Valencia, Hiebert, and Afflerbach (1994):
"1. Assessment should be an ongoing process. Literacy assessment is not a test given at the end of a unit or a block of study, separate from the ongoing daily activities of instruction. Instead, assessment should take place every time a child reads, writes, speaks, listens, or views something. When assessment is viewed as an ongoing part of instruction, it becomes natural and expected.
2. Effective assessment is an integral part of instruction. The best forms of assessment are the routine daily activities of instruction, which tell us exactly how our students are performing. By comparing the work of individual students over time, we can determine patterns of growth. When a student writes a story about her trip to visit friends, you can assess her ability to organize ideas, express herself, and use the various conventions of language. Overall, you get a picture of how effectively she constructs meaning through writing. ...
3. Assessment must be authentic, reflecting 'real' reading and writing. For years, this author has asked teachers in workshops and classes, 'If you want to know how well children read and write, what do you need to have them do?' They have always replied in unison, 'Have them read and write.' Even in the heyday of using isolated skill assessment practices, teachers knew for years that marking, circling, and underlining did not reflect authentic reading and writing. The tasks of assessment in a literacy-centered classroom must reflect and honor the 'wholeness' of language (Harp, 1991). It is possible for learners to be very effective readers and writers and not do well on a test covering an isolated piece of the process. ...
4. Assessment should be a collaborative, reflective process. It should not be viewed as something the teacher does to the students. We know learning is a collaborative process; we learn alongside and with our students and our peers (Collins, Brown & Newman, 1986). If this is true for learning, it is also true for assessment. As students collaborate with their teacher on assessment, they reflect and ask themselves, 'How have I done?' 'What can I do to improve?' 'How can I use what I have learned?' Thus, students should help you assess and evaluate their own progress in literacy. ... Collaboration means students sometimes help select what they want evaluated. This becomes a joint effort in which teacher and students work and think together, and should also involve parents (Dillon, 1990). When students, teacher, and parents collaborate on evaluation, the responsibility is shared, as it should be.
5. Effective assessment is multidimensional. Quality assessment should use several different tasks, such as samples of writing, student retellings, records of independent reading, self-evaluations, and checklists. In making these choices, you need to trust your own intuition based on your knowledge and observations about students. More formal types of assessments have proclaimed their validity and reliability using various statistical procedures. Although many of the techniques being suggested today are more informal, we must still know that they are trustworthy (Valencia, 1990a), and one way to determine this is to use multiple tasks to get a consistent pattern of performance. Cambourne and Turbill (1990) argue that data generated from multiple sources using teacher observations and judgments are just as trustworthy and 'scientific' as those generated by what have been called 'measurement-based' approaches to assessment.
6. Assessment should be developmentally and culturally appropriate. We know children develop literacy and their ability to construct meaning by 'trying out' their reading and writing and making approximations. Therefore, tests or procedures that require absolute mastery at a given level or complete mastery of a given set of words before moving to a new book are completely contrary to how we know children learn. We must select assessment tasks that honor children's developmental levels of learning. At the same time, we must consider the cultural diversity of our classrooms. Children from different cultures have not only different language bases but also different patterns and styles of learning (Au, 1993; Garcia, [G. E.] 1994). We must take these into consideration as we plan our assessment procedures.
7. Effective assessment identifies students' strengths. Children learn to construct meaning by doing what they already know how to do and by getting support in gaining new strategies and techniques. This is using what Vygotsky (1978) calls the zone of proximal development. Effective assessment therefore must help us identify what our students do well. For many years, we have given students tests to find out what they do not know; then we proceeded to plan lessons totally around these weaknesses. This is contrary to how students acquire language and contrary to how they learn to construct meaning.
8. Assessment must be based on what we know about how students learn to read and write. This entire text has focused on how students learn to read and write and construct meaning. Clearly, we know assessment has not kept pace with our knowledge about reading and writing. We know the two processes are similar but different. We also know they develop together and produce benefits that are attainable by neither one alone (Tierney & Shanahan, 1991). And we know reading and writing are both constructive processes. As we plan assessment tasks, we must keep this knowledge in mind, incorporating new knowledge as it becomes available." (pp. 516-518)
Note: From Literacy: Helping Children Contruct Meaning (3rd ed., pp. 516-518), by J. D. Cooper, 1997, Boston: Houghton Mifflin Company. Copyright © 1997 by Houghton Mifflin Company. Used with permission.
Copyright © North Central Regional Educational
Laboratory. All rights reserved.
Disclaimer and copyright information.