Effect of text presentation on reading fluency and comprehension: an exploratory analysis.

Previous research found that the type of text being read affected reading fluency and recall, and students of varying reading skills take different approaches to the task. The current study explored the effect the reader's skill level and format of presentation had on fluency and comprehension assessment by comparing student behavior between conditions in which students read from a book or from a typed single page. No significant differences were found in mean fluency and comprehension scores between the Book and Typed conditions for the two highest reading skills groups. However, the difference between the Typed and Book conditions was significant for both fluency and comprehension among children in the lowest reading group. Students in the lowest reading group read more fluently and with better comprehension from the Book condition than from the Typed condition. Suggestions for further research are included.

**********

Reading continues to be the academic area most frequently identified as deficient in special education eligibility referrals (Joseph, 2002) and a large majority of children identified as learning disabled have a deficit in reading (Lerner, 2003). Thus, many academic intervention models exist for reading and scholars frequently recommend using fluency measures such as curriculum-based measurement (CBM; Deno, 1985) for making instructional decisions or problem identification (Fuchs & Fuchs, 1998; Gresham, 2001; Kaminski & Good, 1998; Shinn, 2002). Moreover, reading fluency measures are also useful tools for large-scale screening of student skills (Berninger, 2002).

Administration procedures for reading fluency probes, as outlined by Shinn (1989), involve having the student read orally for 1 minute from a typed passage written on a page that contains no additional markings, such as pictures. The number of words read correctly within the timed probe is counted to obtain a score that represents reading fluency. Adequate fluency in reading is important because reading fluency and comprehension have consistently been linked and comprehension is the ultimate goal of reading (Burns, Tucker, Hauser, Thelen, Holmes, & White, 2002; Fuchs, Fuchs, Hosp & Jenkins, 2001; McCormick & Samuels, 1979). Fluent reading consists of quick and effortless reading of words presented in text (Carmine, Silbert, Kame'enui, & Tarver, 2004) but how that is measured is a matter of some debate (Burns, 2003).

Reading fluency could be assessed with speed of accurate contextual reading (Nathan & Stanovich, 1991) or by whole word reading without context (Torgesen, Wagner, & Rashotte, 1997). However, the format of the assessment could affect the results (Burns, 2003). Further, Jenkins, Fuchs, van den Broek, Espin and Deno (2003) compared contextual reading fluency and context-free fluency and found that fluency from contextual reading was more strongly correlated with comprehension (r = .83) than context-free fluency such as word lists (r = .54). Fluency of contextual reading predicted reading comprehension, but fluency of reading word lists did not (Jenkins et al., 2003). This finding makes some intuitive sense given that, in general, the instructional environment affects student performance (Ysseldyke & Christenson, 2002). An instructional environment consists of the situational variables that interact with the student to determine the resulting behavior (Ysseldyke & Christenson, 1987). Comparatively, a reading environment would then encompass those situational variables that are specific to the reading task and interact with the student to determine reading behavior. Thus it seems reasonable to infer that environmental variables such as assessment format (e.g., reading words within or without context) would affect reading outcomes.

Although the nature of the reading task affected the outcome in the Jenkins et al. (2003) study, the causal hypothesis intertwined reading task and skill. It was concluded that "context-free reading skill appears to make a larger contribution to context fluency for less fluent readers, whereas comprehension processes contribute more to context fluency for more fluent readers" (Jenkins et al., p. 727). In other words, reading skill level and assessment format seemed to determine what factors limited reading ability (e.g., language development and word identification skills). Previous research also found qualitative differences in how good and low readers typically decode unknown text (Kim & Goetz, 1994; Stanovich, 1986), but none examined the interaction with assessment format.

Previous studies have also shown that students recall information more effectively when reading narrative texts as compared to expository texts, even after controlling for difficulty and students demonstrated higher fluency rates when reading narrative texts as opposed to expository (Petros, Bentz, Hammes, & Zehr, 1990; Zabrucky & Moore, 1999; Zabrucky & Ratner, 1992). Hypotheses as to why these differences in fluency and recall occurred included differences in purpose for reading the texts (Narvaez, van den Broek, & Ruiz, 1999), differences in familiarity and predictability between narrative and expository texts (Zabrucky & Moore, 1999) and difficulty with understanding the logical relations with expository text (Petros et al., 1990). Moreover, expectations that students have about the text before and during reading can differ based on the type of text (i.e., expository vs. narrative; Kameenui & Simmons, 1990) and student expectations of text can affect student attitude, interest, and performance as well (Cunningham & Gall, 1990; d'Ydewalle, & Van Dam, 1988). Therefore, while there may be many plausible explanations for this behavior, it appears that student expectations of their reading environment can affect reading fluency and recall.

Research on the effects of reading environment format on fluency and recall to date is limited primarily to expository and narrative comparisons and other implications seem intuitively reasonable. For example, it could be stated that children expect reading books containing pictures of characters, activity, and novelty, all of which are characteristics of narrative text, to be more interesting than reading material without these components (Anderson, Shirey, Wilson, & Fielding, 1987; Cunningham & Gall, 1990). Moreover, student perceptions of task requirements are often more important to responding than the actual requirements of the task (Hong, 1999), and factors such as assumed text difficulty determine the expectancy for success in the reading task (Stock, Winston, Behrens, & Harper-Marinick, 1989). Therefore, it could be hypothesized that other reading environmental factors such as presentation of reading task (e.g., reading from a book as opposed to reading a single typed page), would also be influenced by expectancy of text and student behavior could change as a result.

Reading is clearly a complex process involving many different yet related skills (e.g., word identification, comprehension). Given the interdependent nature of the various subskills in reading, oral reading fluency was often recommended as a global measure of reading skill (Burns et al., 2002; Jenkins et al., 2003). However, factors such as reading skill and task format could affect reading behaviors of students. Moreover, fluency assessments remain the most widely used and researched reading assessment models in special education, but the effect of text format as it interacts with student skill has not been evaluated. Therefore, the current study examined the effect the reader's skill level and text format for the reading task have on fluency and comprehension, by comparing student behavior between conditions in which students read from a book or from a typed single page. It was hypothesized that the interaction of reading skill and condition (i.e., Book or Typed) would significantly affect measurements of reading fluency and comprehension. Given the lack of empirical investigation in this area, the current study was exploratory in nature with the goal of investigating the need for additional research.

Method

Participants

The study participants were 119 third-grade students from five elementary schools in Michigan. Just over half of the students were male (N = 68, 57.1%) with 42.9% being female (N = 51). A majority of the children (N = 99, 83.2%) were Caucasian, 16 (13.4%) were African-American, one (.8%) was Hispanic, one (.8%) was of Asian descent, and two (1.7%) were Native American.

The participants were divided into three groups based on reading ability (low, average, and high) with similar demographic distributions between them. The low group (n = 30) consisted of 12 (40%) females and 18 (60%) males, and three (10%) African-American students and 27 (90%) Caucasian students. The average group (n = 44) consisted 20 (45.5%) females and 24 (54.5%) males, and five (11.4%)African-American students, 38 (86.4%) were Caucasian, and one (2.3%) was Native-American. Finally, the high group (n = 45) consisted of 19 (42.4%) female and 26 (57.8%) male students, and eight (17.8%) African-American students, 34 (75.6%) Caucasian students, one (2.2%) Hispanic, one (2.2%) Asian-American, and one (2.2%) Native American student.

Independent Variables

Students were randomly assigned to one of two conditions; reading text from a book (Book) or reading the same text that was typed on one piece of white paper (Typed). A relatively equal distribution of conditions was achieved with 58 (48.7%) students in the Book condition and 61 (51.3%) students in the Typed condition. In addition, results of the 2002-2003 statewide assessment, called the Michigan Literacy Progress Profile (MLPP; Michigan Department of Education, 2001) were used to group the students by level of reading skill. MLPP is an assessment tool used to track the progress of students in kindergarten through third grade. Every academic year, teachers are required to assess students in a number of different areas. For the purpose of this study, the students' scores for Oral Reading Fluency and Comprehension of a Narrative Text were used.

Oral Reading Fluency was assessed within MLPP by determining words read correctly per minute from standardized grade-level probes and converting that score to a category score of one to four. The rubric of one to four was intended to provide a qualitative descriptor of reading fluency skills. A 4 indicated that the student demonstrated fluent, phrased reading, read primarily in larger meaningful phrases, read with only a few word-by-word slow downs for problem-solving, demonstrated expressive interpretation at places throughout the reading and gave attention to punctuation and syntax. A 3 indicated reading with a mixture of word-by-word and fluent reading with attention to punctuation and syntax, and an occasional reread for problem solving. A 2 was scored for students who demonstrated word-by-word reading, but with some two-, three-, and four-word phrases, and demonstrated awareness of syntax and punctuation. A 1 on MLPP Oral Reading Fluency indicated that the student demonstrated little fluency with all word-by-word reading, long pauses between words and little evidence of awareness of punctuation.

The MLPP assessed comprehension by having the student read a narrative text and assigning a score of one to four based on the students' retelling of the passage. A 4 suggested that the student retelling included a clear generalization with a correct main idea and a clear statement of all the story elements (i.e., main characters, setting, problem, major events, resolution) in a logical sequence (MDE, 2001). A student was rated as a 3 if retellings included a generalization that stated or implied the main idea, contained a clear statement of most of the story elements and the events were retold mostly in appropriate order. A 2 indicated that the student's retelling contained inaccuracies or incomplete understanding of the main idea, some story elements were detailed with minimal connections to one another and the events were retold in a somewhat disconnected fashion. A score of 1 suggested significant difficulty with reading comprehension in that the retelling included minimal or no understanding of the main idea and minimal story elements with a lack of sequence and/or format of a story.

Student reading scores were derived for the current study by averaging the students' Oral Reading Fluency and Comprehension of a Narrative Text scores from his or her second grade assessment. The resulting mean score was used to categorize the students as high (i.e., 3.0 to 4.0), average (i.e., 2.0 to 2.9), or low (i.e., 1.0 to 1.9) readers. Although MLPP is mandated for all students in Michigan between kindergarten and grade three, no psychometric data for the tool are available. Fluency measures in general have been shown to be reliable (Marston, 1989), but categorical scores, rather than fluency data gathered in MLPP, were reported. Thus, the reliability of this tool can only be inferred and not assessed. However, retell scores using a rubric of 0 to 2 demonstrated an average interscorer reliability coefficient of .84 (Rabren, Darch, & Eaves, 2000). Moreover, the Oral Reading Fluency and Comprehension of a Narrative Text scores for the children in the current study were correlated using Pearson Product Moment and produced a strong correlation (r = .90).

Dependent Variables

Reading fluency and comprehension for the Book and Typed conditions were used as the dependent variables for the study. The same reading passage was used for both conditions (i.e., Book and Typed), and was taken from the Burns and Roe Informal Reading Inventory (BRIRI, 1999). The BRIRI was selected because it contains contextual reading and six comprehension questions for each graded passage, and because it was described as one of the better informal reading inventories (Green, 2001). The readability level of the BRIRI third-grade passages was determined with the Spache Readability Formula, according to the test manual (Burns & Roe, 1999). There are four passages for each grade level of the BRIRI. Passage 3C, which contains 171 words, was randomly selected for this study. Readability was confirmed by the primary author with the Harris-Jacobsen Wide Range Readability Formula (Harris & Jacobsen, 1982), which resulted in a reading level of 3.2.

The 3C passage was taken from the book A Bowl of Sun by Frances Wosmek. The exact same passage was read from the actual book for the Book condition. The font size of the pages was unknown, but the height and width of the letters in both the typed and book stimuli were measured with a ruler and determined to be the same, and thus the passages were apparently equal in size. Common administration procedures for fluency assessments (Shinn, 1989) were used. This involved having the child read orally while being timed, having the administrator state any words that the child did not attempt within 2 seconds, and recording the number of words read correctly. The same procedure was followed in both conditions. The number of words read correctly was then divided by the time required to complete the passage resulting in a words read correctly per minute score, or fluency score. The primary author observed the data collection for 20% of the participants and also scored the responses. The fluency scores from the two sets of observations were then correlated with Pearson Product Moment Correlation to assess interrater reliability, which resulted in a coefficient of .99.

Six open-ended comprehension questions were provided for the passage; three inferential and three literal. After completing the passage, each student was orally presented the six comprehension questions and was asked to provide a verbal answer. The number answered correctly was recorded. A coefficient alpha was computed for the comprehension questions and resulted in a score of .82, which suggested adequate reliability for research purposes. Interobserver agreement was computed for the comprehension items by dividing the number of items that were agreed upon by the two observers and dividing by the total number of items, which resulted in 92% agreement.

Procedures

The principals of six elementary schools in Michigan were contacted via letter and a phone call to recruit their participation. Upon receiving written permission from each school, informed consent forms describing the project were sent home to the parents of the children in all of the third-grade classrooms. Those that returned the signed informed consent forms were included in the study. As stated earlier, 119 students returned the form and were included in the study. The form was sent home to 162 students, resulting in a 73.5% return rate. Teachers in each of the six classrooms were given a $25 gift certificate to a local business supply store, and children who participated were given a book.

Students were seen individually at their elementary school for one session lasting approximately 5 minutes. Students in the Book condition were asked to read orally from one page of the book A Bowl of Sun that corresponded to the passage in the BRIRI. The passage was found on page eight of a 47-page book. These students were presented with the closed book, with the cover facing them, and were asked to turn to page eight. Students from the Typed condition were given one sheet of paper with the passage typed on it. Neither the page in the book nor the typed page contained any pictures or any other markings besides words. Students in both conditions were then asked to (1) "Read this page out loud and do your very best reading," (2) "Keep reading until you are finished," (3) "If you come to a word you do not know, I'll tell you what it is," and (4) "When you are finished, I will have some questions for you about what you have read." Finally, after reading the timed passage, students in both conditions answered the same six open-ended comprehension questions, which were presented and answered orally.

One undergraduate student researcher collected data for this study. The student researcher completed two undergraduate assessment courses, one of which included an 8-hour/week field experience in conducting various educational assessments including reading fluency. Further, the student researcher was individually trained in administration of the study protocol and confidentiality of data during one 2-hour session. The student researcher was also observed while practicing the study protocol with three students before beginning the data collection. The fidelity of protocol implementation was assessed by counting the number of steps correctly implemented. The average percentage of correctly implemented protocol steps cumulatively recorded across the three students was 93.3%.

Results

AMANOVA was conducted to examine if reading skill groups determined by MLPP affected fluency and comprehension scores. Results suggested a significant effect F (2, 116) = 52.60, p <.001. Mean fluency and comprehension scores for the three groups are listed in Table 1. Six planned comparisons were conducted using a Bonferroni correction that lowered the alpha level required for significance to .008. Results of these analyses found that students from the high reading group had significantly higher mean fluency scores than the average t (87) = 4.62, p < .001 and low t (73) = 4.84, p < .001 groups, and the mean fluency score for the average group was significantly greater t (72) = 6.32, p < .001 than the mean fluency score for the low group. Further, the highest group also correctly answered significantly more comprehension questions than the low group [t.bar] (73) = 4.84, p < .001, and the average group [t.bar] (87) = 3.04, p = .003. However, the difference between mean comprehension questions answered correctly between the average and low groups was not significant t (72) = 2.20, p = .03.

After evaluating the reading skill groups, a two-way multivariate analysis of variance was used to examine fluency and comprehension scores using a two (Book or Typed condition)-by-three (high, average, and low reading skills) factorial. The results of this analysis are also listed in Table 1. A significant interaction between the two variables for reading fluency and comprehension F (2, 113) = 7.91, p < .001 was found. Students in the Typed condition whose reading skills fell within the lowest category demonstrated the lowest average fluency (33.7 words/minute) and comprehension (.9 questions correct). Six planned comparisons between groups were conducted using a Bonferroni correction, which adjusted the alpha level required for significance to .008. Results found no significant differences between mean fluency and comprehension scores between the Book and Typed conditions for the two highest reading skills groups. However, among children in the lowest reading group, the difference between the Typed and Book conditions was significant for both fluency t (28) = 3.29, p = .003 and comprehension t (28) = 3.70, p = .001, with students from the Book condition out performing those from the Typed condition on both variables. Cohen's d was also computed as an estimate of effect size to further examine the differences between these two groups. The effect size for the Book and Typed comparison among children who were low readers was 1.25 standard deviation units, and 1.41 standard deviation units for the comprehension means.

Discussion

The purpose of this study was to explore the effects of skill level and text format on reading fluency and comprehension. Results suggested that for average to above average readers, the presentation of the reading material (i.e., book or text) did not significantly affect fluency or comprehension. However, among the lowest readers, students read more fluently and with more accurate comprehension when reading from a book as opposed to one typed page, even though the passage was the same. Using Cohen's (1989) interpretative criteria for effect sizes of .80 as a strong effect, the effects for both reading fluency and comprehension among the lowest readers were strong.

The mean fluency rates for students in the Book and Typed conditions could have meaningful interpretations, given that previous research found a minimum reading fluency rate of approximately 50 words per minute to be necessary for comprehension to occur (Burns et al., 2002). The mean fluency rate for the Book condition among low readers was above that criterion, but the mean fluency rate for the Typed condition was well below 50 words per minute. Further, previous fluency recommendations for reading instruction were 50 to 99 words per minute (Deno & Mirkin, 1977) and 45 to 65 words per minute (Lovitt & Hansen, 1976). The Book condition fell within those recommended levels, but the mean of children in the Typed condition did not. Therefore, it could be assumed that whether or not a child with reading difficulties meets a specified criterion for a fluency assessment could be influenced by the format of the assessment.

Reading fluency data collected over time are also frequently presented as crucial aspects of progress monitoring efforts (Deno, 2002). Thus, given the differences in fluency rates among low readers, data from the current study reinforce the importance of maintaining the reading assessment format throughout the progress monitoring probes. Changing probe format (e.g., between typed pages and pages from a book) could affect fluency rates and practitioners would not be certain that changes in fluency were due to changes in reading skills as opposed to differences linked to assessment format. However, data from the current study should only be used cautiously to make recommendations for practice.

Although the mean score for reading comprehension was higher for low readers in the Book condition than those in the Typed condition, it is unknown if the format actually affected comprehension. Given that the fluency of the Typed condition fell below the aforementioned minimum fluency rate necessary for comprehension, and the link between fluency and comprehension in general, the lower comprehension score could be due to the slower reading fluency. Additional research is needed to examine the effect of expectancy and assessment format on reading comprehension.

The significant differences in fluency and comprehension between the reading skills groups supported the validity of the MLPP classification scores. However, these scores were derived from teacher-administered assessments for which no psychometric data were available. Although these scores were used as part of Michigan's accountability assessment system, relying on these scores for categorization of skills was a limitation of the current study. Future researchers may wish to categorize reading ability with assessments for which sound psychometric data are readily available.

Current results were consistent with previous research that demonstrated differences between adequate and low readers' approaches to reading text (Kim & Goetz, 1994). These differences in reading approaches could be due to expectations of the reading environment (e.g., presence of pictures, characters, or other aspects of the book), even though the selected page lacked pictures, which has been proposed as a reason why students read more fluently from narrative texts as compared to text passages that lack these components (Anderson, Shirey, Wilson, & Fielding, 1984; Cunningham & Gall, 1990). Thus, students may have expected something different from the book as opposed to a typed page, and their behavior was different as a result. However, this hypothesis was not tested with the current data and additional research is needed to accept or reject any proposed causal hypothesis. Perhaps future researchers could replicate this study with low readers and measure expectations of their reading environment, specifically of the reading passage after presenting the passage, in either book or typed format, but before asking them to read the text. Expectations between groups could then be compared.

The focus of this study was to identify areas in need of additional research. The small number of children in the Low reading group (n = 30) resulted in small numbers in the Low Book and Low Typed conditions, which could have affected the results. Thus, future research is needed with more participants. Future research could also focus on children with low reading skills, which may allow for a larger sample of children in the Book and Typed conditions. Once theory regarding the difference in reading performance of children with low reading skills between the different formats of fluency assessment is clarified, implications for practice can be explored. Because fluency assessment is recommended as a screening instrument (Berninger, 2002), it may be useful to determine if differences in identification rates would occur as a result of reading fluency format. Decision-making theories discuss implications of identification error such as false negatives and false positives (Kaplan & Saccuzzo, 2001). If screening efforts are linked to special education eligibility and/or identification decisions, implications of assessment format are critical, but if screening leads to remediation efforts and improved student learning with adequate resources, then over and under identification become less crucial. Thus, implications of assessment format for screening purposes should be evaluated from this perspective.

It is hoped that the current study will lead to additional research that extends and applies theories of text expectancy, reading environment, and reading skill in order to examine implications for school psychological theory and practice. Given that reading continues to be a priority for assessment, interventions and research, further research in this area appears warranted.

References

Anderson, R., Shirey, L., Wilson, P., & Fielding, L. (1984). Interestingness of children's reading material. Technical report no. 323. Urbana, IL: Illinois University Center for the Study of Reading (ERIC Reproduction number ED 248487).

Anderson, R., Shirey, L., Wilson, P., & Fielding, L. (1987). Interestingness of children's reading material. In R. E. Snow & M. Rarr (Eds.) Aptitude learning and instruction, Volume 3: Cognitive and affective process analysis (pp. 287-299). Hillsdale, NJ: Erlbaum.

Berninger, V. W. (2002). Best practices in reading, writing, and math assessment-intervention links: A systems approach for schools, classrooms, and individuals. In A. Thomas & J. Grimes (Eds.) Best practices in school psychology IV (pp. 851-866). Bethesda, MD: National Association of School Psychologists.

Burns, M. K. (2003). Reexamining data from the National Reading Panel's meta-analysis: Implications for school psychology. Psychology in the Schools, 40, 605-612.

Burns, M. K., Tucker, J. A., Hauser, A., Thelen, R., Holmes, K., & White, K. (2002). Minimum reading fluency rate necessary for comprehension: A potential criterion for curriculum-based assessments. Assessment for Effective Intervention, 28, 1-7.

Burns, P. C., & Roe, B. D. (1999). Informal reading inventory: Preprimer to twelfth grade (5th ed.). Boston: Houghton Mifflin.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.

Cunningham, L. J., & Gall, M. D. (1990). The effects of expository and narrative prose on student achievement and attitudes toward textbooks. Journal of Experimental Education, 58, 165-175.

Deno, S. L. (1985). Curriculum-based measurement: The emerging alternative. Exceptional Children, 52, 219-232.

Deno, S. L., & Mirkin, P. K. (1977). Data-based program modifications: A manual. Minneapolis, MN: Leadership Training Institute for Special Education.

d'Ydewalle, G., & Van Dam, G. (1988). Contextual information effects in text processing as a function of the type of test expected and received. Current Psychology of Cognition, 8, 139-152.

Fuchs, L. S., & Fuchs, D. (1998). Treatment validity: Aunifying concept for reconceptualizing the identification of learning disabilities. Learning Disabilities Research & Practice, 13, 204-219.

Fuchs, L. S., Fuchs, D., Hosp, M. K., & Jenkins, J.R. (2001). Oral reading fluency as an indicator of reading competence: A theoretical empirical and historical analysis. Scientific Studies of Reading 5, 239-256.

Green, F. J. (2001). Test review of the Burns and Roe informal reading inventory: Preprimer to twelfth grade (5th ed.). In B. S. Plake & J. C. Impara (eds.) Fourteenth mental measurements yearbook (pp. 195-196). Lincoln, NE: Buros Institute.

Gresham, F. (200l, August). Responsiveness to intervention: An alternative approach to the identification of learning disabilities. Paper presented at the Learning Disabilities Summit: Building a Foundation for the Future, Washington, DC.

Harris, A. J., & Jacobson, M. D. (1982). Basic reading vocabularies. New York: Macmillan.

Hong, E. (1999, April). Effects of gender, math ability, trait test anxiety, statistics course anxiety, statistics achievement, and perceived test difficulty on state test anxiety. Paper presented at the Annual Meeting of the American Educational Research Association. Montreal, Quebec, Canada. (ERIC Document Reproduction Service No. ED432499).

Jenkins, J. R., Fuchs, L. S., van den Broek, P., Espin, C., & Deno, S. L. (2003). Accuracy and fluency in list and context reading of skilled and RD groups: Absolute and relative performance levels. Learning Disabilities Research and Practice, 18, 237-245.

Joseph, L. M. (2002). Planning interventions for students with reading problems. In A. Thomas and J, Grimes (Eds). Best Practices in School Psychology (4th ed., pp. 803-816). Bethesda, MD: National Association of School Psychologists.

Kaplan, R. M. & Saccuzzo, D. P. (2001). Psychological testing: Principles, applications, and issues (5th ed.). Belmont, CA: Wadsworth/ Thomson Learning

Kameenui, E. J., & Simmons, D.C. (1990). Designing instructional strategies: The prevention of academic learning problems. Englewood Cliffs, NJ: Macmillan.

Kim, Y. H., & Goetz, E. T. (1994). Context effects on word recognition and reading comprehension of poor and good readers: A test of the interactive-compensatory hypothesis. Reading Research Quarterly, 29, 178-188.

Lerner, J. (2003). Learning disabilities: Theories, diagnosis, and teaching strategies. Boston: Houghton Mifflin.

Lovitt, T. C., & Hansen, C. L. (1976). Round one--placing the child in the right reader. Journal of Learning Disabilities, 9, 347-353.

Marston, D. B. (1989). A curriculum-based measurement approach to assessing academic performance: What it is and why do it. In M. R. Shinn (Ed.), Curriculum-Based Measurement: Assessing special children (pp. 18-78). New York: Guilford.

McCormick, C., & Samuels, S. J. (1979). Word recognition by second graders: The unit of perception and interrelationships among accuracy, latency, and comprehension. Journal of Reading Behavior, 11, 107-118.

Michigan Department of Education. (2001). Michigan literacy progress profile (2nd ed.). Lansing, MI: author.

Narvaez, D., van den Broek, P., & Ruiz, A. B. (1999). The influence of reading purpose on inference generation and comprehension in reading. Journal of Educational Psychology, 91, 488-496.

Nathan, R. G., & Stanovich, K. E. (1991). The causes and consequences of differences in reading fluency. Theory Into Practice, 30, 176-184.

Petros, T. V., Bentz, B., Hammes, K., & Zehr, D. H. (1990). The components of text that influence reading times and recall in skilled and less skilled college readers. Discourse Processes, 13, 387-400.

Rabren, K., Darch, C., & Eaves, R. C. (1999). The differential effects of two systemic reading comprehension approaches with student with learning disabilities. Journal of Learning Disabilities, 32, 36-47.

Shinn, M. R. (1989; Ed.). Curriculum-based measurement: Assessing special children. New York: Guilford.

Shinn, M. R. (2002). Best practices in using curriculum-based measurement in a problem solving model. In A. Thomas & J. Grimes (Eds.) Best practices in school psychology IV (pp. 671-698). Bethesda, MD: National Association of School Psychologists.

Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21, 360-406.

Stock, W. A., Winston, K. S., Behrens, J. T., & Harper-Marinick, M. (1989). The effects of performance expectation and question difficulty on text study time, response certitude, and correct responding. Bulleting of the Psychonomic Society, 27, 567-569.

Swanson, H. L. (1989). Phonological processes and other routes. Journal of Learning Disabilities, 22, 493-497.

Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1997). Prevention and remediation of severe reading disabilities: Keeping the end in mind. Scientific Studies of Reading, 1, 217-234.

Ysseldyke, J. E., & Christenson, S. (1987). The instructional environment scale; A comprehensive methodology for assessing an individual student's instruction. Austin, TX: PRO-ED.

Ysseldyke, J. E., & Christenson, S. (2002). Functional assessment of academic behavior. Longmont, CO: Sopris West.

Zabrucky, K. M., & Moore, D. (1999). Influence of text genre on adults' monitoring of understanding and recall. Educational Gerontology, 25, 691-710.

Zabrucky, K., & Ratner, H. H. (1992). Effects of passage type on comprehension monitoring and recall in good and poor readers. Journal of Reading Behavior, 24, 373-391.

Rebecca J. Lagrou, Central Michigan University; Matthew K. Burns and Elizabeth A. Mizerek, University of Minnesota; and Jill Mosack, Central Michigan University.

Correspondence concerning this article should be addressed to Dr. Matthew Burns, Associate Professor, School of Psychology Program, 350 Elliott Hall, University of Minnesota, 75 East River Road, Minneapolis, Minnesota 55455.

Table 1
Means, Standard Deviations, and Two-Way MANOVA Results,for
Fluency and Comprehension

                            Fluency       Comprehension

                   N       M       SD      M        SD       F

Low Reading
  Book             17     53.8    18.3    2.6       1.4
  Typed            13     33.7    13.8     .9       1.0
  Total            30     45.1    16.4    1.9       1.2
Average Reading
  Book             20     78.7    31.6    2.4       1.4
  Typed            24     85.9    25.7    3.0       1.7
  Total            44     82.6    28.4    2.7       1.6
High Reading
  Book             20    102.4    31.9    3.4       1.7
  Typed            25    120.2    30.3    4.0       1.7
  Total            45    112.3    31.1    3.7       1.7    7.9 *

* p < .001