The effect of three-strategy remedial teaching for word recognition, comprehension, and fluency of a post primary reader with reading difficulty

(1)

The Effect of Three-strategy Remedial Teaching for Word Recognition, Comprehension, and Fluency of a Post Primary Reader with Reading Difficulty

Shelley Marion Thring B.Ed., University of Victoria, 1992 A Thesis Submitted in Partial Fulfillment of the

Requirements for the Degree of Master of Arts

in the Department of Educational Psychology and Leadership Studies

O Shelley Marion Thring, 2004

(2)

Supervisor: Dr. Lily Dyson

ABSTRACT

A single-case research design was used to test the effectiveness of combining three instructional strategies to improve the word recognition, reading comprehension, and reading fluency of one post-primary child with a history of reading difficulty. The strategies selected, phonological decoding, metacognitive thinking, and rehearsal were shown throughout the literature to be effective on their own and in various combinations for supporting children with reading difficulty. The ten-year old female subject was given instruction in using these strategies in an intensive 1 : 1 setting. The subject met with a Research Assistant for three, 1-1 54 hour sessions each week over a ten week period.

Prior to collecting baseline data, pre-testing was conducted using standardized instruments, the Wechsler Individual Achievement Test 11 (2002) and the Slosson Oral Reading Test (1 963) and a non-standardized instrument, the Classroom Reading Inventory (Silvaroli, 1982). Graded word lists and passages taken from the Diagnostic Reading Program, an inventory prepared for the Alberta Ministry of Education, Student Evaluation Branch (1 986) were used to assess the subject's skills in word recognition, comprehension, and fluency.

During Session #9, the intervention, a combination of phonological decoding, metacognitive thinking, and rehearsal strategies were taught prior to testing and data collection. At Session #15, the intervention was applied to comprehension. Although data was collected for fluency scores throughout the study, the intervention was not directly applied to this variable. Post testing revealed the subject's word recognition and

comprehension skills improved after the intervention was introduced, demonstrated by the generally positive trend on the multiple baseline whereas fluency scores actually dropped as she slowed her reading down to use the strategies for decoding unknown words. The quantitative data collected during post-testing with the WIAT I1 indicates an

increase in percentile and age level scores that the subject's scores in word recognition (13%, 1 year), comprehension (57%, 4.4 years), and pseudoword decoding (lo%, 1 year) and scores on the SORT demonstrate a grade level equivalent growth of 1.2 years.

(3)

Qualitative data in form of daily field notes by the Research Assistant signifl an overall positive change in the subject's view of herself as a reader. The strategies gave her more confidence, she abandoned her key strategy of guessing in favor of

metacognitive thinking and decoding strategies, and focused on meaning, all of which contributed to higher post-testing scores for word recognition and in particular,

comprehension as well as generally positive trends in baselines. After the study was completed, the subject was observed by her classroom teacher to have improved her performance and participation in the reading program and most significantly, was presenting a more positive and enthusiastic attitude toward reading.

It is acknowledged by the researcher that although provision of intensive 1 : 1

instruction in the use of reading strategies is not always feasible for schools attempting to provide quality instruction and programming for students with a wide range of abilities, these strategies, when applied, can significantly improve the word recognition and reading comprehension skills of a post-primary student with a history of reading dificulty although they caused a reduction in reading fluency scores.

(4)

.

Multiple-Baseline Design

b

.

Definitions of Terms Used in this Study

...

c

.

Multiple-Baseline Design Steps

...

d Three-Strategy Treatment

...

3.4 Three-Strategy Treatment

a

.

Phonological Decoding Strategy

...

.

b Metacognitive Strategy c

.

Rehearsal Strategy

...

3.5 Instrumentation

...

a

.

Slosson Oral Reading Test

...

3.7 Method of Analysis of Results

...

a

.

Sample Sessions - Pre-Treatment

4.5 Multiple Baseline Results (Figure 4.0).

5

.

Discussion

...

. .

6 . Bibliography

...

Appendices

...

7.1 Appendix 1 : Student consent form

...

7.2 Appendix 2: Parent consent form

...

7.3 Appendix 3: School Principal consent form

. .. .

.. . .. . .

.. ..

.

.. .. . . . .. . . .. . .

.

..

.

.. ..

63

(8)

ACKNOWLEDGEMENTS

The completion of this thesis and personal goal for this middle-aged educator, would not have been possible without the generous support of three very special people, my supervisor, Dr. Lily Dyson, Kerry Heichert, my Research Assistant, and my partner and friend, Rob Plaskett. Dr. Dyson's expertise in the field of learning disabilities, her encouragement through some challenging times, and her belief in my abilities were invaluable to me as I negotiated my way through the process of writing my thesis. Ken-y's professionalism, hard work, and dedication to the research made the entire pursuit possible. And finally, to Rob Plaskett, for his knowledge of how children learn, his generous gifts of patience and guidance, and especially for his love; it guided me through to the realization of this dream.

(9)

Introduction I

CHAPTER 1 INTRODUCTION

Flor Jenny, the school year would be similar to previous ones. She had spent an extra year in the primary program because her literacy skills were below grade level and now she was saying good-bye to yet another group of friends as she remained behind in the split grade that had the younger children. Her peer group changed almost yearly and she still didn't have a best friend! Her parents and teachers reassured her that she was indeed a Grade 6 student but if that was the case, then why was she made to do the same activities as the Grade 5s? Even the Grade 5 work was too hard for her, especially the reading! And, once again she was spending time in the resource room with Mrs. Smith. She couldn't remember when she hadn't spent time out of the class anld in the resource room! Homework was another challenge she faced; it took Jenny a very long time to get her homework assignments done because she had such a difficult time with the reading part, even the directions! She hated reading and now that she was in Grade 6, she had to read all {of the time. The science and social studies textbooks were especially hard! If it was too hard now, she couldn't imagine how she would deal with the reading students had to do in high school! Jenny was looking at a report that her mom left on the fridge. It said that she was reading at an early grade two level! Would she ever be like her friends and be able to read the cool books they did'? She was so discouraged she wanted to give up.

This story describes a real life situation (the name has been changed) for a child known to the researcher. Sadly, Jenny is not alone in her struggle with reading. She is one of many older elementary-age students who have slipped through the cracks of the early intervention programs of primary grades and is destined to enter middle school reading at a level well below that of what is considered average for children in her age group. Jenny is becoming discouraged because no matter how hard she works, the outcome seems to be the same and she cannot read the same materials her classmates read, even those who are younger than she. Despite the years of support

(10)

Introduction 2 Jenny received in the classroom and resource room, very little progress in reading was realized. Numerous approaches were tried and costly programs purchased to address the needs of Jenny and others who struggled to read during their primary years. A number of different school personn.el worked with these children to try to find the magic solution that would turn them into readers. It becomes clear as the literature is reviewed that this search for a magic solution is still taking place.

Statement of the Problem

Tlhe problem of this study is expressed by the following question: What effect would intensive one-on-one instruction in a three-strategy reading intervention have on improving the phonological decoding, metacognitive thinking, and rehearsal of one post-primary child with a history of reading difficulty? It is hypothesized that it would result in improved outcomes for word rescognition, comprehension, and fluency.

Given that there has been little measurable progress made in addressing Jenny's difficulty with rea.ding, due perhaps to a lack of effective and consistent intervention approaches, a need has arisen to improve her reading through the use of an effective, systematic program of interven.tion. Specifically, Jenny, like many other children who have difficulty with reading, lacks effective strategies for reading, demonstrated by research to be critical to reading success. The strategies include: phonological decoding, metacognitive thinking strategies, and rehearsal strategies (Foorrnan, Francis, Fletcher, Schatschneider, & Mehta, 1998; Mastropieri & Scruggs, 1997; Swanson, 2000; Swanson & Alexander, 1997; Torgesen, Alexander, Wagner, Rashotte, Voeller, & Conway, 2001; Wong, 1986). Although Mastropieri and Scruggs, Swanson and Alexandler, and Torgesen et a1 suggest that reading difficulty be addressed through instruction which combines more than one reading strategy, few have actually done so. This study's target population is post-primary children with learning disabilities that experience difficulty primarily in the area of reading.

Jeinny has learning disabilities, making reading very challenging for her. In developing the intervention to be used in this study, it is necessary to have a greater understanding of what a

(11)

Introduction 3 learning disability is and how it affects the reading process. The term learning disabilities itself emerged out of a need to identify children who did not match the existing definitions for exceptionality (Reid & Hresko, 1981). In much of the research, this population of children is differentiated by their substantial difficulties in learning despite average or above-average aptitude (MacMillan & Gresham, 1998).

There are other reputed characteristics that appear throughout the literature as well. Some of the other descriptors applied to learning disabilities are their intrinsic nature and their location in basic psychological processing. In addition, the discrepancy between intellectual aptitude and actual achievement is generally seen to be in one or more of the following areas of psychological functioning: basic skill in reading, reading and listening comprehension, oral and written

express ion, mathematics calculations, or mathematics reasoning (Sternberg & Grigorenko, 2001). Other behavioral correlates such as distractibility, hyperactivity, andlor lack of impulse control as compared to peers are often included in the definition (Reid & Hresko, 1981).

Excluded from the most commonly held definitions of learning disabilities are individuals with mild mental retardation (MMR), emotional or behavior disorder (EBD), and learning problems attributable to sensory or motor impairment. The Individuals with Disabilities Education Act or IDEA in the United States includes the above descriptions in its definition along with conditions such as perceptual disabilities, brain injury, minimal brain dysfunction, dyslexia, and developmental aphasia (Kavale & Forness, 2000).

The definition of learning disability as used in the present study is that of Lerner (2003). To paraphrase her definition, it is a disorder in one or more of the basic processes required for processiing spoken or written language. It can include problems in listening, thinking, speaking, reading, writing, spelling or an ability to do math. The seven areas where difficulties are

encountered are: ( I ) receptive language, (2) expressive language, (3) basic reading skill, (4) reading comprehension, (5) written expression, (6) mathematics calculations, and (7)

mathematics reasoning. Those exempted from the definition are children whose difficulty is due to: emotional disturbance, or to environmental, cultural or economic disadvantage (p. G-5).

(12)

Introduction 4 In terms of prevalence of learning disabilities among school age children, the U.S.

Department of Education in 1997 stated that the number of children with learning disabilities has grown to represent over 50% of all students with disabilities and 5% of the school population overall (Kavale & Forness, 2000). According to the Canadian Council on Social Development (200 I), 4.1 % of school age children were diagnosed with learning disabilities in 1996, compared with 3.:7% in 1994. The actual number of children represented by these statistics was 97,000 in

1996, as compared to 85,500 in 1994, an increase of 12,000 children.

Specifically regarding the issue of how children learn to read, Anderson (1980) indicates two models that describe cognitive processes as taking place in either a "top-down" or "bottom- up" fashion. In the former model, the reader constructs meaning or makes sense of what is on the page based on higher-level general knowledge. These processes include semantic memory, metacognition, and verbal information processing (Mastropieri & Scruggs, 1997). In contrast, Anderson's description of the "bottom-up" model suggests the reader uses critical features such as letters, words, and pictures to process what is on the printed page. Anderson and Swanson and Alexander (1 997) contend that these processes, used simultaneously are the most efficient way to read.

Information processing and reading are seen as interacting with several of these processes, both lovver (as defined by the bottom-up model) and higher (as defined by the top-down model). Once an. understanding of this interaction is reached, the specific processing deficits of readers with learning disabilities can be identified and treated (Swanson & Alexander, 1997). There is general agreement in the literature that reading problems can be attributed to deficits in higher and lowler processing abilities, but as Swanson and Alexander point out, there is a tendency for the emp:hasis to be on assessment and intervention with one or the other and not both. There is a need to :Find effective remediation strategies for children with reading disabilities attributed to higher a-nd/or lower processing difficulties.

According to Foorman, Fletcher, Francis and Schatschneider (l998), one in five children will have difficulty learning to read. Furthermore, if identification of children at-risk for learning

(13)

Introduction 5 to read doesn't take place very early, they could leave primary grades reading well below the levels considered average for their age and grade. Felton and Pepper (1 995) cite longitudinal studies that indicate that failure to acquire word recognition skills in early elementary grades often results in a long history of struggle with reading. One study in particular, conducted by Felton and Wood (1992), indicated that students in the third grade who had poor skills in word identification did not significantly improve their basic reading skills as they advanced to the fifth and eiglhth grades. Even with the best methods currently available, Torgesen (2000) suggests that 2%

- 6% of children would still have inadequate word reading skills in the primary grades. These

children, termed "treatment resisters" by Torgesen (p.55) match the criteria for participants in the current study. This population of students will undoubtedly continue to experience difficulty with reading in later grades unless effective remedial programs are put in place to support them. There is growing realization that we must intervene early with struggling readers so that they leave school as literate young adults.

In the past decade, researchers and educators have focused more attention on the goal of having children read well by their middle years of elementary school. According to Torgesen (2000), this motivation arises from the understanding that early reading failure has a negative impact on overall cognitive development. As well, he argues, children who enter their adult years with low levels of literacy are at a disadvantage in a society that is placing increasing emphasis on effective reading skills in the workplace. Adult outcomes for reading disabilities are important long-ten- consequences to consider and will hopefully inspire us to work hard with those

students who slip through the cracks and are still reading at levels significantly below what would be considered average when they begin middle elementary grades. As well, there are important affective consequences for children who experience difficulty with reading and those have even more emotional impact in the upper elementary and high school grades.

By middle elementary levels, there is less likely to be direct instruction of reading as it is often assumed by teachers that most children will be reading by then. Swanson (2000) identifies the combination of direct and explicit instruction in the use of strategies as the best model for

(14)

Introduction 6 effectively providing more support for students with learning disabilities when compared to other instructional models. The suggestion that both lower- and higher-order processing skills be viewed as important when designing treatments is also supported in the study done by Swanson and A1e:xander (1997). Their study examined how cognitive processes interrelate to predict perfom~ances in word recognition and reading comprehension of readers with learning disabilities. They discovered that these children had difficulty accessing and coordinating phonological, orthographic, semantic, metacognitive, and working memory processes.

If children with reading disabilities have some level of weakness in all of these areas, it makes sense to address this by providing remedial instruction aimed at improving performances in these processing areas. Much of the literature, however, focuses primarily on one processing area, ph.onologica1 awareness and decoding being the most popular (Bowey, Cain, & Ryan,

1992; B;ruck, 1992; Ehri, Nunes, Willows, Schuster, Yaghoub-Zadeh, & Shanahan, 2001; Gang

& Siegel, 2002; Spector, 1995). One of the distinct differences between unskilled and proficient readers is the lack of awareness that poor readers have of the reading process itself, the purpose, goals, and the strategies that constitute metacognition (Meloth, 1990). Both Meloth and Wong (1986) discuss the possibilities of teaching metacognitive strategies to children with reading disabilities.

As mentioned earlier, children with learning disabilities have a difficult time accessing working memory processes. Performance differences between children with and without learning disabilities can be measured in part by their limitations in working memory (Swanson, 1989). Verbal tasks are particularly difficult to remember for children with learning disabilities (O'Shaughnessy & Swanson, 1998). These authors maintain that memory deficits do not improve with age. In their study, when shown a rehearsal strategy, children with learning disabilities did not use it spontaneously. This implies that children with learning difficulty need direct instruction in using effective strategies for practice. The present study will address reading difficulty specifically and the use of strategy instruction as a means of providing remedial

(15)

Introduction 7 G h e n the information gleaned from the literature on learning disabilities and disabilities specific: to reading, it is clear that any study that attempts to address the challenges faced by post- primary children is a relevant one. There is also considerable support for the concept that reading disabilities originate in the child's ability to process information at various levels of

sophistication. Word recognition is inarguably fimdamental to success in reading (Gang &

Siegel, 2002; Spector, 1995) and comprehension and fluency cannot be achieved until this area of processing is addressed. As well, there is enough support for the premise that children with learning difficulties need sufficient rehearsal of newly learned information to master it

(O'Shaughnessy & Swanson, 1998). For many children who have difficulty with reading, however, the remedial programs designed for them may not be characterized by a consistent program or approach and sufficient time to rehearse the skills being taught. Such was the case for Jenny; there were too many, often contradictory approaches or programs used in an attempt to teach her to read. It might be assumed that this would prove to be overwhelming for her. Sufficient rehearsal of any one or combination of strategies is not likely to have taken place.

A~dditionally, with older children like Jenny struggling in reading there is an assumption made by teachers at this level that by the time they reach intermediate and middle school grades, fundamental reading skills have been mastered and therefore, reading lessons do not include direct instruction of reading strategies. It would seem as well, that costly programs that may or may not be available to cash-strapped schools are not the answer but rather, that specific reading strategies, proven in studies to work for children with reading disabilities, be combined and taught to children like Jenny in intensive programs involving one-to-one instruction; a shorter term investment of time and money but a long term pay-off for students like Jenny.

(16)

Introduction 8

Purpose of the Study

The purpose of this study is to test the possibility that a combination of strategies for phonological decoding, metacognitive thinking, and rehearsal will help to eliminate the deficits in the information processing of one child with difficulty in reading. Specifically, the study will test the effectiveness of combining these three strategies to improve the word recognition, comprehension, and fluency of the reader. These strategies have been shown throughout the literature to be effective on their own and in various combinations for supporting children with reading difficulty. This study will contribute to the body of research available on reading difficulty and provide an alternative to the single strategy approaches frequently cited. The intervention tested in this study could improve a child's reading, leading to greater progress academically, and perhaps have a subsequent positive effect on his or her social development as well. Learning Assistance and classroom teachers may find this study useful for guiding their reading instruction with this population of students. Parents can be trained in the use of the strategies as a means of supporting their children's reading at home.

Summa y

In summary, an issue arising from the literature on learning disabilities is that children with reading difficulties appear to lack a "tool box" of reading strategies that they can apply when confronted with an unfamiliar word. If these children were to be taught how to use the strategies selected for this study, phonological decoding, metacognitive thinking, and rehearsal as

strategies for reading, their reading might improve and consequently, the image they have of themselves as learners. It is important for teachers to understand how they can intervene to promote and remediate reading difficulty by teaching the student with learning disabilities how to use such strategies. As is evidenced by Jenny's experience, having difficulty with reading has a negative effect on many aspects of the child's school experience. Addressing reading difficulty in students with learning disabilities then is critical to their academic and perhaps as

(17)

Introduction 9 In Chapter 2, information processing theory, the framework for this study will be reviewed with particular regard to phonological processing, metacognitive processing, and the effect of rehearsal on short-term memory as they relate to reading difficulty and remediation.

(18)

Literature Review 10 CHAPTER 2

LITERATURE REVIEW

In Chapter 2, the literature will be analyzed more extensively in an effort to address the research question: What effect would intensive one-to-one instruction in a 3-strategy reading intervention have on improving the Phonological Decoding, Metacognitive Thinking, and

Rehearsal of one post-primary child with a history of reading difficulty? It is hypothesized that it would result in improved outcomes for word recognition, comprehension, and fluency. The review i~tself will be divided into the following sections: (a) The History of Constructivist Information Processing, (b) Information Processing and Reading, (c) Word Recognition's Role in the Reading Process, (d) Short-term MemoryIWorking Memory and Word Recognition, (e) Rehearsal Strategies and Word Recognition, (f) Metacognition and Word Recognition, (g) Reading Comprehension and Fluency, and Multi-strategy Approaches in an Information Processling Framework of Remediation.

The His tory of Constructivist Information Processing Theory

During reading, simultaneous use of bottom-up information processes, i.e., distinguishing the critical features of letters, words and pictures (Anderson, 1980) and top-down processes including semantic memory, metacognition, and verbal information processing (Mastropieri &

Scruggs, 1997) is viewed by many theorists to be the most efficient way to read (Anderson; Swanson & Alexander, 1997). To better understand the growing popularity of this view, it is helpful to look at the historical development of the constructivist information processing theory.

The metaphor of human beings as processors of information has been the dominant view in educational psychology since the rebirth of cognitive psychology in the late 1950s (Mayer,

1996). It is the second of three metaphors described by Mayer in his examination of the role of information-processing theories in the history of educational psychology. The first metaphor,

(19)

Literature Review 1 1

known as "response strengthening" (Mayer, p. 152), a very simplistic representation of

learning popular in the first half of the twentieth century, was based predominantly on laboratory research with animals. In this theory, the teacher's role was to give out rewards and punishments for good and bad performance, respectively, while the learner's role was to receive them. The instructional model involved was one that focuses on drill and practice of basic skills.

Information processing, the second metaphor, grew popular in the late 1950s and continued to gain popularity through to the 1970s. Mayer (1996) likens this metaphor to an historic bridge, acting to connect the associationist metaphor prevalent during the first half of the century to the constructivist metaphor that would become dominant in educational psychology up to the present time. The information-processing model was applied to research with humans perfoming artificial tasks. In this metaphor, the teacher would dispense the information using textboolks and lecture-style teaching and the students would receive and theoretically, learn the information in this way. Mayer points out that coinciding with advances in computer technology, information processing acquired a computer-like analogy to explain how the human mind

processes information. Samuels (1987) described it well, making the learner analogous to the computer operator, the hardware to the physiological factors affecting learning, and the software to the sk.ills and strategies necessary to learn. A literal interpretation of this theory is that

cognitive processing is a procedure by which information is inputted and operated on, then outputteld as new information. In short, it is a mental computation (Mayer, p. 156).

The third and most modern view of learning and the model used in the development of this research study is the constructivist interpretation of information processing as described by Mayer. A key difference between constmctivist theory and the more traditional information processing metaphor is the role of memory representations. According to Mayer, whereas the latter infixmation processing model views memory as representing information, the

construc1.ivist model sees them as representing knowledge. In Mayer's comparison, knowledge, unlike information, can be schematic or planned, based on a general understanding of the world around US, or influenced by others. Another change in focus and key difference between the two

(20)

Literature Review 12 emphases is the contribution the constructivists made to an important paradigm shift in

psychology where a more inclusive view of the learner would evolve; the learner as an active participant in his or her learning, cognitively and emotionally. In short, constructivists would judge the human-as-computer analogy of the previous information processing theory as an

incomplete view of the learning process.

hlayer (1996) describes how findings in studies on learning in the 1980s and 1990s in psychollogy led to the incorporation of this more inclusive view of the learner. When researchers began to consider mental representations and the nature of mental processing, they were obliged to study cognition in educational settings. Thus the constructivist model of information-

processing theory was born.

Learning would now be viewed as "active, planned and efforthl" (Mayer, 1996, p. 158). In this sense, the constructivist view of information processing is a revision of the previous

information processing theory, building on its basic premises and involving the learner more in the process. In practice, this model perceives learners as constructing knowledge as they try to make sense of the information they are given. Mayer describes the teachers' role as a guide to their exlploration of academic tasks through instructional methods that include discussion, guided discoveiry and supervision of assigned academic tasks. The constructivist view of information processing is the one dominating the field of educational psychology today and can be

success~ully applied to the development of reading skills. The information-processing model as it is applied to reading acquisition will be examined next. Throughout the remainder of this review, when information processing is referred to, it will be associated with that of the constructivist metaphor of the past two decades illustrated by Mayer.

Constructivist Information Processing and Reading

When adopting a constructivist information processing approach to learning, the learner is viewed as the most important player in the teaching-learning situation (Reid & Hresko, 198 1). Of all the many skills learned, reading is thought to involve the most complex processing (LaBerge

(21)

Literature Review 13 (1997) maintain that all learners, disabled and nondisabled, progress through several

stages of higher (working memory, metacognition) and lower-order cognitive processes (phonological processing and rehearsal) in order to activate meaning of the written word (LaBerge & Samuels, 1974; Swanson, 1987b). These stages of cognition as described by Swanson (1987b) include encoding, organizing, storing, retrieving, comparing and generating information.

When children experience difficulty at any one or combination of these cognitive stages, their potential for success is seriously jeopardized. If, as Swanson and Alexander (1997) suggest, this difficulty in reading is caused by a deficit in information processing, how are the deficits identified and where do they originate? I return to information processing theory and the view of the human mind as a computer. Although it has been found to be incomplete within the context of the constructivist model of information processing, it is a usekl tool for examining the

specifics of cognitive processing deficits in learning. I once again examine an earlier description of information processing as Samuels (1987) presents it.

Siimuels (1987) uses the computer analogy to explain the sources of deficient processing: ( I ) the individual could have a problem that is physiological in nature; this would trace it back to the hardware system of the mind, or (2) the difficulty could be in the learner's failure to learn the skills and strategies necessary to learn; this would trace it to the software system of the mind. Swansoin (1987a) also argues that the problems could coexist in both the hardware and software of the brain. According to Swanson, regardless where the problem lies, any deficits in processing will have a negative impact on the child's ability to learn new skills and information and

remedial instruction will be required.

The remedial instruction Swanson (1987b) describes covers a broad continuum. At one end, the teacher acts as a model and questions the child's strategic thinking, a more top-down approach to instruction. The teacher's involvement lessens as the learner internalizes these self- regulation skills. At the other end of the instructional continuum, Swanson describes the focus as being on the processing skills and subskills that must be performed automatically and therefore,

(22)

Literature Review 14 practiced until mastered, a more bottom-up approach to learning. In a subsequent study

done b y Swanson and Alexander (1 997), the researchers recommend that both ends of the continuum be combined for instruction creating a learning environment that focuses on

facilitating the child's use of metacognitive (thinking about your thinking) strategies at a more automatic level. This picture of information processing theory resembles the constructivist view of the learner who is guided by the teacher in performing complex tasks through a process of determining which strategies are appropriate, orchestrating them, and then monitoring their effectiveness in solving the problem (Mayer, 1996).

In summary, the constructivist information-processing model combines all of the

processing components of the human mind. It implies that deficits in processing will manifest in reading difficulty. Deficits in memory will have negative consequences for a child's potential to master reading. Strategies for compensating for deficits in processing are central to intervention for reading difficulty. Mastropieri and Scruggs (1 997) support the teaching of reading strategies to childiren who have deficits in information processing. They suggest that the best overall reading programs combine strategies for compensation in both bottom-up processes of reading such as phonological awareness and fluency with top-down processes such as semantic memory, metacognition and verbal information processing. There is considerable potential for information processing to provide a theoretical framework for the cognitive task performance of children with learning disabilities (Swanson, 1987a).

Word Recognition's Role in the Reading Process

To explain how phonological processing fits in the information-processing model, the diagram (p. 16) provided by Bos and Vaughn (1991) will be used (see Figure 2.0). In word recognitmon, the word itself is the stimulus from the environment. It is taken in first by the senses. For the most part, this is done through visual and auditory processing. They remain in sensory storage for approximately a second, just long enough to be attended to and perceived by the brain. If the word is not recognized, i.e., retrieved fiom long-term memory, the next step is to perceive the knowledge based on the critical features of the stimulus and the meaningfulness of

(23)

Literature Review 15 the context in which it is presented. At this point, the learner will try to apply

phonological decoding skills. This analysis of the features of the word describes the bottom-up process of perceiving the word in terms of the smallest units of information it provides.

Phonological processing is characteristic of this level of processing. Proficient readers also use the top-down processes of context or background knowledge to guide these perceptions so that they can interpret the information accurately.

The majority of readers with persistent deficits in reading have difficulty at the word recognition level and the greatest discrepancy between good and poor readers is in their ability to Figure 2.0

An Information Processing Model of Learning

An Information Processing Modelof learning

-

_-

--L --c - - Unobservable Thinkmg to self lnlormat~on Observable Lost by Response, Fading and lnefleclive Motor Slrategier .Speech Feedback ₁

Source:Bos, C. S. & Vaughn, S. (1991). Strategies for teaching students with learning arzd behavior problems (2"d Edition). Massachusetts: Allyn & Bacon

decode (Spector, 1995). In a meta-analysis of phonemic awareness instruction, Spector argues that poor readers have difficulty in using the sounds of language when processing oral and

(24)

Literature Review 16 written information. She concludes that phonemic awareness training is key to

correcting poor decoding skills but acknowledges that mastery of this area of processing does not necessarily generate improved vocabulary, comprehension or sight word recognition. There are no suggestions given as to the inclusion of other strategies that might augment a program in phonemic awareness and create a more positive correlation between decoding and other reading skills.

Stanovich (1988) attests that both children with dyslexia and those with less severe reading difficulty exhibit the same deficits in phonological decoding but for the dyslexic child, the deficit may be more pronounced. As with Spector's (1995) view, this focus on phonology tends to ignore other deficits that occur at higher levels of processing such as metacognition or working memory. In that sense, Spector's view and the phonological-core variable-deficit model that Stanovich supports seem incomplete.

Bowey, Cain and Ryan (1 992) suggest that children with an impaired understanding of the alphabetic principle rely more heavily on acquiring a sight word vocabulary. These authors point out that accessing words through a visual process is slower and less efficient than phonological decoding. The results of their study suggest more delay in the acquisition of phonological analysis and decoding skills for the less skilled older students. These findings are surprising considering that more direct instruction in phonetic decoding would be taking place in earlier grades. Bowey and associates (1 992) conclude that the inferior word reading ability of less skilled older readers as compared to younger readers is reflected in the significant differences they found in the phonological analysis and decoding skills of the respective groups.

A study conducted by Foorrnan, Francis, Fletcher, Schatschneider, and Mehta (1 998) demonstrated the effectiveness of direct instruction in the alphabetic principle as a means of improving word recognition skills. In their study, three groups of Grade 1 and 2 children with special education designations were selected and divided into three groups. Each group received instruction in phonological processing skills. One group received explicit or direct instruction in matching decodable text. According to Foorman et a1 (1998), an instructor provides explicit

(25)

Literature Review 17 instruction when he or she clearly states what is to be taught and explains what is

required of the student. A second group in the study received both implicit and explicit

instruction in connected text and the third received implicit instruction in connected text. Implicit instruction involves the instructor pointing out alphabetic principles in context and allowing the child to construct his or her own understanding of those principles rather than breaking the task down and teaching them directly. The results of the study indicated that children at risk for reading failure in the primary grades benefited more from direct instruction in the phonological processing skills. The word reading skill of the group receiving direct instruction in the

alphabetic principal approached national averages in decoding and comprehension as measured by percentile scores, the 43rd percentile and 45th percentiles respectively. It is important to note, however, that group differences were less robust in comprehension and little variation occurred in scores for spelling and vocabulary.

Torgesen, Alexander, Wagner, Rashotte, Voeller, and Conway (200 1) integrated several types o I? instruction in more than one area. In this study, two different programs that

incorporated explicit instruction in phonemic awareness, phonemic decoding and sight word recognition were used to assess the immediate and long-term outcomes of intensive instructional approaches characterized by one-to-one instruction in 50 minute sessions twice weekly for eight weeks. Sight word recognition as well as the added elements of error-correction routines to establish which word identification strategies worked best (a higher-order, metacognitive process), and opportunities to practice (rehearsal) contributed to better results. Both programs reportedly produced improvement in generalized reading skills and half of the children attained average-level reading skills when reassessed after the follow-up period. Although some children still demonstrated severe impairment in reading, they too made considerable progress. Prior to the study, these children had not been able to acquire sufficient skills through general and special education classrooms. Torgesen et al. make note of the power that can be created through

(26)

Literature Review 18 acquisition (through intensive rehearsal), scaffolding activities and error-correction

routinels (metacognition).

Torgesen et al. (200 1) further reiterated the importance of sufficient time to practice, a key component for improving working memory skills. Although they had not shown a strong

correlation between direct teaching of comprehension and their intervention, Torgesen et al. point out that in other research done by Mastropieri and Scruggs (1 997) and Swanson (1 999) gains are shown in comprehension once word recognition skills improve (these studies will be examined in more detail later). In the case of the group who, despite their marked improvement in wordl reading skill, still fell short of agelgrade expectancy levels, these authors maintain that additional instruction in effective comprehension strategies would make a measurable difference. They were committed to including such instruction in their next study on intensive intervention (P-56).

In a meta analysis done by Ehri, Nunes, Willows, Schuster, Yaghoub-Zadeh, and Shanahan (200 l), the authors cautioned that phonemic awareness instruction, while an important

component to helping children learn to read, is not a "magic bullet" (p. 279), to be taught in isolation and without connection to other types of reading instruction. Rather, the authors consider it to be a means to an end. They also question the usefulness of phonemic awareness instruction for older children while at the same time stressing the importance of continued instruction for children with disabilities who may still have poor understanding and skills in this processing area.

Bmck's (1992) study compared the phonological awareness skills of dyslexic children and adults with childhood diagnosis of dyslexia against good readers at various age levels. She notes that phonological deficits are often seen as being the most dramatic in individuals with dyslexia, a statement that receives support from Stanovich (1988) in his discussion on the strong

similarities between individuals with dyslexia and those with other categories of reading disability. The results of Bruck's study indicate that even as reading skills increase over time, the development of phonemic awareness appears to be arrested due to failure in using

(27)

Literature Review 19 orthographic information when making decoding judgments. This supports the

contention of this study that time is of the essence; if children emerge from primary grades with weak information processing skills for acquiring reading, intensive instruction must take place in order to realize any significant improvement. Further research needs to be done in this regard with post-primary age groups.

As Bruck (1 992) points out, the focus tends to shift away from phonological and

orthographic codes as children age and enter adulthood. She is concerned that the inability to use these skills successfully may inhibit the development of more fluent word recognition skills throughout the lifespan. The results of these three studies appear to support the contention that learners with reading disabilities need instruction in phonological awareness and the use of strategies for decoding.

Short-Term Memory/Working Memory and Word Recognition

Although there is a great deal of evidence to demonstrate that children with reading disabilities and dyslexia have difficulty processing phonological information as indicated in the literature presented here, there is cause for considering other cognitive processes related to reading as well (Swanson & Alexander, 1997). Once a word has been inputted through the senses and is perceived, it can move into short-term or working memory. Working memory is activated memory because it represents the information that is easily accessed (Bos & Vaughn,

199 1). Swanson and Alexander describe working memory as a processing resource of limited capacity relevant to the reading process. Tasks measuring working memory are those that require the person to hold information in his or her mind for a short time while simultaneously carrying out other operations. Applying this to word recognition, the child is presented a word unknown to him or her and is told what it is. If there is no rehearsal of the word, it will fade from short- term memory. If on the other hand, sufficient rehearsal takes place, it may be stored in long-term memory and if it is worked with (more rehearsal), retrieval will be possible.

Children with reading disabilities are less able to retain information in memory while simultaneously processing the same or other information (Seigel & Ryan, 1989; Swanson, 1989).

(28)

Literature Review 20 These pervasive memory deficits are correlates of reading deficiencies

(O'Shaughnessy & Swanson, 1998). These authors note that children with learning disabilities and reading problems consistently perform poorly on digit-span tests and other measures of short-term memory. In his study of children with reading and math disabilities, Swanson (1993) hypothesized that children with problems primarily in the area of reading would have difficulties with language-related working memory tasks and children whose primary problems were in arithmetic would demonstrate difficulties in visual-spatial working memory tasks. He discovered that there were generalized memory deficits in both reading and math performance groups, which he attributed to processing problems in the executive system. The executive system is responsible for metacognitive thinking. It is the processing system where learners coordinate, monitor and determine which strategies they should employ for any given task (Bos & Vaughn,

199 1).

Memory deficits have serious implications for remedial instruction. The best conceivable strategies will fail if sufficient rehearsal is not provided, creating a greater likelihood that the information being learned can be moved from short-term to long-term memory storage.

Rehearsal Strategies & Word Recognition

Torgesen, Alexander, Wagner, Rashotte, Voeller, and Conway (2001) found a positive correlation between the use of intensive instruction and measurable progress in word recognition skills. This further validates the importance of sufficient practice for children with reading difficulty. O'Shaughnessy and Swanson (1998) indicated that children with reading disabilities do not apply strategies spontaneously and therefore, direct instruction and practice would be necessary to correct this problem. If that type of instruction has not been a primary component of the reading instruction that children with reading disabilities experience in their earliest years of acquiring reading, it makes sense that they would continue to struggle as they enter upper elementary and middle school grades. Although rehearsal strategies would be considered a lower-order process in the information-processing model, they are discussed here because of their important role in memory storage. Given the involvement of both phonological and

(29)

Literature Review 2 1 memory processes in word recognition, it would appear important that strategies for

making short-term or working memory more efficient during reading be taught to children with reading disabilities. Rehearsal is a way to facilitate memory, especially short-term memory (Bos

& Vaughn, 1991). Some commonly used rehearsal strategies are repeatedly saying aloud what is to be remembered. Other senses can be incorporated into the rehearsal process as well. These approaches, called multisensory, combine various sensory inputs for the purposes of

remembering the new information. With respect to word recognition, using a variety of sensory input, i.e., visual, auditory, kinesthetic, and tactile, is characteristic of programs such as Spalding Method, Orton-Gillingham method, and the Slingerland method (Gang & Seigel, 2002).

Automaticity through rehearsal (drill and repetition) may be necessary for children with reading disabilities, particularly those with dyslexia, to ensure encoding of sounds and sound combinations for the purposes of phonological decoding (Gang & Siegel, 2002; Torgesen,

Alexander, Wagner, Rashotte, Voeller, & Conway, 2001). Swanson (1982) is also a proponent of extensive rehearsal in the interest of having children with reading disabilities become more automatic in reading subskills such as phonological processing and memory. He describes automatization as a repetition of "routine experiences" (p. 8 1). Because short-term memory capacity is limited (in children without disabilities as well), it is important to maintain as much information as possible at a level that is automatic. If this can be accomplished, Swanson contends, there will be more time for the learner to engage in more strategic reading behavior. He identifies the need for extensive practice in reading sub-skills as integral to the successfid integration of strategic knowledge (metacognition) and recommends both as part of an integrated program of instruction for children with reading disabilities.

In their study, Bowey, Cain, and Ryan (1992) make the observation that due to poor phonological processing skills, children with reading disabilities rely heavily on having a sight word vocabulary. However, inefficient rehearsal strategies and short-term memory capacity make this method of acquiring words challenging. Bowey et al. note that visual access is slower than efficient phonological decoding for identifying unfamiliar words. This holds true for

(30)

Literature Review 22 children with and without reading disabilities. This having been said, however,

Swanson's (1982) support for better rehearsal strategies makes the acquisition of a sight word vocabulary plausible. Ideally, rehearsal of non-phonetic, high frequency sight words should be part of the toolbox of strategies taught to students with reading disabilities. The final area of information processing to explore is that of the higher-order process of metacognition.

Metacognition and Word Recognition

Attention, perception, working memory, and long-term memory are all information processes controlled by executive functioning. Brown (1 980) describes this executive functioning or control and coordination of learning processes by its more popular name, metacognition. Metacognition basically describes the learner's awareness of what skills, strategies, and resources are necessary to perform a cognitive task. As well, the learner can access self-regulation strategies so as to monitor thinking processes. According to Brown, these may include self-correcting strategies when what is being applied is not working well. Wong (1986) argues that older, poor readers are still equating reading with decoding. Their peers in the meantime are developing increasingly sophisticated metacognitive strategies as they engage in a broader range of reading experiences. Wong sees this deficient awareness in the poor, older reader as an indication of a need for instruction in self-monitoring procedures. It is her hope that not only will this help their academic skills in general to improve but also that they will become more autonomous like their non-disabled peers (p.22).

Wong (1 986) also presents a compelling argument for including metacognitive strategy instruction as a means of compensating for poor phonemic awareness. She is not convinced that the emphasis Torgesen (1999) and Stanovich (1988) place on phonological processing means that this is the only source of potential help for these children. She argues that the failure of students with learning disabilities to learn metacognitive strategies is not the issue, but rather mastery of the strategies to such an extent that they can be generalized to new learning is the problem. Wong suggests that metacognitive instruction include "mediated mindfulness" (p. 1 10) or self-awareness and self-regulation training as well as direct instruction in how to generalize

(31)

Literature Review 23 what is learned to new learning. This is in line with the earlier writing of Brown

(1980). She points out that children with learning problems tend to be passive in their learning for fear of failure. The acquisition of metacognitive skills that would put them into the center of their own learning could prove to be very empowering and fits well with the information- processing model used as a framework here. As is true for learning any new skill, children with learning difficulties require a scaffolding of instruction, a strategic model promoted by Wong. In terms of a complimentary fit to the information processing framework used for this study, the image of a scaffold of skills beginning with lower-level phonological processing and building up to higher level processes such as metacognition is appropriate.

Metacognition is often written about in the context of higher-order reading skills such as comprehension (Wong, 1986) but Bos and Vaughn (1 99 1) discuss its role in word identification as well. When a normally developing reader comes across a word that he or she does not

recognize, metacognitive processes are automatically applied to determine what the word is. This may involve making a decision to associate the word with a picture, rehearsing the word, or doing both simultaneously. It can also involve looking at the word for familiar patterns. Following this metacognitive word analysis, the learner can decode it more efficiently than if these thought processes did not take place. Metacognition is proven to be a powerfid tool when used in combination with phonological decoding in the following study.

The study conducted by Lovett, Lacerenza, Borden, Frijters, Steinbach, and De Palma (2000) attempted to assess the efficacy of a combination of phonological and metacognitive (also referred to as strategy-based) remedial approaches for reading disability. The treatment program took place over a seventy-hour period and included different combinations of approaches involving phonological decoding and metacognitive thinking. The intent of the study was to use a sequential crossover design to address the question of whether a phonologically based

remediation approach is sufficient to achieve the best outcomes possible. The flip side to that question was the second one addressed by the study: would a combination of phonological and metacognitive approaches result in superior remedial outcomes?

(32)

Literature Review 24 The results indicated that those children receiving instruction in using both

approaches had superior learning outcomes in word recognition. Another significant finding was that older children with disabilities demonstrated improvement in phonological analysis, sound blending, and letter-sound associations as a result of the intensity of the remediation. This study offers hope for older and younger children with mild to severe reading disabilities. The

effectiveness of combined strategy approaches in conjunction with the intense instruction appears to be worthy of future research.

In summary, the literature presents diverse views on which area of processing is the weakest in children with disabilities with regards to word recognition. Many studies cite Stanovich (1988) and support his hypothesis that most children with reading disabilities have deficits in phonological processing. Research done by those taking more of an information processing/constructivist view, in particular, (Swanson) based on studies and reviews from 1982-

1999, examines both lower- and higher-level cognitive processing deficits and recommend remedial programs that teach strategies in more than one area.

Despite these differences, there seems to be some consensus that remediation needs to begin at the word recognition level. Early identification and intervention of potential reading disabilities is of course of optimal importance (Felton & Pepper, 1995; Santa & Hoien, 1999), but for those children who advance to later grades still reading well below the level considered average for their age, remediation programs that provide intensive strategy instruction are critical. There is certainly no disagreement in the literature about the superiority of programs characterized by intensive instruction in reading strategies but which strategies should be taught and whether or not they should be addressed as isolated skills or an integral component of other skills to be taught remains unclear.

Thus far, phonological processes, working memory, rehearsal, and metacognitive

processes have been discussed in relation to their contributions to word recognition in reading. If effective word recognition strategies can be taught to children with reading disabilities, it is worth exploring whether or not they will generalize to reading comprehension and fluency. The

(33)

Literature Review 25

following section will present the findings in the literature related to the

comprehension and fluency of readers with learning disabilities. The implications for instruction will also be examined.

Reading Comprehension and Fluency

Reading comprehension has been defined as the process of constructing meaning from written text by integrating the information provided by the author with the reader's background knowledge (Bos & Vaughn, 1991). Mastropieri and Scruggs (1 997) report that students who have learning disabilities typically demonstrate substantial deficits in reading comprehension. Their difficulty in remembering what they have read may include the general idea or gist of the passage as well as the facts and details presented. Bos and Vaughn attribute comprehension problems to a number of factors. For one, many children with reading disabilities are already expending considerable energy to word identification so there is little attention left for

comprehension. Other children with reading disabilities consider reading to be saying the words correctly, a phenomenon referred to as "word calling" by the authors. Children who word call may read with reasonable fluency but as they do not attend to the meaning of the passage, comprehension is compromised. Syntax and semantic processing skills can also be the culprit. If children do not have a good understanding of word meanings and grammatical structures, comprehension will suffer. Some children fail to use metacognitive strategies when they read; they don't integrate the information into their existing schema to make sense of it or they fail to monitor their comprehension and/or word identification skills as they read.

Fluency or the ability to read in a smooth, flowing, and seemingly effortless manner occurs when the reader is able to automatically process information at both the visual and phonological levels. This automaticity in word recognition frees their attention to focus on the meaning of what they are reading and integrate this information with their background knowledge or existing schemas (LaBerge & Samuels, 1974; Samuels, 1987). Fluency then, appears to result from automatic decoding. LaBerge and Samuels note that if children can become more skilled at decoding individual words and this becomes more automatic for them, they can devote their

(34)

Literature Review 26 cognitive resources to reading for meaning. There appears to be a reciprocal

relationship between fluency and comprehension; a reader must read fluently to comprehend what is read and at the same time, as the child is able to comprehend what he or she is reading, the more fluent the reading becomes (Dowhower, 199 1 ; Shreiber, 199 1).

Dowhower (1 99 1) suggests that all three components of fluency - rate, accuracy, and

expression contribute to improved reading and comprehension. She points out that although rate and accuracy tend to be focused on more than expression due to the tendency of most individuals to read silently as they age, expression or prosody should not be overlooked as aids to

comprehension. The melodies and rhythms of expression will, in her view, contribute to the meaning and enjoyment of the reading whether it is read aloud or silently. From that point of view, expression, as a function of fluency, becomes a worthwhile strategy to teach children with reading disabilities and indeed, all developing readers. Schreiber, (1991) suggests that the use of rehearsal strategies for fluency such as repeated reading would improve fluency because of the opportunities it creates for children to attend to syntax in the passages they read.

Homan, Klesius, and Hite (1993) compared the effectiveness of repeated readings and teacher-assisted, non-repetitive reading methods on the comprehension of at-risk students. Samuels's (1979) method of repeated readings involves the following steps: 1) the student reads a short passage aloud while the teacher times the reading and records the miscues, 2) the passage is then reread silently or orally several times, 3) the student rereads the passage aloud again and the teacher once again records the time and miscues, and 4) the teacher and/or student prepares a graph to depict the growth in performance from the first to the last reading. Three re-readings appear to have the most positive affect on fluency, accuracy and story retellings (Sindelar, Monda, & O'Shea, 1990). In their study, Sindelar et al. found that student retellings were better after three re-readings than the seven recommended in Samuel's method.

The teacher-assisted, non-repetitive reading strategies that Homan et al. (1993) describe are considered non-repetitive because the child does not repeat the text in the same way as the repeated reading method described above. The results of the Homan et al. study on the use of

(35)

Literature Review 27 repeated reading and assisted reading strategies indicated that both methods were

equally effective at improving fluency and comprehension. Homan et al.'s findings support the value of allocating time for students, particularly those with learning disabilities to engage in reading real text. The authors do caution against overuse of repeated readings however. They fear that an overuse of either method will create a negative effect on the reading attitudes of these students. The suggestion is made that if the text is expository in nature and the purpose for reading is to study, then the redundancy that comes from repeated reading might be beneficial.

Rehearsal strategies such as repeated reading and teacher-assisted non-repetitive reading strategies could be considered lower-level processes in the information-processing model. Theoretically, they work on lower-level processes to promote automaticity, allowing more cognitive resources to become available for processing the meaning of the text, hence, improved comprehension. O'Shea, Sindelar, and O'Shea (1985) conducted a study to explore the effect of attentional cues on reading fluency and comprehension. Students with learning disabilities were cued to either read quickly and accurately or to read and read and remember as much as they could. Those students cued to remember, read with greater recall than those who were cued to read quickly and accurately. In their review of best practices for facilitating improved

comprehension in children with learning disabilities, Mastropieri and Scruggs (1 997) identified one potential shortcoming of the repeated reading approach: although it provided greater reading fluency, the students did not receive the important information with regard to how they could use the cognitive resources that had been freed up for comprehension. As part of a rehearsal strategy used in this study, repeated reading is worthy of consideration.

In their discussion on other comprehension practices in the field, Mastropieri and Scruggs (1 997) conclude that regardless of the orientation of the researcher; behavioral, cognitive, or constructivist (information processing), comprehension improved when instructional procedures were intentional and logically related to the specific process of comprehension. As noted earlier, they too, support the contention that children with learning disabilities suffer deficits in both top- down and bottom-up cognitive processes. Intervention that specifically targets one or more of