By MELISSA STEINEGER

he third-graders at Stevenson Elementary School in Washington dutifully read Baby Wolf, a text with all the style and substance of the old Dick-and-Jane primers: See baby wolf. See baby wolf sleep. Yawn, baby wolf, yawn.

It was the young readers who were yawning when Dr. Lesley Thompson jolted them with a story about a sled dog named Cookie and her litter of pups. Thompson, a reading-assessment expert at the Northwest Laboratory, had given the youngsters a chapter titled "Nativity" from Gary Paulsen’s book Puppies, Dogs, and Blue Northers: Reflections on Being Raised by a Pack of Sled Dogs. The text, written at about a sixth-grade level—three grade levels above their other materials—dramatically describes how Cookie dealt with the death of one of her puppies.

As they responded to Thompson’s prompts and queries, the students showed their ability to synthesize what they had read—that is, to see how parts of the story work together to build meaning, noting similarities and differences between the two stories, and finding meaning beyond the texts’ literal boundaries. To probing questions such as "How do you know that Cookie was a good mother?" the third-graders eagerly shared their feelings, all the while exploring ideas from the text. Even later, when the official inquiry into the text was over, the students returned to the story during free periods, reread it on their own, and talked about it with their classmates. In explaining the difference between the stories—why Paulsen’s narrative captured the young readers so completely—one student replied: "‘Nativity’ is real to me. Baby Wolf is just a story."

These students, who were part of a 1997 pilot project on reading assessment, had become what Thompson calls "engaged readers." Engagement with text—something that research indicates is critical to becoming a good reader—can be fostered in the primary grades by using informal reading assessments, Thompson and other researchers say.

Simply put, informal reading assessments involve understanding an individual reader—his or her strengths, weaknesses, interests, attitudes toward reading—and then using that understanding to help engage the student in reading. The goal is to create readers who challenge themselves to read frequently and who tackle books above their grade level. Eager readers, like the third-graders Thompson worked with, are the end result of effective classroom assessments.

The strength of informal assessments in the overall assessment mix is their ability to give specific clues to a young reader’s personal set of strategies for decoding words and gleaning meaning from text, says NWREL researcher and reading specialist Dr. Jane Braunger. These clues give the teacher the insight he or she needs to plan instruction that will move that student further along the reading continuum.

"You want a window to the child’s mind," says Braunger. Opening that window, she says, is what informal assessments are all about.

Braunger, coauthor with Dr. Jan Lewis of the 1997 book Building a Knowledge Base in Reading, jointly published by NWREL, the National Council of Teachers of English, and the International Reading Association, identifies the key benefits of informal assessments:

They provide indepth understanding of an individual’s development as a reader. This allows a teacher to design instruction to meet the needs and interests of all the individuals in a class, create flexible reading groups that change as skills change, choose reading materials appropriate for the needs of the individuals, and move the individual along his own continuum of reading development.

They help a teacher see a youngster’s background knowledge, motivation to read, and other hard-to-measure elements that have a profound effect on how well a youngster reads.

They help a teacher link teaching and testing by allowing her to build on students’ base of knowledge and skills in designing instruction.

Noted researcher Dr. P. David Pearson of Michigan State University advocates using informal testing, especially in the younger grades. Pearson, author of numerous books on reading, says teachers would use their time better by informally assessing individuals rather than by giving standardized tests to whole classes, particularly in the primary grades where students may not understand directions or may be unable to concentrate for the duration of the test. Teachers will gain the clearest picture of a youngster’s reading development, he stresses, by using a variety of assessments rather than relying on any single type of evaluation. (For more of Pearson’s views on informal assessments, visit his Web site at http://ed-web3.educ.msu.edu/ cspds/home.htm. Select Works in Progress and then Past Archives to find his essay, "Standards and Assessments: Tools for the Reform of Early Reading Instruction.")

Following are descriptions of three tools—running record, miscue analysis, and cloze procedure —and one task—oral retelling —that teachers most often use to informally assess students’ reading strategies and comprehension assessment. Assessment tools, says Thompson, are, like hammers and wrenches, instruments that are applied in the same way over and over. A task, in contrast, is variable and adaptable to specific needs or settings. Thompson recommends that teachers receive training in the use of these techniques in order to fully realize the benefits they can bring.

Running record
a cumulative account of selected behavior, as of that of a student noted by a teacher over time.*

The running record of text is a system developed by New Zealand educator Marie Clay and detailed in the 1993 edition of her book, An Observation Survey of Early Literacy Achievement. A running record is the teacher’s record of everything the child says when reading a text aloud.

Reading samples are 100 to 200 words, or enough to take the child three to 10 minutes to read. Samples should come from readily available reading materials used within the regular classroom and should be at three levels of difficulty: easy, instructional, and hard.

Easy texts are those children have previously read successfully. Instructional texts have some familiarity for the child, but the reader must problem-solve to read at 90 percent to 94 percent accuracy.

A harder text may have been previously introduced or never seen before, and the child will read it at 80 to 89 percent accuracy. These levels of difficulty provide insights into how a young reader orchestrates effective reading (easy text), how a youngster problem-solves reading material (instructional text), and how effective processing breaks down (harder text).

There are several options for recording the information: using a blank sheet of paper, creating a form, or making a duplicate of the text with space between lines. Although running records don’t need to be tape recorded, recording may help teachers who are learning the technique to review how well they’ve captured errors.

Clay offers a shorthand method of capturing the child’s reading. It goes like this:

For every correctly read word, the teacher makes a checkmark.

If the reader misreads a word, the teacher writes the wrong word over the correct word, like this:

(Child) Spit

(Text) Spot

This is counted as one error.

If the youngster tries several times to read a word, each attempt is recorded:

(Child) Spit / Splat / Spat

(Text) Spot

This is counted as one error.

If the teacher must tell the child the word, the teacher records a "T" for told:

(Child) Home ___

(Text) House/ T

This counts as one error.

If the youngster makes several attempts, eventually getting the word correct, the teacher writes SC for "self-corrected." SC is not an error.

If a child reads a word that is not in the text, the teacher writes the word and beneath it draws a dash. If the child skips a word, the teacher writes the word and above it draws a dash. Either instance counts as one error.

If a youngster appeals for help, the teacher writes an "A" for appeal. If the teacher must tell the child the word, the teacher adds a "T" for told. Either situation alone or combined counts as one error.

If a child gets into a state of confusion, the teacher may intervene by saying, "Try that again," perhaps indicating where the child should start anew. Such an instance is marked as TTA (try that again). If the confusion lasts over several words or a phrase, the teacher brackets the entire portion and counts it as one error.

Repetition is marked as an R, but is not counted as an error. If the child repeats a phrase, the teacher writes an R and draws an arrow back to where the youngster begins anew. If the child reads the phrase correctly, the entire sequence is marked as SC and is not an error. Sometimes a child will reread text and correct some but not all errors. Any repetitions are not counted as errors, but new or continued misreadings are.

No error is counted for trials |that are eventually successful—for instance, the child who says want/ won’t/went for went.

Other non-errors include reading multiple errors and then going back and self-correcting, broken words (a way for away), and mispronunciation. If there are alternate ways to score, score so that the fewest errors are recorded.

To determine the student’s error rate, compare the number of errors with the number of words in the text. A student making 15 errors in a 150-word reading selection has an error rate of one to 10 (one error to 10 words).

To calculate the percentage of accuracy, divide the number of errors by the total number of words in the text. Then subtract that number (the error rate) from 1. In the example above, the student has a 10 percent error rate, hence an accuracy rate of 90 percent. Children who are reading text at the appropriate level should score at 90 percent or higher.

"If there is more than 10 percent of error in the record rate this is a ‘hard’ text for this child," Clay writes. "When children read a book with less than 90 percent accuracy, it is difficult for them to judge for themselves whether their attempts at a word are good ones or poor ones. They need easier material which they can attempt at a rate of not more than one error in 10 words at the time they begin the new book. For the average child there is movement from 90 percent accuracy when he is first promoted to a book to 95 percent or more as he completes his learning on that book."

A long analysis is not necessary. But a teacher should at least ask what led the child to make the error and use that information as an aid in instruction. More difficult texts will produce higher error rates; with more practice, teachers improve their ability to detect errors—also raising the error rate.

While learning to take a running record may seem complicated, Clay says most teachers need two hours or less to learn the basics. She suggests selecting three average readers as case studies. Try out the procedures on these children, score and analyze the results, and summarize the observations. Some teachers fit running records into their schedules by taking a running record of one child each day—which means each child has a running record every three to four weeks.

Miscue analysis
a formal examination of the use of miscues as the basis for determining the strengths and weaknesses in the background experiences and language skills of students as they read.

A reading miscue inventory and analysis combines aspects of the running record with oral retelling. The inventory is a listing of words a child may substitute when faced with a word that is difficult to pronounce or decode. By analyzing a child’s substitutions as the child reads aloud, the teacher can look for specific areas where a reader is having trouble decoding words, phrases, or ideas and later provide the appropriate instruction.

A "miscue" is defined in The Literacy Dictionary of the International Reading Association as "a deviation from text during oral reading or a shift in comprehension of a passage." The dictionary adds the important note that "miscues are not random errors, but are attempts by the reader to make sense of the text." For this reason, they "provide a rich source of information for analyzing language and reading development."

The general approach is for the teacher to choose a text of interest to the reader, but one the reader has not read. The piece should form a cohesive whole, whether it is a story, poem, or other text of about 500 words, or enough text to allow for 15 to 30 minutes of reading. Several pieces may be chosen to reach that duration. The piece should be predictable—that is, the reader should be able to provide some meaning without a previous reading—yet be of sufficient difficulty that the reader will make miscues. (If the reader does not miscue, the student has mastered the reading-skill level of the chosen text.)

The teacher marks identical copy that is triple-spaced, but has the same number of words per line and page so that format miscues can be identified. After the child reads the text and the teacher marks the miscues, the reader retells the story orally, without prompting, if possible. When the retelling is finished, the teacher delves more deeply into the story with open-ended questions such as these: "Tell more about (a character named by the student);" "After (an incident mentioned by the student) happened, what came next?" and "Why do you think (a character named by the student) did that?"

Analyzing the miscue inventory and the retelling can provide insight into a child’s ability to use context to establish meaning, drive to seek meaning, ability to self-correct, growth in developing fluency, and ability to read with fluency and expression.

Cloze procedure
any of several ways of measuring a person’s ability to restore omitted portions of a text by reading its remaining context.

The cloze process can help a teacher assess a student’s reading comprehension and mastery of language. Students use clues from the context of the passage to fill in words that have been deliberately omitted. A traditional cloze exercise omits words systematically, say every fifth or 10th word, regardless of the word. But words also can be deleted more selectively, by category. Teachers may wish, for example, to omit verbs, prepositions, or special vocabulary. Research suggests that selective word deletion is more useful in providing clues to teachers than systematic deletions.

LinguaLinks (http://www.ethnologue.com/LL_docs/contents.asp), an electronic resource for language learners and teachers, recommends leaving intact the first two or three sentences of the passage to provide enough context clues for the reader. Here is what a cloze exercise with selected verbs omitted might look like with a passage from "Nativity."

Cookie slept hard, was absolutely sound asleep, and I thought I would take the body now, take it to the house and dispose of it so she could not find it. But when I reached across the hut to get it, her eyes opened and her lips moved to clear teeth, and again she looked directly into my eyes. "I will ____ your sled," she said, "and love you and ____ the team and ____ your life and be loyal to all that you are and ____ you in all things until I cannot, but if you ____ my pup you die." I left the pup and it was not for three days, almost four, when the still-frozen pup was clearly not going to come back to life, that she finally surrendered to her grief and let me take it away.

LinguaLinks offers these guidelines for preparing texts for a cloze procedure:

Choose texts that provide a lot of clues and supporting information to aid word identification

Use cloze tests along with other kinds of tests for reading comprehension

Test the text with a fluent reader before using it

Use texts at an appropriate reading level for each learner

In reviewing students’ completed cloze exercises, Thompson says, teachers should look for complexity and sophistication in the student’s word selection, the student’s ability to come as close as possible to the original text, the ability to demonstrate literal comprehension of the original passage, and a willingness and desire to create meaning.

Retelling
requires the student to construct a personal text and make inferences both from the original text and prior knowledge.

Oral retelling measures how well a child understands a text and can help youngsters improve their reading comprehension. This assessment also helps engage children in reading by allowing them to capture the flavor of a piece they’ve read by using their own language.

"An analysis of the retelling can help teachers identify problems that do not surface when students are simply asked to answer questions," the Oregon Department of Education notes.

Oral retellings are straightforward. Aware that he or she will be retelling the story, the youngster reads a designated text. To help the student along, the teacher may preface the reading with some instruction that will help the child construct a good retelling. She might, for instance, instruct the reader to give a general introduction to the story, describe the main characters, and explain any obstacle the characters must overcome.

Typically, the teacher asks the child to tell the story as if he were telling it to another child who has not read the story. Older readers may retell the story in writing. For any age, teachers may choose to prompt the retelling with questions about the plot, characters, and significant ideas.

To evaluate, the teacher looks for the child’s knowledge of the gist of the story and the main ideas; accurate reporting of the events; sequence accuracy; the number and accuracy of direct quotes from the text; the ability to relate information to personal knowledge; the presence of the beginning, middle, and end of the story in the child’s retelling; precision of vocabulary; the presence of characters and setting; and the use of detail. This may be recorded on an appropriate scale—say, "low," "moderate," and "high."

Teachers also can probe for understanding by asking the reader to make inferences rather than simply recalling the text. They can invite a personal response to the text which helps young readers extend their ability to connect the text with other texts and experiences.

The teacher can model how to explore a text by making notes of the main ideas of the text, then sharing these with readers. He can invite the readers to identify what the text made them think about, and encourage them to ask questions about things they didn’t understand.

Oral retellings also can engage an entire class. Braunger once asked a first-grade class to retell a story, allowing students to relate their favorite passages or events while she wrote down the responses. In an article titled "Retelling: Reading Assessment That’s Also Good Instruction" from the Handbook for Student Performance and Assessment, a 1996 publication from the Association for Supervision and Curriculum Development, Braunger describes how the students were eagerly raising their hands or calling out their favorite parts. When the retelling was over, the entire story had been captured from the classroom full of engaged readers, she says. The experience also provided a model for the students when they later gave individual retellings of other texts.

Research has found that both low- and high-skilled readers have increased comprehension of setting, theme, plot, and resolution after participating in only four retelling sessions. n

*NOTE: Definitions of assessment terminology are from The Literacy Dictionary: The Vocabulary of Reading and Writing, edited by Theodore Harris and Richard Hodges and published in 1995 by the International Reading Association; and from Reading Assessment: Grades K-4, Third Grade Benchmark, published by the Oregon Department of Education.

What are the characteristics of a good reader—that is, one who reads critically, deeply, and with comprehension?

Having identified six traits shared by good readers, Dr. Lesley Thompson of the Northwest Laboratory, has developed an assessment strategy for the traits. Every reader falls somewhere on a continuum of progress for each trait, Thompson says. Students may have greater or lesser skill, but they can always be helped when a teacher informally assesses the skill level the student has reached and shapes instruction to further the youngster’s development.

The traits, along with examples of advanced accomplishment, follow:

1. Decoding conventions of writing, organization, and genre. Advanced students are able to recognize correct grammatical constructions, understand the function of punctuation, and have an awareness of spelling conventions. They can identify the title, author, and components of the text, such as the table of contents and chapter headings. And they can identify the genre.
2. Establishing comprehension. Students can state or write a thesis statement; name major and minor examples of the thesis; identify the turning moments with facts and examples; and can connect the turning points to the main thesis.
3. Realizing context. Students can use examples from the text to discuss the author’s intentions and inferred meanings, both implicit and explicit.
4. Developing interpretation. Students can identify problems in texts and resolve them using clues and evidence.
5. Integrating for synthesis. Students can connect text with other texts, subjects, and experiences.
6. Critiquing for evaluation. Students can, with insight and evidence, critique ideas and perspectives found in the reading.

Also available is a newly published video, which presents the six traits of an effective reader and shows how they can be applied in the classroom. For details and ordering information on The Journey of a Reader, call (503) 275-9535.

New Policy Guidelines for Reading: Connecting Research and Practice
by Jerome Harste (National Council of Teachers of English and ERIC Clearinghouse on Reading and Communication Skills, 1989). Intended to help educators develop improved policy in reading instruction and research, this book offers a useful discussion of the role of evaluation in reading curriculum.

This document's URL is:

Home | Up & Coming | Programs & Projects: Northwest Education | People | Products & Publications | Topics

© 2001 Northwest Regional Educational Laboratory

Date of Last Update: 9/28/01
Email Webmaster
Tel. 503.275.9500