Printer Friendly

Improving correct and error rate and reading comprehension using key words and previewing: a case report with a language minority student.

Language-minority students are "non-native English speaking students who lack full proficiency in English" (Scarcella, 1990, p.181). Many language-minority students who might be considered fluent in their use of English for interpersonal communication are not proficient enough readers of English to succeed in school (Carey, 1987). Lack of reading proficiency could be a major factor in the over representation of language-minority students in special education classes (Miramontes, 1987; Chinn & Hughes, 1987).

Much has been written about the relationship between the rate at which a student reads and comprehension of the reading material. Perfetti and Hogaboam (1975) supported the view that links fluent decoding and comprehension:
 If a reader requires considerable processing capacity to decode a
 single word, his processing capacity is less available for higher
 order integrated processes--for example, memory of the just
 previously coded word may suffer, memory for the preceding phrase
 may decrease and the subject's ability to 'predict' what he is yet
 to encounter on the printed page may diminish. (p. 461).

The research literature cites numerous studies linking reading fluency and comprehension (Deno, Mirkin, & Chang, 1982; Fleisher, Jenkins, & Pany, 1979, LaBerge & Samuels, 1994; Shapiro, 1989; Skinner, Cooper, & Cole, 1997). For example, Harris (1970) noted that slow readers do poorly in comprehension due to the fact that their many repetitions and hesitations break up the continuity of thought. "Unless readers become automatic with the alphabetic code, the time and attention required to identify a word directly limits the cognitive resources available to process the meaning of the sentence in which the word appeared" (Kameenui, 1998, p.329).

Efforts to increase decoding skills and, as a result, comprehension skills, have included repeated readings (Dowhower, 1987; Gilbert, Williams, & McLaughlin, 1996; Samuels, 1979), neurological impress method (Heckelman, 1986), and precision teaching methods such as timed and charted measures of student performance (Downs & Morin, 1990). Repeated readings involve having students orally reread passages until a satisfactory level of fluency is reached. Neurological impress method involves having the student read aloud and simultaneously with the teacher. The precision teaching method provided one minute timed readings, charting the student's best daily reading, providing feedback and having the student compete with himself.

Previewing reading material has been shown to increase oral reading proficiency among low achieving students (Sachs, 1984; Skinner et al., 1997). Listening-previewing is defined as any method that provides an opporunity for a learner to read or listen to a selection or passage prior to instruction and/or testing (Daley & Martens, 1994; Rose, 1984a). This has been shown to be effective with elementary general education students as well as with students with disabilities (Rose, 1984a,b,c; Rose & Sherry, 1984).

Discussion of key words has also been effective in increasing both factual and inferential reading comprehension as this strategy provides the reader with relevant prior knowledge of the subject (Roberts, 1988; Rousseau, Tam, & Ramnarain, 1993). Discussion of key words is defined as the teacher's discussing the meanings of key words from the reading passage prior to the student's reading the passage aloud (Rousseau & Tam, 1991). Discussion of key words helps to expand the vocabulary and comprehension of the learners and enables them to grasp the meaning of the passage more readily.

The results of a study carried out by Rousseau and Tam (1991) showed that two treatments, discussion of key words and listening-previewing, when presented together were more effective than either treatment presented alone. This would suggest that a combination of two interventions should be used for increasing oral reading proficiency and reading comprehension in language minority students with speech and language deficits.

Experiment I

The purpose of the study was to determine if the combination of listening-previewing and discussion of key words would increase the oral reading proficiency and the reading comprehension of a language minority student, therefore providing replication of Rousseau and Tam (1991). However, the study was extended to determine if the effects of the strategies maintained over a six month period. A second experiment which replicated the first with the same student documented the continued effectiveness of the methods across time.


Participant and Setting

The student, Ralph, was a 10 year old fifth grader in a regular education classroom at an inner city elementary school in eastern Washington. According to teacher reports, he was a very bright, motivated and friendly young man. The Chinese-American student who is the subject of this study is not considered to have a speech or language deficit. However, he was considered at-risk because there was evidence that his ability to think and to process in English and, therefore, to fully understand what he read, was lacking. In addition, he interacted very well with his classmates and appeared to be well liked. The staff felt he was a pleasure to work with and he had much support from his family and the faculty at his school.

The participant was born in Hong Kong and came to this country as an infant. He was raised in a home where his mother did not speak English and his father spoke only halting English. His father solicited help for his son from his fifth grade teacher who in turn sought to obtain a tutor. The experimenter spoke with the classroom teacher and father who were both concerned with the child's reading fluency and comprehension. A specific area of concern was the area of context and syntax in his reading. Information from his teacher indicated that {,A subject read at an acceptable pace but his level of comprehension was low. He was able to memorize the vocabulary words in isolation, but when he encountered them in a narrative or expository story, he did not have the experiences or tools to garner their relevance in context. In reviewing his standardized scores on the State Wide Achievement Tests given to fourth graders in January of 1998, he scored in the thirteenth percentile in reading, and his scores in all subjects were at least 50 points below the passing level.

The study took place after school in the school library or in the art room. The sessions took place two to three times per week and lasted for one hour.


Reading materials and sessions were presented in English. The reading materials used were taken from an Instructional Fair Reading Comprehension Booklet. Level 5, by Brewer (1984). Each passage averaged approximately 290 words (range: 266 to 303 words). Words such as and, a, an, and the, as well as prepositions of less than four letters, were not counted in the total number of words, nor were they counted in number of words read correctly. Stories from the beginning, middle and end of the booklet were used in each phase of the study. The correct reading rate and error reading rate were determined by marking errors on a separate written copy of the reading passage. Total number of words read correctly, total number of errors and answers to five comprehension questions were tallied on a data collection sheet (See Figure 1). The measures involved one minute readings which were timed by use of a stopwatch which was reset after each reading.


Dependent Variables and Measurement Procedures

The dependent variables were words read correctly and correct answers to comprehension questions. Mispronunciations, omissions, substitutions, repetitions, incorrect word endings and neglecting to pause at a period were counted as errors. Words read true to the text and words which were self-corrected within three seconds were counted as correct.

Correct answers to comprehension questions were defined as oral answers given by the student which corresponded to the experimenter's answer key. Five literal comprehension questions were written by the experimenter for each of the passages. The number of comprehension questions remained the same throughout the entire study, (i.e., five questions per session).

Reading scores were based upon one-minute timings of reading passages written at the fifth grade level. The one minute timings over reading material were recorded on a tape recorder for later scoring. The experimenter listened to the recordings after the sessions and the rate of words read correctly per minute and errors were calculated. Words read correctly were identified by subtracting words read incorrectly from the total words read. The number of comprehension questions correct and incorrect were also determined by listening to the tapes and comparing Ralph's answers with the answer key. For the comprehension questions, the student was given no prompts and was given five seconds to answer each question. His answers were recorded as correct or incorrect. No partial credit was given. The results for each session were recorded on a data collection sheet.

Design and Conditions

The experimental design employed for the study was an ABAB reversal design (Kazdin, 1982). Baseline lasted four sessions. Then intervention was implemented for 10 sessions. A return to baseline was implemented for four sessions. Intervention was implemented again for six sessions. Follow-up was measured on three days, six months after the last intervention session. The instructional aim for Ralph was 175 to 200 correct words per minute. This was determined by total number of words read less number or errors. The aim for comprehension was five correct answers for each story read. The passages were different each time he read, but were taken from the same book.

Baseline. During baseline, the student was given his copy of the passage for the session. The student was asked if he was ready to begin and then was directed to start. He was timed with a stop watch and told to stop after one minute. Errors were not corrected. The last word read was marked. Immediately after he finished reading he was orally asked five short answer questions about what he had just read. He answered the questions orally. The number of words read correctly was tallied and recorded. Also, the number of correct answers was noted.

Key words and previewing. Prior to each session, the experimenter and the student discussed the story they were about to read to make sure the student related to the topic. Before reading the passage, the experimenter identified key words in the passage. Key words were defined as words with which the student may not be familiar or which might present difficulty with definition, pronunciation, or the way they were used in the story. Some key words were hung, samurai, lacrosse and parliament. The experimenter said the word and asked the student to repeat it. Definitions for unfamiliar words, as well as the context in which the words were used in the story, were discussed. Questions were asked to determine if the student understood the meaning of the word. If he said that he knew a difficult word and did not need to review it, he was asked to define the word and was corrected as needed. The student was given an opportunity to ask questions about each of the words. After this analysis of key words, the author read the entire passage and the student followed along silently. He was encouraged to follow along with his finger under each word. This was done to ensure that the student was attending.

After the experimenter read the passage, the student was told to read the same passage aloud as fast as he could while still understanding what he read. The last word read was marked. Immediately after reading the passage, the student was asked five comprehension questions relating to the passage.

Follow-up. Approximately six months after the initial study was completed, the experimenter met with the student again to check for retention of skills. The setting was the same, however the student was now in the sixth grade. The experimenter randomly selected three of the original passages which had been used in the key words and previewing sessions and gave them to the participant to reread. There was no prior discussion of key words or topic, nor was there any previewing of the story. The student was given the passage to read orally, was timed and was asked five new, unfamiliar questions about the passage. The number of words read correctly was tallied and the number of correct answers was noted.


Interobserver agreement data were collected by means of a tape recorder which was used to record the student's readings and the answers to the questions. These recordings were used by both the experimenter and an independent observer who listened to them separately. Interobserver agreement was calculated for the number of correct words read in the story, the number of errors, and the number of comprehension questions answered correctly. Interobserver agreement was recorded for 42% of all sessions throughout the study. The total mean interobserver agreement for all conditions and behaviors was 93.3% with a range of 89.8% to 99%. Mean interobserver agreement was 92% (range 89% to 99%) on the number of words read correctly in the story, 89.8% (range 81% to 100%) on the number of reading errors, and 98% (range 83% to 100%) on the number of comprehension questions answered correctly.


The results of this study are displayed in Figures 2, 4, and 6. They illustrate Ralph's progress in reading fluency, reading errors and comprehension.



For baseline, the total number of words read correctly in one minute ranged from 135 to 154 with an average of 142.5. Errors ranged from 13 to 65 with an average of 35 errors during a one minute timing. In the area of reading comprehension, the range of accurate answers for the five questions ranged from 0 to 2 correct with an average of 1.0 correct.

With the return to baseline, Ralph's scores were similar to the previous baseline. His total number of words read correctly in one minute ranged from 155 to 166 with an average of 158.5 words. His number of errors ranged from 7 to 20 with an average of 14.8 errors per minute. The correct answers to the comprehension questions also declined. The range was from 0 correct to 3 correct with an average of 1.75 correct.

Key words and Previewing

When key words and previewing were first introduced, the total number of words read correctly in one minute ranged from 153 to 208 with an average of 181 words. The number or errors ranged from 4 to 12 with an average of 7 errors during a one minute timing. In the area of comprehension, the participant's scores ranged from 2 to 5 correct with an average number of correct answers of 4.4 over the ten sessions.

When the key words and previewing were reintroduced, the participant's reading and comprehension again improved. Over the four sessions, his number of words read correctly ranged from 169 to 181 with an average of 174 words. The number of errors ranged from 5 to 9 with an average of 7.75 errors during a one minute timing. Correct answers to the comprehension questions were consistently high. The range was from 4 to 5 answers correct with an average of 4.75 answers correct.


The results of the follow up portion of the study showed the total number of words read correctly in one minute ranged from 169 to 177, with an average of 172 words per minute. The number of reading errors ranged from 3 to 4 with an average of 3.7 per one minute timing. In the area of comprehension, his scores ranged from 3 to 4 correct with an average number of 3.7 answers correct.


The purposes of Experiment II were to replicate the findings from the first experiment and to extend those results through the next school year.


Approximately two weeks after the completion of the maintenance section of the original study, the experimenter again met with the student. These sessions were held twice a week in the same place in the same school as in the original study. The experimenter again used the Instructional Fair Reading Comprehension Booklet, Level 5. Data collection and reliability measures were conducted in the same manner as was employed originally. Baseline and key words and previewing conditions mirrored those of the original experiment. Ralph was given a novel set of passages to read.


Interobserver agreement was recorded for 40% of all sessions, once per condition. The total mean interobserver agreement for all conditions and behaviors was 93% with a range of 88% to 97%. Mean interobserver agreement was 95% (range 95% to 97%) on the number of words read correctly in the story, 88% (range 50% to 100%) on the number of reading errors, and 97% (range 80% to 100%) on the number of comprehension questions answered correctly.



The student was tested during three sessions of baseline. During these sessions his total number of words read correctly ranged from 133 to 165, with an average of 151 words read correctly. His number of errors ranged from 5 to 9, with an average of 6.3 per one minute reading. In the reading comprehension category, Ralph answered from 1 to 3 questions correctly, with an average of 2.3 questions correct.

When the key words and previewing were removed, Ralph read the three passages at an average speed of 171 words per minute. The range was from 166 to 177 words per minute. He made an average of 5.3 errors per passage with a range of 4 to 7 errors per minute. He answered an average of 3 questions per passage correctly His range was from 2 to 4 questions answered correctly (See Figures 3, 5 and 7).


Key Words and Previewing

During the first three sessions of key words and previewing, the student read the passages at an average rate of 171 words per minute. The range was 168 to 174. He made no errors on one passage and he made one error on two passages. This averaged .7 errors per passage. When asked the 5 questions over each passage, the student answered 4 questions correctly for one passage and 5 questions correctly for the other two passages. His average number correct was 4.7.

When the key words and previewing were reintroduced, Ralph averaged 188 words correct per minute. The range of words read per minute was 186 to 191. He made an average of .7 errors per reading, with a range of 0 to 1 errors per passage. He answered all 15 questions relating to the 3 passage correctly.


The results of this study demonstrated that by previewing through discussion of key words and having the student practice reading as the experimenter read and the student followed along, improvement was shown. The student's number of words read correctly increased while his number of errors decreased, and the level of comprehension increased. It was helpful to compare the results from Experiment I and Experiment II in each of the three skill areas measured.

In the area of correct words read in one minute, in the first experiment, the student began in baseline reading below 170 words per minute. In intervention his speed increased to above 170 fairly rapidly and maintained throughout this phase. When the key words and previewing were discontinued, his rate went down but not to the previous baseline level. When intervention was reintroduced, his speed again increased to a level above 170 words per minute. In follow-up, which was six months after the original study and covered previous material, his rate of reading was at or about the 170 words per minute level. These figures show a definite generalization effect over time. In the second experiment, his initial baseline score was approximately the same as the original baseline of the first experiment. However, it rapidly increased in baseline to almost 170. In the first intervention of experiment two, his rate averaged 171 words per minute. In baseline it went down but not to the level of the original baseline. Finally, in the last intervention, his rate of reading rose rapidly to above 180 words per minutes.

When looking at the number of errors, it was evident that this procedure over time continued to decrease errors. The baseline levels in all four instances, show that the number of errors was less in each succeeding baseline. Furthermore, the key words and previewing resulted in progressively fewer errors and ultimately resulted in an average of less than one error per reading in the second experiment. The results found in the follow-up portion most likely were effected by the fact that the student had seen the material before and had an additional 6 months to practice reading. This is born out by comparing his follow up scores in this area (number of errors) with those of the first baseline in experiment two.

In examining the results of the comprehension question portion of the study, it was interesting to note that throughout both experiments, it took the student less time to get all 5 questions correct with each succeeding intervention. In the first study, he maintained a relatively high rate of correct answers in follow-up, but he did not realize the same level he had reached with key words and previewing. This was most definitely influenced by the fact that, although the reading passage was familiar, the questions asked were not the same questions used initially.

It was noteworthy that the student's actual number of words read was quite high, even in baseline. However the number of reading errors was also very high. After the intervention the number of words read per minute remained high while the number of words read incorrectly decreased. When looking at the number of correct answers to the comprehension questions, it is easy to draw a correlation between familiarity with the vocabulary and the context in which it is used, correct pronunciation of the words, and understanding what was read. Understanding the meaning of the key words enables the learner to use contextual information to facilitate word recognition (Stanovich, Cunningham, & Freeman, 1984). In baseline, he "read" the words but the errors intefered with his comprehension. Actually hearing the words read and understanding their meaning in the context of the passage before he read them, increased his comprehension to a level commensurate with his rate of oral reading.

As the student was not enrolled in a resource room but, rather, a regular classroom, the procedures would require the teacher to devote a fair amount of time to implement them. Unless the teacher had a group of four or five students at the same level, these procedures may not be practical. However, the procedures could be employed by other adults (e.g., volunteers, para-educators, teacher-aids, etc.).

Since the experimenter was focusing on increasing both number of words read correctly and comprehension levels, the decision was made to use both previewing and discussion of key words. These findings support previous studies, particularly those done by Rose (1984a,b,c), Rousseau and Tam (1991), and Skinner et al. (1997). However, it would be interesting to see if discussion of key words alone or previewing alone would increase comprehension rates.

For this language minority student with good phonics skills but a vague sphere of reference when faced with unfamiliar vocabulary, key words and previewing is a beneficial intervention. In both correct words read and comprehension, when the treatments were removed, Ralph did not return to the low levels of the original baseline. These results point to the reality that there could be some lasting and perhaps permanent effects from employing these strategies.

The techniques described in this study have been proven effective but time consuming. However, we must remember that children in our schools learn in diverse ways and many
 ... constantly face the tyranny of time in trying to catch up with
 their peers, who continue to advance in their literacy development.
 The pedagogical clock for students who are behind in reading and
 literacy development continues to tick mercilessly, and the
 opportunities for these students to advance or catch up diminish
 over time. (Kameenui, 1993. p. 379)

We owe it to each child to do what we can to give that child a better chance in life.


Brewer, P. (1984). Instructional fair reading comprehension homework booklet. Grand Rapids, MI: Instructional Fair.

Carey, S. T. (1987). Comprehension in the first and second language of immersion and Francophone students. Canadian Journal for Exceptional Children, 34, 103-108.

Chinn, P. C., & Hughes, S. (1987). Representation of minority students in special education classes. Remedial & Special Education, 8, 41-46.

Daly, E. J., III, & Martens, B. K. (1994). A comparison of three interventions for increasing oral reading performance: Application of the instructional hierarchy. Journal of Applied Behavior Analysis 27, 459-470.

Deno, S. J., Mirkin, P. K., & Chiang, B. (1982). Identifying valid measures of reading. Exceptional Children. 49, 36-45.

Dowhower, S. L. (1987). Effects of repeated readings on second grade transitional readers' fluency and comprehension. Reading Research Quarterly, 15, 30-48.

Downs, J. & Morin, S. (1990). Improving reading fluency with precision teaching. Teaching Exceptional Children, 22, 38-40.

Fleisher, L., Jenkins, J., & Pany, D. (1979). The effects on poor readers' comprehension of training in rapid decoding. Reading Research Quarterly, 15, 30-48

Gilbert, L, Williams, R. L., & McLaughlin, T. F. (1996). Use of assisted reading to increase correct reading rates and decrease errors of students with learning disabilities. Journal of Applied Behavior Analysis, 29, 255-257.

Harris, A. (1970). How to increase reading ability. New York: David McKay.

Heckelman, R. G. (1986) NIM revisited. Academic Therapy Quarterly, 21, 411-420.

Kameenui, E. J. (1993). Diverse learners and the tyranny of time: Don't fix blame; fix the leaky roof. The Reading Teacher, 46, 376-383.

Kameenui, E. J. (1998). The rhetoric of all, the reality of some, and the unmistakable smell of mortality. In J. Osborn & F. Lehr (Eds.), Literacy for all (pp. 319-338). New York: Guilford.

Kazdin, A. E. (1982). Single case research designs: Methods for clinical and applied settings. New York: Oxford University Press.

LaBerge, D., & Samuels, S. J., (1994). Toward a theory of automatic information processing in reading. Cognitive Psychology, 6, 292-323.

Miramontes, O. (1987). Reading miscues of Hispanic students: Implications for assessment of learning disabilities. Journal of Learning Disabilities, 20, 627-632.

Perfetti, C., & Hogaboam, T., (1975). Relationship between single word decoding and reading comprehension skills. Journal of Educational Psychology, 6, 461-469.

Roberts, T. A. (1988). Development of pre-instruction versus previous experience: Effects on factual and inferential comprehension. Reading Psychology: An International Quarterly, 9, 141-157.

Rose, T. L. (1984a). The effects of previewing on retarded learners' oral reading. Education and Treatment of the Mentally Retarded, 19, 49-52.

Rose, T. L. (1984b). Effects of previewing on the oral reading of mainstreamed behaviorally disordered students. Behavioral Disorders, 10, 33-39.

Rose, T. L. (1984c). The effects of two prepractice procedures on oral reading. Journal of Learning Disabilities, 17, 544-546.

Rose, T. L., & Sherry, L. (1984). Relative effects of two previewing procedures on LD adolescents' oral reading performance. Learning Disabilities Quarterly, 7, 39-44.

Rousseau, M. K., & Tam, B. K. Y. T., (1991). The efficacy of previewing and discussion of key words on the oral reading proficiency of bilingual learners with speech and language impairments. Education and Treatment of Children, 14, 199-209.

Rousseau, M. K., & Tam, B. K. Y. T., & Ramnarain, R. (1993). Increasing reading proficiency of language-minority students with speech and language impairments. Education and Treatment of Children, 16, 254-271.

Sachs, A. (1984). The effects of previewing activities on oral reading miscues. Remedial & Special Education 5, 45-49.

Samuels, S. J. (1979). The method of repeated readings. The Reading Teacher, 32, 403-408.

Scarcella, R. C. (1992). The tapestry of language learning: The individual in the communicative classroom. Boston, MA: Heinle & Heinle.

Shapiro, E. S. (1989). Skills Problems: Direct assessment and intervention (2nd. Ed.). New York: Guilford.

Skinner, C. H., Cooper, L., & Cole, C. L. (1997). The effects of oral presentation previewing rates on reading performance. Journal of Applied Behavior Analysis, 30, 331-334.

Stanovich, K. E., Cunningham, A. E., & Freeman, D. J. (1984). Relation between early reading acquisition and word decoding with and without context: A longitudinal study of first-grade children. Journal of Educational Psychology, 76, 668-677.

Preparation of this paper was in partial fulfillment of the requirements for a Master's Degree in Special Education, Gonzaga University, Spokane, WA. Address correspondence to Kimberly P. Weber, Special Education Program, Gonzaga University, Spokane, WA 99258-0001 (E-mail:

Patricia O'Donnell

Kimberly P. Weber

T. F. McLaughlin

Gonzaga University
COPYRIGHT 2003 West Virginia University Press, University of West Virginia
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:O'Donnell, Patricia; Weber, Kimberly P.; McLaughlin, T.F.
Publication:Education & Treatment of Children
Date:Aug 1, 2003
Previous Article:The achievement of students with developmental disabilities and their peers without disabilities in inclusive settings: an exploratory study.
Next Article:The effects of three techniques on student participation with preschool children with attending problems.

Related Articles
Cognates and other clues: strategies for comprehension. (Language Teaching & Learning).
Reading comprehension and fluency: examining the effects of tutoring and video self-modeling on first-grade students with reading difficulties.
Use of story-mapping to increase the story-grammar text comprehension of elementary students with learning disabilities.
A comparison of the effects of two rates of listening while reading on oral reading fluency and reading comprehension.
Fostering the development of vocabulary knowledge and reading comprehension though contextually-based multiple meaning vocabulary instruction.
Charging Up! Reading comprehension program: ETA/Cuisenaire.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters