Purpose/Types

An assessment of writing provides an indication of how well students communicate in the written language. Teachers can determine student progress in writing through direct measures of functional literacy, such as writing samples, process writing, and dialogue journals. Writing samples are the most commonly used performance assessment of writing ability and will be described further.

Design

To construct a writing sample, the following steps are recommended:

(1) Select writing prompts that are developmentally appropriate. Use prompts that will elicit writing from students on topics that are familiar and interesting to them and that reflect writing styles commonly practiced at their grade level. Writing prompts should not depend upon a student's knowledge of a content area.

(2) Provide a choice of topics. By giving options, teachers increase the possibility that students will be interested in and capable enough to select one of the topics and write on it (Hamp-Lyons, 1990; Reid, 1990). However, because the scoring of writing samples may differ depending on the genre, the topics provided in a single assessment should be limited to one genre, preferably one to which students have already been exposed. Representative topics may differ by grade level and can include autobiographies, biographies, short stories, reports, and critical analyses.

(3) Check the topics for cultural bias. If the topics consistently assess knowledge of a particular culture rather than of the English language, they are probably culturally biased. Writing prompts may address relatively neutral themes, such as asking students to describe a favorite relative. Alternatively, they may include more controversial topics designed to provoke students to take a position, such as students' attitudes toward violence in school. Topics can be checked for bias with cultural informants (other school staff or parents) who share or have experience with the culture of the students.

Administration

Give clear directions. Students should know the amount of writing required, how much time they have to write, and how their writing will be scored. A minimum of thirty minutes should be provided to allow students to plan, organize, and revise their work. A minimum of two paragraphs should be required of students. Decide whether students can use dictionaries or other resource materials during the writing process and apply this rule consistently.

Scoring

In preparation for a writing assessment, students should be told what the scoring criteria are, be given model papers that illustrate the range of scores assigned to writing samples, and be provided opportunities through cooperative learning groups to discuss their work-in-progress relative to the scoring criteria (Kolls, 1992). Students should be given opportunities to re-write their products after receiving feedback from scoring.

Scoring writing samples of students in various grade levels means developing criteria to apply to different levels and determining what scores mean with regard to student progress. At a minimum, at least three categories should be established within each grade and level of English language proficiency with regard to writing ability: low, intermediate, and high levels of writing. For example, a student might be a high level writer at an intermediate level of proficiency for sixth-graders.

To score a writing sample, use holistic scoring criteria which focus on the communicative nature of the writing (see Figure 3). Other options include assigning relative weights to overall fluency, organization, mechanics, and grammar. Depending on the goals of instruction, teachers may want to assign lower weights to language conventions (e.g., mechanics and spelling) and higher weights to expression (coherence, effective narrative strategies) in order to encourage students to communicate freely through writing.

The most reliable way to score writing samples is to allow at least one other teacher or staff member to score the sample separately, after both raters have discussed and agreed upon the scoring criteria and have obtained similar scores on a few writing samples. Raters need to establish consistently high interrater agreement levels with regard to the appropriate classification of each student but not necessarily with regard to exact scores. A minimum interrater agreement level of 80% is recommended. This means that two teachers rating a sample of student papers should agree on the overall rating for at least 80% of the students (see Figure 3).

Significant differences in the classification of students require negotiation, clarification of the scoring criteria, and, perhaps, re-scoring the writing sample. If differences persist, then raters should look at other evidence of student writing in order to determine progress and decide whether a student should be placed at a higher or lower level of instruction. The usefulness of the writing sample depends upon the quality of the scoring procedure used and the degree of interrater agreement established.

ANECDOTAL RECORDS

Purpose

Anecdotal records are notes based on teacher observations of how students learn. They can be used to determine a student's integration of reading and writing processes. The open-ended nature of the anecdotal record allows teachers to describe in rich detail each student's development in literacy, to integrate these observations with other available information, and to identify instructional approaches that may be appropriate (Rhodes & Nathenson-Mejia, 1992).

Design

Anecdotal records can be produced by following three guidelines:

(1) describe a specific event, process, or product;

(2) report rather than evaluate-save interpretation for later; and

(3) relate the material to other known facts about the student.

Administration

Students can be observed while engaged in learning activities which produce the behaviors to be assessed. In completing the anecdotal record, the teacher will describe the specific learning event and what the student actually did in that situation. For example, teachers may develop anecdotal records of academic language skills by observing students as they work on math word problems in cooperative learning activities. Students may also be observed and asked appropriate questions while reading aloud in order to determine their level of reading comprehension on different types of materials. In addition to observations, teachers can review student products as a means of developing anecdotal records. For example, teachers can keep anecdotal records on student writing samples to document progress in literacy.