Reading Skills of Students At-Risk for Academic Failure in High School
Linda O’Neill, M.S.
Department of Speech Pathology and Audiology
Illinois State University
Normal, Illinois
Heidi Harbers, Ph.D., CCC-SLP
Department of Speech Pathology and Audiology
Illinois State University
Normal, Illinois
Tom Parton, M.S., CCC-SLP
Normal West Community High School
Normal, Illinois
Correspondence to Heidi Harbers
hmharbe@ilstu.edu
Abstract
This study examined the reading skills of thirty-six 8th grade students identified by their teachers as “at-risk” for high school failure. Reading rate (fluency), oral reading and reading comprehension measures were collected. Results indicated that 6 of 10 measures were significantly (p <.01) correlated. However, reading rate (a measure of reading fluency) was not found to be associated with measures of comprehension in older children.
Introduction
Fluency has been described as effortless reading (Lipson & Lang, 1991), reading with automaticity and accuracy (Nathan & Stanovich, 1991) and rapid and accurate oral translation of text (Fuchs, Fuchs, Hosp, & Jenkins, 2001). Fluency has also been described as a goal, ingredient, instrument, prerequisite, and by-product of skilled reading (Lipson & Lang, 1991). The National Reading Panel (2000) defined fluency as the immediate result of word recognition competence. Historically, fluency was viewed along a continuum leading to reading competence, and was one of many tools used to teach reading. Today, fluency is viewed more as a product of reading, and the term “fluency” itself is used to define a characteristic of the reader, separate and distinct from comprehension. (Stayter & Allington, 1991).
There are many benefits to being a fluent reader. When fluent, the reader provides little attention to decoding tasks (Wolf & Katzir-Cohen, 2001), punctuation, or prosody (National Reading Panel, 2000). As a result, the fluent reader’s brain is free to use higher-level cognitive skills (Fuchs, et al., 2001). These higher-level cognitive skills include understanding, synthesizing, and thinking about the material (Nathan & Stanovich, 1991).
Reading fluency improves with practice. As a result of being fluent, readers read more, which in turn, increases fluency (Rasinski, 2000). Referred to as the “Matthew Effect” (Stanovich, 1985), fluency and the amount of time spent reading are cause and consequence of one another. Being a fluent reader allows one to manage text with more ease. Stanovich purported that superior readers are better decoders and as a result, understand text better. Further, disfluent readers have less comprehension due to insufficient decoding skills. For the disfluent reader, reading becomes laborious and learning from text is limited.
Because word recognition is a high capacity demand task (Nathan & Stanovich, 1991), readers who are just learning or who are disfluent use most of their cognitive resources for letter and word recognition (Wolf, 2001). Skilled readers automatically recognize words and occasionally use controlled effortful decoding strategies, whereas unskilled readers must always rely on controlled, effortful decoding strategies (National Reading Panel, 2000). Due to the time requirements for holding text in his/her memory, the reader may be unable to remember what he has read. (Mastropieri, Leinart, & Scruggs, 1999).
Meyer and Felton (1999) presented three explanations as to why a reader may have fluency difficulties; 1- The disfluent reader may recognize words slowly, 2- the disfluent reader may lack awareness of prosodic cues, or 3- the disfluent reader may fail to connect meaning with spelling. The effects of these three possibilities (respectively) include; 1- Slow recognition of words coupled with limited working memory leading to reading comprehension problems. 2- lack of awareness of the prosodic elements of reading, leading to poor understanding of the syntactic structure of the material, and 3- inability to connect orthography (spelling) and semantics (meaning), leading to poor recognition of patterns in words and sentences, preventing efficient reading.
Because fluency is related to reading ability, some researchers believe reading rate should be included as a diagnostic measure when assessing reading ability (Rasinski, 2000). When measured, fluency is usually defined as the number of correct words read per minute (Fuchs, et al. 2001). As might be anticipated, the “level” of text being read affects fluency. A child may be fluent when reading below his ability level, but may be disfluent when reading at a frustration level (Lipson & Lang, 1991).
Studies that examined the use of oral fluency rates to measure reading ability have used a variety of participants. These groups included normal developing children, children with learning or reading disabilities, and mixtures of children (Deno, Mirkin, & Chiang, 1982; Fuchs, Fuchs, & Maxwell, 1988; Jenkins & Jewell, 1993; Shinn, Knutson, Good, Tilly, & Collins, 1992). Findings suggested that oral reading rate is strongly correlated to several different valid formal tests for reading ability: The Stanford Diagnostic Reading Test, The Gates-MacGintie Reading Test, The Metropolitan Reading Test, The Iowa Test of Basic Skills, and The Stanford Achievement Test. First through eighth grade participants were included in the studies, and the validity of oral reading rate held. Correlations between oral reading rate and standardized measures were found to be stronger than the relation between the standardized measures and other more commonly used measures of comprehension including: word list reading, comprehension questions, passage recall, and cloze tests (Deno, et al., 1982; Fuchs, et al., 1988).
Many texts on reading comprehension have not included reading fluency as a measurement of comprehension. If one supports the view that fluency only relates to decoding, then assessing fluency would only provide information relative to decoding abilities. However, curriculum-based measurement research indicates fluency is indeed more than just decoding. There is a long history of using curriculum-based measurement of reading fluency which indicates that oral fluency rate is reliable and valid when used to measure reading skills, including comprehension (Shinn, et al., 1992).
Previous studies have found that reading rate is associated with reading comprehension in elementary school children (Deno, et al., 1982; Jenkins & Jewell, 1993; Shinn, et al., 1992) and junior high school students (Fuchs, et al., 1988). Very little is known about this association in high school. Moreover, most studies that examine reading fluency use students receiving general education or special education instruction. Little is known about the older at-risk student.
The purpose of this study was to examine the relationship between three informal reading measures used to assess fluency, comprehension, and word list reading abilities of eighth grade students at-risk for academic failure.
Method
Participants
Participants included 36 eighth grade students from a socioeconomically mixed middle school, located in Normal, Illinois. Teachers selected participants for the reading testing to determine if they qualified for a special high school program for students at-risk for academic failure. Of the 36 participants, 23 (64%) were girls and 13 (36%) were boys. All participants were from regular education classrooms and did not receive special education services.
Procedures
All participants completed three informal reading measures: oral reading rate (fluency), word list reading, and oral answering of comprehension questions. These measures were selected from The Basic Reading Inventory (Johns, 1997). Each word list and comprehension task had a previously established grade level.
To measure participants’ oral reading rates, each participant was individually taken to a quiet private room and asked to orally read a 100-word passage written at the eighth grade equivalent level. The examiner, a graduate student majoring in speech language pathology, used a stopwatch to time the task. When the participant finished reading the passage, the examiner recorded the lapsed time and calculated words read per minute. Errors were not subtracted. The participants’ reading rates were recorded in total number of words read per minute.
The participants then completed the word list reading measure. Each word list was comprised of 20 words. Each participant met with the school’s speech language pathologist in a private room and read the sixth grade list first, while the examiner noted any errors. When the participant finished reading the list, the examiner determined the number of words read correctly. To determine the reading ability level, the total number of words read correctly was used to assign one of the following reading levels:
95%-100% = independent reading level,Depending on the results of the reading of the sixth grade word list, easier or more difficult graded word lists were read, until the examiner determined reading lists that provided an instructional level and an independent level for each participant.
70% to 90% = instructional level, and
65% or below = frustration reading level.
Each participant silently read a passage at a sixth grade equivalent reading level. After the passage was read, the examiner orally asked 10 open-ended comprehension questions. The participant orally responded to each question. The questions related to fact, evaluation, inference and vocabulary. The scoring guide from The Basic Reading Inventory (Johns, 1997) was used to determine reading levels based on the number of questions answered correctly. Depending on the results of the first set of questions, easier or more difficult reading passages were read with the related comprehension questions answered until instructional and independent reading grade levels were determined and recorded.
Data Analysis
Five scores were collected for each participant. These included oral reading rate (i.e., total number of words [correct and incorrect] read per minute), instructional grade level for word list reading (i.e., grade level of word list read with 70-90% accuracy), independent grade level for word list reading (grade level of word list read with 95% accuracy). Additionally, instructional and independent grade level for reading comprehension scores were obtained by the number of comprehension questions accurately answered out of 10 (independent: 9-10; independent/instructional: 8-8.5; instructional: 7.5; instructional/frustration: 5.5-77; frustration: 5 or less). Means and standard deviations for each variable were obtained. Paired t-tests were performed between variables that had grade level scores. Pearson product correlation coefficients were computed for all variables.
Results
Table 1 displays the means and standard deviations of each reading measure. The participants demonstrated a wide range of reading rates (82 – 190 wpm). Paired t-tests between the word list and comprehension variables indicated a significant difference between reading comprehension at the independent level and reading comprehension at the instructional level (t = -12.27; p < .001). Reading comprehension at the instructional level was more than one grade level above reading comprehension at the independent level. There was also a significant difference found between word list reading at the independent level and word list reading at the instructional level (t = -10.55; p < .001). The instructional level was more than two grades above the independent level for these participants.
Table 1: Means and Standard Deviations of the Reading Measures
Note: Oral Comp – Ind = answering of oral comprehension questions at the independent level; Oral Comp – Inst = answering of oral comprehension questions at the instructional level; Word Rdg – Ind = word list reading at the independent level; Word Rdg – Inst = word list reading at the instructional level.
When comparing comprehension and fluency measures, a significant difference was not found between the participants’ performance of word list reading (at the independent and instructional level) and reading comprehension at the independent levels. However, a significant difference was noted between word list reading at the independent level and reading comprehension at the instructional level (t = -2.67; p < .01). Reading comprehension at the instructional level was .70 grade levels above the word list reading grade level at the independent level. Another significant difference was found between word list reading at the instructional level and reading comprehension at the independent level (t = 6.67; p < .001). Word list reading at the instructional level was almost three grade levels above reading comprehension at the independent level. Word list reading at the instructional level was also found to be significantly different than reading comprehension at the instructional level (t = 4.09; p < .001). Word reading at the instructional level was 1.5 grade levels above reading comprehension at the instructional level.
Pearson correlation coefficients for all measures are reported in Table 2. No relation was found between word list reading (independent or instructional levels) and reading comprehension at the independent level. In addition, no relation was found between reading comprehension (instructional or independent levels) and reading rate.
Results indicated that 6 of the 10 correlations were moderate to large (.42 - .92) and were statistically significant (p<.01). The strongest associations were found between reading comprehension at the independent level and reading comprehension at the instructional level, and also word list reading at the independent level with word list reading at the instructional level.
Additionally, there were five moderate correlations; word list reading at the independent level correlated to both reading comprehension at the instructional level and oral reading rate, word list reading at the instructional level was related to oral reading rate and reading comprehension at the instructional level, and a correlation was found between reading comprehension at the instructional level and oral reading rate.
Discussion
The purpose of this study was to examine the relationship between three informal reading measures (i.e., word list reading, oral answering of comprehension questions, and oral reading rate) used to assess the reading abilities of thirty-six 8th grade students considered to be at-risk for academic failure. All measures, except word list reading at the instructional level, were found to be below grade level.
In this study, oral reading rate (i.e., total number of words read per minute) was not related to comprehension but was related to word list reading. In addition comprehension at the instructional level was related to word list reading.
Results indicated that the mean fluency rate of the participants was 132 words per minute. Hasbrouck and Tindal (1992) determined from their sample of fifth graders, a reading rate between 100 to 151 words correct per minute. Assuming eighth graders should be faster readers than fifth graders, the participants in the current study performed below grade level.
Word list reading at the instructional level was the only measure for which the participant groups mean performance was at or above their grade. Participants performed from one half of one grade to nearly two grades below their grade level for other reading measures. These findings were primarily due to the participant selection criteria, as all participants were considered at risk for academic failure as measured by below grade level performance.
The findings from this study indicated that oral reading rate was not related to comprehension (as measured by the answering of oral comprehension questions.) This finding is in opposition to previous research findings. Fuchs et al. (1988) found a strong positive relationship between oral reading rate and reading comprehension questions. Their findings may be due to the use of a different participant group and different measurements for oral reading rate. The participant group studied by Fuchs et al. was comprised of students with learning, emotional or cognitive disabilities whereas the participants in the current study were at-risk for academic failure. In addition, oral reading rate, as recorded by Fuchs and colleagues, was determined by words read correctly per minute, whereas the current study followed the procedures from Johns (1997) and recorded reading rate as total words read per minute. Total words read per minute included words read correctly and incorrectly. The measure selected for this study resulted in higher reading rates. If the current study measured reading rate as words read correctly, the mean reading rate would have been lower which could have resulted in stronger correlations.
Another finding from this study indicated that oral reading rate and word list reading were related for students considered at-risk for academic failure. This finding supports previous research conducted by Deno et al. (1982) that found an association between oral reading rate of a passage and oral reading of word lists despite the fact that their participant pool was younger (grades 1-5) and different measures were used.
Although past studies examined the association between word reading and performance on comprehension subtests of standardized reading tests (Deno et al., 1982), no previous study examined the relation between word list reading and the informal measure of answering open-ended comprehension questions. The findings of the current study indicated that comprehension (as measured by the oral answering of comprehension questions at the instructional level) was related to word list reading at the independent and instructional levels.
It is unclear why only oral answering of comprehension questions at the instructional level was related to word list reading at the independent and instructional levels. The correlation coefficients between oral answering of comprehension questions at the independent level and word list reading at both levels were not significant (.07 and .21). This could have resulted from the wide range (1 - 9, mean of 6.2) of comprehension abilities at the independent level. In contrast, the range of grade levels found when participants orally answered comprehension questions at the instructional level was narrower, (4 - 10, and mean of 7.5).
Several limitations should be considered when generalizing the results of this study. Only one grade level of participants was used. These participants were identified by their teachers as being at-risk for high school failure due to their social and academic performance. In addition, the way in which oral reading rate was measured (total number of words read per minute) produced a higher reading rate than if oral reading rate was measured as the number of words read correctly per minute. Finally all the measures used were informal rather than standardized measures.
Many studies indicate that oral reading rate is a valid measure of reading ability for children in lower grades (Deno et al., 1982; Fuchs et al., 1988; Jenkins & Jewell, 1993; Shinn et al., 1992). More research is needed to determine whether informal reading measures are valid for children in eighth grade and beyond. There is a lack of normative data as well as information about fluency performance for students in the upper grade levels (i.e., beyond fifth grade). Future research also needs to address best practice when evaluating the reading skills of older students who are performing below, at, or above age level. Moreover, research is needed to address how fluency skills should be defined and measured in this population.
Because Fewster and MacMillan (2002) found that oral reading rate was a valid measure to use when making academic placements, it would be useful to examine the current study’s participants’ performances with their future academic grade achievements. In addition, studies are needed to determine what reading rate truly measures (i.e., decoding, comprehension, or both) for older students.
Despite the limitations of this study, the findings suggest word list reading is an important tool for assessment and should accompany other measures when assessing reading progress. Graded word list reading can provide insight to obstacles of unfamiliar, more complex word forms. This becomes more important as students progress with respect to grade level, and text becomes more difficult, less redundant and less predictable.
Additionally, when creating reading profiles of older children, decoding and comprehension abilities should be included. Although at-risk students may continue to display core deficits in effective decoding, special education services and support may not be indicated for at-risk students due to adequate skills in language comprehension. The student’s success with higher grade level text and curricula may ultimately suffer as a result of inefficient decoding. Decoding and automaticity continue to impact reading development into the upper grade levels and should be considered for continued skill development. When targeting fluency as an intervention goal, word list readings (timed and untimed) should be included.
References
Deno, S. L., Mirkin, P. K., & Chiang, B. (1982). Identifying valid measures of reading. Exceptional Children, 49, 36-45.
Fewster, S., & Macmillan, P. D. (2002). School-based evidence for the validity of curriculum-based measurement of reading and writing. Remedial and Special Education, 23, 149-156.
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.
Fuchs, L.S., Fuchs, D., & Maxwell, L. (1988). The validity of informal reading comprehension measures. Remedial and Special Education, 9, 20-28.
Hasbrouck, J. E., & Tindal, G. (1992, Spring). Curriculum-based oral reading: Fluency norms for students in grade 2 through 5. Teaching Exceptional Children, 41-44.
Jenkins, J. R., & Jewell, M. (1993). Examining the validity two measures for formative teaching: Reading aloud and maze. Exceptional Children, 59, 421-432.
Johns, J. L. (1997). Basic Reading Inventory. Dubuque, IA: Kendall/Hunt Publishing Company.
Lipson, M. Y., & Lang, L. B. (1991). Not as easy as it seems: Some unresolved questions about fluency. Theory into Practice, 30, 218-226.
Mastropieri, M., Leinart, A., & Scruggs, T. (1999). Strategies to increase reading fluency. Intervention in School and Clinic. 34, 278-283, 292.
Meyer, M. S., & Felton, R. H. (1999). Repeated reading to enhance fluency: Old approaches and new directions. Annals of Dyslexia, 49, 283-306.
Nathan, R. G., & Stanovich, K. E. (1991). The causes and consequences of differences in reading fluency. Theory into Practice, 30, 176-184.
National Reading Panel. (2000). Report of the national reading panel teaching children to read: An evidence-based assessment of the scientific research literature of reading and its implications for reading instruction. Washington D.C.: National Institute of Child Health and Human Development.
Rasinski, T. V. (2000). Speed does matter in reading. Reading Teacher, 54, 146-151.
Shinn, M. R., Knutson, N., Good, R. H., Tilly, W. D., & Collins, V. L. (1992). Curriculum-based measurement of oral reading fluency: A confirmatory analysis of its relation to reading. School Psychology Review, 21, 459-479.
Stanovich, K. E. (1985). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21, 360-407.
Stayter, F. Z., & Allington, R. L. (1991). Fluency and the understanding of texts. Theory into Practice, 30, 143-148.
Wolf, M. (2001). Dyslexia, Fluency, and the Brain. Timonium, MD: York Press.
Wolf, M., & Katzir-Cohen, T. (2001). Reading fluency and its intervention. Scientific Studies of Reading, 5, 211-239.