Printer Friendly

Strategy usage among deaf and hearing readers.

Strategy Usage Among Deaf and Hearing Readers

Fluent readers use metacognitive strategies while reading. These strategies are often referred to as comprehension monitoring strategies (Baker & Brown, 1984). Metacognition, in reading, refers to the awareness and control that people have over their own reading comprehension. For example, experienced readers will know when they are not understanding parts of a passage and will often use contextual information or go back and reread a previous section to retrieve the meaning of the text. Fluent readers may use many such metacognitive strategies to assist in comprehending a text.

There is some evidence that deaf readers use many of the same strategies to comprehend printed material as do hearing readers (Ewoldt, 1977; Soderbergh, 1985). Case studies of deaf children involved in the reading process show that they use contextual information to "reconstruct" the text using their semantic knowledge. In general, however, reading comprehension is extremely difficult for most deaf children and youths.

Why is reading so difficult for deaf students? Several reasons have been given. One is that deaf readers lack background knowledge or prior experience about many topics found in commercial texts. Deaf children often do no have a communication system until after the age of 2 or 3 when deafness is diagnosed. Thus, many deaf children--especially those with hearing parents who cannot communicate with them very well--are deprived of having someone explain their environment to them. And as they get older, their gaps in world knowledge and conceptual understanding widen.

A second reason frequently cited in the literature is the poor English linguistic skills of deaf readers. Few ever achieve fluency in reading or writing English despite years of schooling (Moores, 1987). More specifically, they have smaller reading vocabularies, know few multiple meanings of words and idiomatic expressions, have difficulty understanding figurative language and syntactical forms, and have difficulty making inferences of "reading between the lines" (Quigley & Paul, 1984).

A third reason for deaf students' reading problems is that the form of sign language, typically the language of most deaf children, is structurally different from English. Thus, deaf readers must recode print into their own language. Hearing readers have the advantage of matching their speech to print (Conrad, 1979). Although some deaf readers are able to use sound recoding for reading and memory (Conrad, 1979; Lichtenstein, 1984), most deaf readers instead use kinesthetic codes such as fingerspelling or signs, or they use the visual representation of the letters or even a combination of these codes (Lichtenstein, 1984; Treiman & Hirsch-Pasek, 1983). Research has yet to determine how effective these alternative codes are to reading comprehension for deaf youths.

Instructional techniques in metacognitive comprehension strategies may be needed to improve deaf youths' poor performance in reading comprehension (Wilbur, 1987). Yet it is difficult to offer instruction without understanding the comprehension strategies deaf youths use when reading expository texts. Our approach is to compare deaf readers with two comparison groups of readers. Hearing youths who are at the same reading level and age offer one comparison, and younger children who are making normal progress offer another.

The purpose of this study is to examine closely the strategy usage among a small sample of deaf and hearing youths. The students were asked to read three expository passages in which key words were missing. The task was for the youths to figure out an appropriate replacement for the deleted parts and then explain how they selected the words. This technique is a type of protocol analysis (Olshavsky, 1976-77) using a cloze procedure. Language and reading research with deaf subjects has relied on a cioze procedure (LaSasso, 1978; Marshall, 1970; Moores, 1967). However, because the cloze procedure by itself would require experimenter interpretation of strategy use, we modified this procedure by asking the students to explain how they chose the words. Their self-reports were in the form of sign language for the deaf youths and verbal reports for the hearing youths (Ericsson & Simon, 1980; Garner, 1982).



Fifteen white males who resided in the southeastern United States made up the sample. The five males who were deaf were enrolled in the high school department in a state residential school for the deaf and ranged in age from 17 to 20 years. They were selected based on these factors: (a) profound hearing loss of 90 dB or greater (American National Standards Institute, 1969), (b) prelingual deafness, (c) born of hearing parents, (d) normal intelligence as measured by the Wechsler Intelligence Scale for Children-Revised with no additional disabilities that would impede learning, and (e) one subject each at these reading grade levels: second, third, fourth, fifth, and sixth, as measured by the Standford Achievement Test--Hearing Impaired Edition. None of the subjects could discriminate speech well enough to be measured. All were skilled users of American Sign Language (ASL), as reported by their school principal, who interacted with them daily. The average number of years attending the state school was 13.7 years for this group. Table 1 summarizes this background information.

Two groups of hearing youths were used as comparison groups. One group was composed of five white males from a public elementary school. This group was reading on grade level as measured by the California Test of Basic Skills. Their ages ranged from 8 to 11 years, and they were selected by their teachers as being average readers for the second, third, fourth, fifth, and sixth grade.

To obtain a hearing group of comparable age, five hearing white males in classes for students with learning disabilities were selected by the principal from a public high school. All subjects had normal intelligence but had at least a 4-year delay in acquiring reading skills, as measured by the Brigance Reading and Comprehension Test. Their ages ranged from 14 to 18 years, and their reading levels ranged from the second through the sixth grade.

Although there is no straightforward way to match the reading and language abilities of deaf and hearing youths, an approximation of reading from first through sixth grade provided a rough means of equating students and a way to describe the strategies used by students of various reading proficiencies.

All 15 subjects were from low to low-middle socioeconomic-status families. The family breadwinners worked on farms, in coal mines, with the railroad, or in small businesses. A few were unemployed. None had college degrees.


The beginning 10 to 14 sentences of seven expository texts (Houghton Mifflin, 1981) were selected. The topics included libraries, grasshoppers, corn, swamps, sea mammals, and horses. The seven passages were designed by Houghton-Mifflin to represent reading grade levels 1 through 7. Evaluation by the Fry readability formula showed there was some variation, though all the texts were at or just below the designated reading grade level. Linguistic characteristics of the texts, such as vocabulary and sentence structure, were considered when selecting the passages but were not formally assessed.

For all seven passages, the title and one picture from the text illustrating the topic were placed on the front page. For example, the first-grade text about the library showed a drawing of a room in a library with the title. One or two sentences of the passage followed on each page but with a word or phrase, usually the last word of the last sentence on a page, blanked out for the student to figure out. For instance, in the "Library" passage, one sentence read, "You can find good [underscore]." Here, the word "books" was deleted. Successive pages included the previous sentences(s) from the passage with the blank filled in. Page 1 of the "Library" passage, for example, had the tile of the passage and one sentence with the last word missing. Page 2 had a new sentence plus the previous one, "You can find good books." and so on. The student, then, could refer to all previously read information each time he read the new sentence and use that to help fill in new deleted words and phrases. Each text was organized in the same way. Each student read three texts: first, the text below his reading level, then the one on his reading level, and finally the one just above his reading level.


Before each session, the experimenter spent time making sure the subjects were comfortable by talking about their interests, hobbies, and school life. Next, a demonstration story was given to the students to ensure that they understood the directions and were comfortable with the audiotaping and videotaping. The experimenter gave the student the passage and then discussed the topic of the picture. The hearing students used voice communication; the deaf students used manual communication, either alone or with voice. The experimenter encouraged the students to describe and talk about the pictured topic. Then students read the title and the first sentence to the point where the text was omitted and they were told any words they could not read. They predicted what the missing word or phrase might be and discussed their rationale for making the prediction. If the students became silent after reading a sentence stem before filling it in, the experimenter would say, "What are you doing now?" or "What are you thinking?" or "How are you finding the answer?" These probing questions were asked using voice with the hearing students and in ASL with the deaf students. Probing questions were used to increase the explicitness of the report (Afterback & Johnston, 1984).

Following this discussion, the experimeter showed the students which word was blocked off by turning to the next page (where it was filled in) and had subjects continue reading. The deaf students were audiotaped and the hearing students were audiotaped. The experimenter was fluent in American Sign Language. Each session took approximately 1 hour per subject as they read and discussed the deleted words for the three texts--one below, at grade, and above grade reading level. All reading sessions occurred at the student's schools.


The audiotapes were transcribed by graduate students who were trained and monitored by the experimeter. The videotapes were transcribed from a slow-motion playback videotape by a mother of a deaf teenage boy. She was skilled in ASL and manually coded English. A second transcriber, a deaf teacher, interpreted the unclear signs.

Classification Scheme for Word Deletions. There were 29 or 30 word replacements attempted by each subject (9 or 10 for each passage). These responses were scored from strict to lenient: text-exact match, semantic associate, contextual, and unrelated or no response. The strict scoring was the exact response, what the author of the passage intended. So as not to penalize the deaf students, the response did not need to follow exact English syntax (ASL does not use the same order). A less stringent score is a close semantic associate such as a synonym ("hop" for "jump"), a superordinate ("flowers" for "dandelion"), or a cohyponym ("July" for "June"). A lenient score is a response that fits the context of the passage in general but does not fit the sentence frame. Unrelated and no response were coded as incorrect. See Table 2 for examples of how the responses were coded.

Classification Schemes for Strategies. From the written transcriptions, the experimenter studied the verbal reports of the hearing students and the signed reports of the deaf students, in which each student explained to the experimenter how he was filling in the deleted words. These statements yielded six strategies: the subjects' background knowledge of the text topic, rereading the sentence containing the blank, looking back to previous sentences, looking ahead to words beyond the blank, identifying contextual cues in the sentence, and identifying cues from the title. Table 3 shows strategy types, along with a definition and an example for each.

Interrater Agreement. Coded word deletion responses were rated by an independent observer, who coded the responses after studying a list of the levels with definitions and examples with copies of the written transcriptions. Interrater agreement of 94% was calculated using this formula: agrements/agreements + disagreements X 100. The types and frequencies of the strategies from the students' explanation were rated by an indenpendent observer using the same procedure. An 83% interrater agreement was found.

Word Replacements

Table 4 shows the number and percentage of all responses by the three groups. The exact-replacement scoring is too stringent for analysis since only about 26% of the responses would then be correct. Similarily, the contextual coding is too lenient since if exact and semantic contextual are allowed to be correct, then 83% of responses would be correct. Thus, we chose to accept an exact and semantic response, so that 48% of the responses were scored as correct. Coded responses of the deaf high school group were different from the other two groups, particularly because the deaf subjects were less likely to give a text-exact-match response. The hearing elementary students obtained 61% correct responses; high school students with learning disabilities, 50%; and deaf students, 32%.

Strategy Types

Examination of strategy type among the three groups revealed that the deaf group used 3.8 different strategies, whereas the others used 4.7 different strategies. All three groups used background knowledge, most frequently followed by rereading the sentence and looking back into the text. The use of context clues (the surrounding words) in the sentence and the title were also used by the hearing elementary and hearing learning disabled students. Looking ahead was seldom used by any group. The less able deaf readers were more likely to simply reread and use their background knowledge. The other two groups and the more able deaf readers made use of those two strategies, as well as title and context clues. Thus, the hearing students appeared to be more able to find text information.

Figure 1 shows that the number of strategies increased with the reading grade level of all subjects. That is, the more skilled readers had a larger number of strategies. Surprisingly, the deaf reader at sixth grade used a larger number of strategies than did any of the hearing elementary students. Conceivably, to be an effective reader, deaf students need to learn more strategies than do hearing students.

That the more skilled elementary group made less use of the background knowledge strategy and more of text-connected strategies might be explained in part by the higher rate of exactmatch responses. The greater use of context and title clues could be associated with a deeper or more accurate understanding of the text. At and above the third-grade level, both groups of hearing subjects made regular use of context clues, while this strategy appeared only with the two deaf subjects reading at and above the fifth-grade reading level.

Word Analysis Strategies

The focus of this study was for students to explain metacognitive strategies for reading expository texts. The approach used was to have the subjects tell us how they were filling in the deleted parts of the texts. While the deaf subjects were reading, however, we made some interesting observations concerning how they analyzed words to get at the meaning. Before the study began, we had decided to supply the students with words they could not read in the text if they asked us. For example, if they could not read a word before the deleted protion we simply told them the word in sign language. During this process, we observed the deaf readers using a variety of word analysis strategies as they attempted to read words they did not know.

The most skilled deaf reader used sound recoding to a great extent. He would sign the text and use voice. But before filling in the deleted part, he would mouth silently all the words while he was reading. He also frequently used a sign recording strategy. For example, when he came to the word "whatever," he signed DOESN'T MATTER, which is the ALS equivalent to this word. (Note that here and in the following text, ASL signed words appear in all capital letters, and fignerspelled words appear in capitals with hyphens.) The most common strategy among all the deaf subjects was fingerspelling. Typically, they would fingerspell words they did not know. Sometimes, they would first fingerspell the word, then follow it up with the sign equivalent. For instance, "teeth" was fingerspelled T-E-E-T-H, then signed TEETH. Graphemic strategies were also used by all five deaf readers. However, this word-attack strategy more often resulted in an incorrect response. The list a few instances, the word "there" was signed THE, "father" was signed FATHER, and "that" was signed THANK YOU. After the more skilled readers made these predictions, based on graphemic similarity of words, the students would usually go back and recorrect themselves. Thus, the more skilled deaf readers showed an ability to reconstruct the meaning of the sentence based on their semantic knowledge.

Combinations of word analysis strategies were most apparent when the reader attempted compound words or multimorphemic words. For instance, the word "wasteland" was read with two signs WASTE + LAND. A frequent combination was using both signs and fingerspelling; for example, "nearby" became the sign NEAR with the fingerspelling B-Y; "farming" was FARM + I-N-G, "useful" was USE + F-U-L, "sounds" was SOUND + S.


The data show that the hearing readers were more able to come close to actual word or phrase replacements than were the deaf readers. Given the linguistic advantage of the hearing students, this was expected. To figure out the information, all readers used strategies of background knowledge, rereading the text, looking back at preceding print, looking ahead at print following the sentence they were reading, and using context clues. The hearing group also reported using the title as information to fill in the deleted word or words. However, the deaf readers used context clues infrequently and never used the title of the passage. The strategies that seem to predict correct word replacements for the deaf readers were the background knowledge and rereading strategies. Yet, when these strategies were compared with more text-dependent strategies such as context clues and the use of the title, it was found that they were not as efficient in predicting the correct answer.

Returning to the reasons given earlier for deaf readers' comprehension difficulties, we found examples of all these problem areas while observing the five deaf youths reading the expository texts. For instance, the sea mammal passage was difficult for the two skilled deaf readers because they had little background knowledge about this topic. Linguistic difficulty was apparent at the word, sentence, and intra-sentence levels. For example, the multiple meaning of "pups" (meaning baby seals as well as baby dogs) had to be explained to one reader. The metaphor "dandelions are like little suns" created a comprehension obstacle for another reader. Sentences that required that understanding of the comparison of two or more ideas were problematic. Recoding the print into silent voice or fingerspelling or bassing a word meaning on a graphic similarity to another word often did not lead to a correct understanding of the word. It is clear from this study that deaf youths need more experience reading through texts with a skilled teacher who can help them identify and use appropriate strategies and then use them on their own.

An analysis of text-comprehension strategies offers one way of describing the processes used by deaf students while reading. It demonstrates that deaf youths can use their own language of signs and fingerspelling with their prior experiences or background knowledge to construct meaning from print. This analysis also shows that these existing strategies need to be further developed while other more efficient strategies need to be explicitly taught.

Another significant aspect of the analysis is in its potential contribution to reading instruction. For instance, deaf students may profit from direct instruction on how to monitor their own comprehension, using the strategies of skilled eaf readers as models. Recent self-monitoring training studies have shown that direct instruction of these metacognitive skills, indeed, increases the reading comprehension of underachieving readers (Palinscar, 1984). Future research needs to address the use of these strategies, given other comprehension formats than our use of a modified cloze procedure. In addition, the word-analysis strategies we found among the deaf readers could be developed into some kind of systematic teaching procedure.

Teaching metacognitive comprehension strategies to deaf students (or any students) will not necessarily guarantee successful reading comprehension. To acquire English proficiency and to fill in informational gaps are major obstacles facing deaf youths. Nevertheless, effective instructional approaches might be developed to enable deaf students to use these strategies as tools or aids to comprehend English texts.

The reading teacher plays a critical role for deaf students. During our observations of the deaf youths reading, our review of the videotapes, and our study of the written transcriptions, we observed that the experimenter functioned as "an explainer of English" to the deaf students. Much time was spent engaging in conversations with the youths and explaining vocabulary by giving the sign equivalent or providing an explanation in sign language to meanings of words, phrases, and sentences. Analogies were often used to explain meanings and concepts not familiar to the deaf students. This leads us to believe that the reading teacher or more skilled deaf reader must spend time individually with the deaf student, together working through the texts. The more expert reader must be able to explain difficult sections to the student using language the student can understand. Thus, we see reading comprehension and strategy use developed through collaboration on text meaning through explaining and interpreting texts. Only through this kind of ongoing, interactive dialogue can the deaf student acquire understanding of complex English texts.


Afterback, P., & Johnston, P. (1984). Research methodology on the use of verbal reports in reading research. Journal of Reading Behavior, 16, 307-322.

American National Standards Institute. (1969). American National Standard Specifications for Audiometers (ANSI 53.6-1969). New York: American National Standards Institute.

Baker, L., & Brown, A. (1984). Metacognitive skills and reading. In D. Pearson (Ed.), Handbook of reading research (pp. 353-394). New York: Longman.

Conrad, R. (1979). The deaf school child: Language and cognitive function. London: Harper & Row.

Ericsson, K., & Simon, H. (1980). Verval reports as data. Psychological Review, 87(3), 215-249.

Ewoldt, C. (1977). A psycholinguistic description of selected deaf children reading in sign language. Unpublished doctoral dissertation, Wayne State University.

Garner, R. (1982). Verbal-report data on reading strategies. Journal of Reading Behavior, 14, 159-176.

Houghton Mifflin Reading Program (1981). Boston: Houghton Mifflin.

LaSasso, C. (1978). An investigation of the validity and reliability of the cloze procedure as a measure of readability and comprehension for prelingually profoundly deaf students. Unpublished doctoral dissertation, University of Maryland.

Lichtenstein, E. (1984). Deaf working memory processes and English language skills. In D. Martin (Ed.), Cognition, education and deafness (pp. 111-114). Washington, DC: Gallaudet University Press.

Marshall, W. (1970). Contextual constraint on deaf and hearing children. American Annals of Deaf, 115, 382-389.

Moores, D. (1967). Applications of the "cloze" procedures to the assessment of psycholinguistic abilities of the deaf. Unpublished doctoral dissertation, University of Illinois.

Moores, D. (1987). Educating the deaf: Psychology, principles and practices. Boston: Houghton-Mifflin.

Quigley, S., & Paul, P. (1984). Language and deafness. San Diego: College-Hill Press.

Olshavsky, J. (1976-77). Reading as problem solving. An investigation of strategies. Reading Research Quarterly, 12(4), 654-674.

Palinscar, A. (1984). The quest for meaning from expository text: A teacher guided journey. In G. Duffy, L. Roechler, & J. Mason (Eds.), Comprehension instruction: Perspectives and suggestions (pp. 251-264). New York: Longman.

Soderbergh, R. (1985). Early reading with deaf children. Prospects, 15(1), 77-84.

Treiman, R., & Hirsch-Pasek, K. (1983). Silent reading: Insights from second generation deaf readers. Cognitive Psychology, 15, 39-65.

Wilbur, R. (1987). Reading and writing. In J. Cleve (Ed.), Gallaudet encyclopedia of deaf people and deafness (pp. 146-151). New York: McGraw-Hill.

JEAN F. ANDREWS (CEC Chapter #241) is an Associate Professor in the Department of Communications, Lamar University, Beaumont, Texas.

JANA M. MASON is a Professor at the Center for Study of Reading, Department of Educational Psychology and Department of Curriculum and Instruction, University of Illinois, Champaign.
COPYRIGHT 1991 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Andrews, Jean F.; Mason, Jana M.
Publication:Exceptional Children
Date:May 1, 1991
Previous Article:Longitudinal analysis of time, engagement, and achievement in at-risk versus non-risk students.
Next Article:Evaluating the fairness of special education hearings.

Related Articles
Babies sound off: the power of babble.
Teaching as transaction: Vygotskian perspectives on deafness and mother-child interaction.
Processing of bottom-up and top-down information by skilled and average deaf readers and implications for whole language instruction.
The perils of assimilation in modern France: the deaf community, social status, and educational opportunity, 1815-1870.
Applying social critical literacy theory to deaf education.
DB-link: National Information Clearinghouse on Children Who Are Deaf-Blind. (Better Speech and Hearing Month)(Organizational Spotlight).
Early intervention collaboration-deaf role models.
ESL REDEUX: when sign lanquage is first.
Repeated reading as an instructional intervention with deaf readers: effect on fluency and reading achievement.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters