Writing issues in college students with learning disabilities: a synthesis of the literature from 1990 to 2000.
In 1989, Professor of English Carolyn O'Hearn brought attention to the lack of research and scholarship on college student writers with learning disabilities (LD), saying that "very little has been written specifically about the LD college writer" (p. 295). O'Hearn (1989) conducted a review of the literature at that time, in an effort to assist the instructors who work with this growing college population. More and more students with LD were entering college, and thus enrolling in composition courses and laboring through written assignments. These students were encountering instructors who might not have known anything about their difficulties with written language. Additionally, the literature about these issues was sparse. In O'Hearn's words, "the relative absence of scholarship in this area is indeed unfortunate because composition is crucially important to the success or failure of the LD college student" (p. 295).
The difficulties of students with LD have been explored by many researchers (e.g., Gregg & Hoy, 1990; Hughes & Smith, 1990; Leuenberger & Morris, 1990), even if the specific causes have not been addressed. For example, Blalock (1981) claimed that 80% to 90% of adults with LD experience written language difficulties. Furthermore, writing problems have become a major concern of students with LD and their instructors. Indeed, they are believed to exceed students' other academic difficulties (Gajar, 1989; Ganschow, 1984; Gregg, 1983; Plata, Zelhart, & House, 1995; Scott, 1991). A review of the literature shows that the writing difficulties range from mechanical aspects of writing, that is, spelling, punctuation, and capitalization, to content aspects of writing, such as organization and coherence issues. In addition, students with LD are also likely to experience perceptual difficulties, suffer from test anxiety, demonstrate poor study and planning skills, and have weak revision skills (Higgins & Zvi, 1996; Sills, 1995; Stracher, 1993). These writing difficulties tend to be "exacerbated in secondary schools, where more complex curricular demands and higher teacher expectations compound the difficulties of adolescents with LD' (Hallenbeck, 1996, p. 108). If the secondary environment increases the demands on students with LD, then, as Norton asserts (1992), the next educational step, the postsecondary experience, must place its own unique and intense demands on writers with LD. As Norton writes, many institutions of higher education "acknowledge that learning disabled students at the postsecondary level present unique needs" (p. 106).
With this reality in mind, and in light of the increasing percentage of students with LD who progress to the postsecondary level, the authors asked the following questions: How far have we come in the decade since O'Hearn's literature survey in exploring the specific issues that face college student writers with LD? Much information is necessary to successfully meet the needs of these students. Has this information been culled and disseminated since 1990 in a way that O'Hearn claims it had not been before that time?
Charged through the Program to Enhance and Ensure Learning for Students with Disabilities (PEEL) grant with conducting a survey and synthesis of the literature on college students with LD, the authors searched 67 peer-reviewed journals in both LD and composition/writing since 1990 to see if the dearth of research on this population and its writing issues that O'Hearn discovered had been addressed.
The PEEL grant charge limited the scope of the search to postsecondary students; therefore, literature on younger students with LD was not included. Further, the literature review did not seek to provide a comprehensive critique of the work that has been done in this time period, although critical comments are provided in presenting the articles. Rather, the purpose was to synthesize the literature of the last decade from both the LD and the composition/writing fields that focuses on college students with LD, in order to determine whether or not the lack of research that O'Hearn identified has been addressed, and to spark dialogue in these two fields on how to proceed to ameliorate the challenges that face this student population.
As noted in the current literature review, various names have been assigned to college students with LD, such as postsecondary learning disabled students, college students with writing disabilities, LD college students, college writers with LD, college students with dysgraphia. Further, the review of literature also revealed that only two articles specified their samples as college students with writing disabilities as opposed to the broader label of learning disabilities. In this article, the authors uniformly use the term "college students with LD/WD" when referring to students with LD and writing difficulties in or from a four-year or a two-year postsecondary college, except when a specific term by the researchers is used as their study is presented and discussed in this article.
In the studies reviewed, some researchers did not provide a definition for learning disability. Among those who did, most conform to the definition by the National Joint Committee on Learning Disabilities (e.g., Raskind & Higgins, 1995). Some researchers (e.g., Neff, 1994; Sills, 1995) use the definition from Public Law (PL) 94-142 and its recent update, PL 99-45, the Individuals with Disabilities Education Act (IDEA). IDEA defines learning disability as a "disorder in one or more of the basic psychological processes involved in understanding or in using spoken or written language, which may manifest itself in an imperfect ability to listen, think, speak, read, write, spell or to do mathematical calculations." Learning disabilities include such conditions as perceptual disabilities, brain injury, minimal brain dysfunction, dyslexia, and developmental aphasia. Learning disabilities do not include learning problems that are primarily the result of visual, hearing, or motor disabilities; mental retardation; or environmental, cultural, or economic disadvantage. Whereas IDEA is geared toward a younger LD population (under age 22 and attending high school or lower grade levels), the Rehabilitation Act of 1973, Section 504, mandates that two-year and four-year colleges provide accommodations to students with LD, such as extended test time, a scribe, a reader, and the right to tape a lecture.
As mentioned, little effort has been made to delineate subgroups of learning disabilities, and studies have not reported on the different characteristics of subgroups of learning disabilities (i.e., math, reading, spelling, written expression, spoken expression). Thus, it appears that that the term "writing disability" is still not specifically defined and the diagnostic criteria vary across studies, with most studies conforming to local institutional, state, and national definitions and criteria. For example, in a study on the assessment of the writings of college students with and without LD, Morris-Friehe and Leuenberger (1992) identified their students with LD in accordance with their state's special guidelines. The specific criteria for LD are a Full Scale IQ of 85 or higher and a 1.3 or greater standard deviation discrepancy between ability and achievement in one or more of five areas (writing expression, spoken expression, math or listening comprehension, and reading) on at least one standardized measure. Similar criteria were used in a study by Leuenberger and Morris (1990), who also included the criterion of "no sensory, mental, emotional or environmental deficiencies." The students with LD in the study by Raskind and Higgins (1995) were identified based on the criteria specified by the chancellor's office of the state university system and also by the definition of the National Joint Committee on Learning Disabilities. Specifically, they required that students with LD have a Full Scale IQ of 90 or better and a significant intracognitive discrepancy and/or achievement and ability discrepancy; in addition, students were to have an average or greater score (25th percentile or above) in at least one academic area. Instruments to determine the ability and achievement discrepancy are mainly the Wechsler Adult Intelligence Scale-Revised (Wechsler, 1981), Woodcock-Johnson Psycho-Educational Battery (W-J) (Woodcock & Johnson, 1977), and Test of Adolescent Language (TOAL) (Hammill, Brown, Larson, & Wiederholt, 1980). Some researchers used either the WJ or TOAL as the standard achievement score in the same study without further validating the concurrent validity of the two instruments, which may negatively affect the results.
The studies reviewed in this article were selected using the following steps. First, a thorough computer search was conducted using various on-line databases and search engines, such as ERIC, PsycInfo, Northern Light, and the local university library system. Next, articles on writing and learning disability in postsecondary students were identified. The second author, the supervisor of the writing project, also provided journal articles and book chapters on similar topics. The authors searched the contents of 67 peer-reviewed journals from 1990 to 2000 from both the LD and composition/writing fields. For each article meeting the criteria (i.e., the content is related to college students with both LD and/or writing difficulties), the reference section was carefully examined to identify additional studies to be reviewed. The search yielded a total of 38 articles on college students with LD/WD.
SEARCH RESULTS AND SYNTHESIS
Based on the search results, four content categories appeared to be most prevalent in the related literature: (a) overview of the available assistive technology for college students with LD/WD; (b) empirical studies of the effectiveness of assistive technology for college students with LD/WD; (c) characteristics and error patterns in the writings of college students with LD/WD; and (d) instructional methods for this student population. Reviews of the literature for each of the four categories of studies are presented individually in the following sections. A list of the four categories with the authors' names and publication year is provided in Table 1. Finally, full citations of these articles appear in the reference section.
The following reasons may account for why educators and policy makers put more emphasis on the four categories identified than others. First, more and more students with LD are going to college, with 67% of the 100,000 students with LD exiting high school every year planning to attend secondary institutions (Office of Special Education Programs, 1992; White, Alley, Deshler, Shumaker, & Warner, 1982). Thus, there is an urgent need for postsecondary institutions to provide programs to help these students succeed, and it is essential to educate general faculty and other professionals about the learning features of students with LD and the areas where they most need help (e.g., Gregg, Hoy, Mcalexander, & Hayes, 1991). Second, after getting to know the features and error patterns of students with LD/WD, LD practitioners put a lot of effort into exploring instructional strategies and testing what strategies are most effective in teaching these students. Third, no more than 15% of educators use computer technology in instruction and many do not use it at all (Hunt-Bert & Rankin, 1994), due to ignorance of the advanced capability of computer systems and existing educational software. Thus, there is a need to "inform educators of the technology available that can offer advanced program in current computer system" and help students with LD/WD in the various stages of the writing process ((Hunt-Bert & Rankin, 1994, p. 169). Finally, the four categories are essential for students with LD/WD to know what is available when they need help.
Thus, it may be mainly due to pragmatic concerns that the majority of the studies fall into the four categories presented above. Most of the empirical studies reviewed here, especially those examining the effectiveness of assistive technology and characteristics and error patterns, were funded by the Office of Special Education. A more comprehensive examination is needed to determine if the topics and publications are influenced by the nature of funding competitions.
The four categories are further classified into four types of articles based on their purpose, main content, and research method: research report, literature review, informational, and program description. The types of articles, name of authors, year of publication, and number of articles in each type are presented in Table 2.
The following is a synthesis of the studies and conclusions discussed in these 38 articles organized by the four major themes.
Overview of Available Assistive Technology
Articles that provided an overview of available assistive technology take up about 15% of the literature found on college students with LD/WD (6 articles). Electronic technology of various kinds has been used to help college students with writing disabilities to compensate for their difficulties. This type of technology is usually referred to as assistive technology (Raskind & Scott, 1993). Many factors have contributed to the wide use of assistive technology (AT) with postsecondary students, including the rapid growth of electronic technology; the increased number of college students with LD attending college; the awareness of technology developers of the huge potential market for software that addresses the needs of individuals with LD; and the passage of federal laws to protect people with disabilities, such as the Americans with Disabilities Act (ADA) of 1990, the Individuals with Disabilities Education Act (IDEA) of 1990, and the Technology-Related Assistance for Individuals with Disabilities Act of 1988 (Raskind & Higgins, 1998). In addition, research institutes, organizations, and others have established high-tech centers and programs for students with disabilities, thereby becoming formidable forces in promoting the use of AT by students with disabilities, including learning disabilities.
Many researchers have realized the importance of assistive technology in helping students with LD (e.g., Collins, 1990; Hunt-Bert & Rankin, 1994; De La Paz, 1999; Raskind & Higgins, 1998), but few studies have focused specifically on the use and availability of AT for college students with writing difficulties. According to the six articles found with this focus, word processing, speech recognition systems, speech synthesis systems, and multimedia technology are the most frequently used tools for helping this population.
Word processing. Students with writing disabilities experience two major types of difficulty when composing, (a) with the content of the text, such as coherence and organization and (b) with the mechanical aspects of writing, such as spelling and editing (Graham, Harris, MacArthur, & Schwartz, 1991a; Isaacson & Gleason, 1997). Problems with the mechanical aspects frequently inhibit students' success with the higher-order demands associated with planning and organization of the content. Thus, it has been postulated that college students with LD/WD can focus on writing more meaningfully if the mechanical burden is removed. Current word processing technology works to alleviate this burden (De La Paz, 1999; Hunt-Bert & Rankin, 1994). Besides the features known to most computer users, such as storing, copying, pasting, cursor control, spell check, and inserting and formatting text, word processing possesses features that are uniquely important to people with writing difficulties. For example, software has been developed that combines both visual and auditory input, offering cues in either or both modes, such as with Write Out Loud (Don Johnston Developmental Equipment Inc., 1993), which cues the writer with an alert sound or flashing of the word. Anticipated spelling programs such as Cue Write (Communication Skillbuilders, 1988) and Write Away (Institute on Applied Technology, 1991) prompt writers to choose correct spellings from a menu of word choices that appear at the bottom of the screen after they type the first letter of an intended word. However, some questions remain unaddressed. For example, how many words with the same initial letter would appear? Where and how would the wanted word appear? Is it difficult for a writer to find the wanted word from the list, and to what extent will this word search interrupt the writer's thought flow?
Word prediction is regarded as a significant application of word cueing (Hunt-Bert & Rankin, 1994). Word prediction programs, such as Cue Write, can offer words based on word frequency, recency of use, grammatical correctness, and word association. The prediction program displays a predicted word in a word list that appears on the screen separate from the text. While not specifically created for writers with LD, these word processing programs offer some potential benefits for this population in that they can achieve more word accuracy, promote better typing efficiency, and provide words richer in meaning (Hunt-Bert & Rankin, 1994; Raskind & Scott, 1993).
In addition to the basic mechanical help that word processors can provide, some word processing programs offer organizational assistance and grammar correction to help students with LD/WD to produce and organize text with appropriate grammar and style. College students with LD/WD often have difficulty in generating topics and putting their ideas on paper (Hughes & Smith, 1990; Hunt-Bert & Rankin, 1994; Raskind & Scott, 1993). Word processing programs, like Process Writer (Scholastic Software, 1990), help users to generate topics through a variety of brainstorming activities and topic suggestions. The Process Writer program also provides an organizational framework that helps writers develop cohesive texts and allows writers to select a particular writing style. Some outlining programs allow students to "dump" ideas into computers, and then sequence them in a meaningful order with major headings and subheadings (Raskind & Scott, 1993). According to Raskind and Scott (1993), such programs are especially useful for writers who either tend to be so overwhelmed by detailed description that they fail to see the big picture or those who tend to lack detail in their writing, since the program enables such writers to see both the "forest" and the "trees," or either the "forest" or the "trees."
Finally, word processing allows writers, including college students with LD/WD, to benefit from grammar correction either through the built-in features of a word processor or through add-on programs. Both options help identify grammatical errors and provide alternatives. Grammar tutoring programs, such as Correct Grammar (Word Star, 1991), facilitate the actual learning of correct grammatical rules in addition to helping accomplish a writing project (Hunt-Bert & Rankin, 1994).
Speech recognition systems. Speech recognition is regarded as one of the "most exciting and important breakthroughs in technology available to post-secondary students with LD" (Raskind & Scott, 1993, p. 249). Referred to as speech input technology, this technology allows users to speak to the computer instead of writing on it, and therefore can be an excellent compensatory aid to writers with LD, who usually possess oral communication skills superior to their writing capabilities (Kupersmith, 1990; Raskind & Scott, 1993; Sills, 1995). Currently, affordable continuous speak recognition systems are available, and their error-correction procedures have been made more effective with Dragon Naturally Speaking Systems, which allow a user to move around a document, edit and format a text, and play back recorded speech (De La Paz, 1999; Raskind & Scott, 1993).
Speech synthesis systems/screen reading. Computer scientists and technologists have not only made it possible for users to talk to computers, they also make computers "talk" or read what the user has written. Speech synthesis systems are computerized voice output systems, and they can be powerful writing tools when used together with a word processing system. Together with "screen reading" software, a speech synthesizer can display a text on a computer screen and read back the text aloud to the author, allowing him or her to detect the grammatical, spelling, and punctuation errors that might otherwise have gone unrecognized if the writer had to rely on visual editing alone (Raskind & Higgins, 1998). While every PC or Macintosh computer has built-in spell and grammar checks, not every computer has speech recognition and synthesis systems/screen reading, which are specially designed for people with auditory, visual, and/or learning disabilities. As a result, due to the extra cost involved in purchasing such programs, there might be an availability problem for students with LD/WD in institutions that are financially constrained.
Multimedia. Besides the many computer programs designed to help college students with LD/WD to compose meaningful and cohesive texts, researchers have recommended alternatives to traditional written assignments, such as essays and research papers (Speziale, 1993). For example, according to Speziale (1993), instead of having college students with LD/WD write a lengthy research paper in the traditional format, instructors may allow them to construct a presentation or research report by using multimedia, integrating audio, visual, graphic, and textual information, as well as hyperlinks. According to Speziale (1993) and Speziale and LaFrance (1992), multimedia technology may cultivate creativity and motivation in college students with LD/WD, give them an opportunity to break up information into manageable units, and provide effective alternatives to traditional writing. In these studies, multimedia was not used as assistive technology but as an alternative to traditional writing processes and formats for college students with LD/WD. There is, however, an extremely limited literature on the application of multimedia with college students with LD/WD, either descriptive or empirical. Thus, the authors' claims of its advantages warrant further and wider range of exploration and verification.
On the whole, word processing, speech recognition, and speech synthesis software are the main assistive technologies used to help college students with LD/WD. While researchers have recommended the use of these programs as compensatory strategies to improve the writing quality of this student population, they did not elaborate in their articles on the availability and accessibility of AT on campus, considering the increasing number of students with LD entering college. Furthermore, researchers did not reveal the potential problems in using these programs. For spell check and grammar check, for example, the red lines indicating misspelled words, punctuation, and poorly constructed sentences can cause frustration and anxiety, especially when students cannot spell or identify the right word from the word list in a way that allows the spell check to make a reasonable choice.
Empirical Studies of the Effectiveness of Assistive Technology
In our review of literature from 1990 to 2000, we noted that empirical studies on the effectiveness of AT for college students with LD/WD are still sparse, with only four studies found. In addition, the same group of researchers, using the same group of samples and similar research methods and procedures, conducted most of the existing studies. Using the same sample may affect the reliability of results due to practice effect and may also affect generalizability of findings to a larger population. Lack of comparative studies and cross-validity between studies may make the findings less likely to practically impact the field.
Collins (1990) examined the effectiveness of word processing technology in helping college students with LD in a mainstream first-year college writing course. Over the course of three years, 57 students and 8 teachers in 18 separate teaching sections were involved in the research. The writing instruction was conducted in a computer classroom, where LD and non-LD students had access to their preferred computer types (Apple II, Macintosh 512, Zenith 159, and IBM PCs), and different word processing software, such as MacWrite for Macintosh, Applewriter, Appleworks, Fredwriter, Wordstar, Volkswriter, and PCWRITE, or products students brought for their own use. The impact of the AT use was measured by comparing LD and non-LD students on the basis of course completion rates, course grades, changes in attitude towards writing, and writing fluency. The results revealed little difference between LD and non-LD students in course completion rates and course grades, suggesting that word processing software was effective in helping writers with LD. Furthermore, Collins (1990) reported that the apprehension of students with LD was reduced significantly from the first course to the second course in all three replication cycles, while the non-LD group, which began with a low apprehension level, did not show significant apprehension reduction. Writing fluency was measured by means of holistic scores, total number of words generated, and spelling. The results showed that non-LD students did better, especially in spelling, although both groups experienced improvement. However, LD students made significantly more spelling errors than the non-LD group. As Collins pointed out, the word processing software can function in the same ways as writing instructors, helping students with LD build up self-confidence and avoid the obstructions encountered with pen/paper writing. However, many of these LD/WD students have repeated the program four or five times just as they might have worked with the same teacher who had had similar homework requirements each semester. Thus, the practice effect may have contributed to their elevated grades. If the author had distinguished LD students who took the class for the first time from those who had taken it many times, the result might have been more informative and convincing.
Higgins and his colleagues (Higgins & Raskind, 1995; Raskind & Higgins, 1995; Higgins & Zvi, 1996) studied the compensatory effectiveness of the more advanced forms of AT, focusing on the impact of speech synthesis and speech recognition programs on proofreading efficiency and written composition performance of college students with LD/WD. In their speech synthesis study (Raskind & Higgins, 1995), the LD participants were instructed to generate an essay by hand or word processor. Two weeks later, students returned to proofread and locate errors in their essays under three separate conditions: (a) using the speech synthesis/screen review system (SS), (b) having the text read aloud by a human reader (RA), or (c) with no assistance (NA). Participants were able to detect a significantly higher percentage of total errors under the SS condition than under the other two proofreading conditions. In addition, under the SS condition, the participants identified a significantly higher percentage of capitalization, spelling, usage, and typographical errors.
One explanation of these results offered by the researchers is that the synthesizer may provide easy access to the text without preoccupation with decoding, the major task of proofreading, thus enabling subjects to spend more time locating errors. Furthermore, Raskind and Higgins (1995) believed that students might have been motivated by the novelty of the assistive technology in proofreading, making them more careful in their editing.
According to the researchers, one observation worth noting is that the subjects identified significantly more grammatical and syntactical errors when a human reader read the text to them (RA condition) than in the SS and NA conditions. Moreover, subjects in the no-assistance condition outperformed those in both SS and RA in detecting content and organization error, although the difference was not significant. This seems to suggest that while SS can be very useful in helping students with LD/WD proofread and detect basic mechanical errors, independent reading and human readers helped to identify errors of higher-order writing concerns, which can be difficult for computers to detect. This finding suggests limited use of SS since regular computers can do the spell check and grammar check. What is more, SS costs much more. Another problem with SS is its flat reading style, without appropriate pauses at the right syntactic junctures. It is presumed that assistive technology needs to be combined with extensive support from LD specialists to be cost-effective (Collins, 1990). Further improvement of SS and making it more "human like" in terms of reading meaningfully with appropriate pauses at right places warrants exploration.
Although SS helps students with LD/WD, to some extent, in proofreading, better proofreading does not guarantee better correction and better written products. Higgins and Raskind (1995) and Higgins and Zvi (1996) investigated the effectiveness of speech recognition software on the written composition performance of college students with LD. Participants wrote three essays under three conditions: (a) using a speech recognition system (SR), (b) dictating an essay to a human transcriber (TR), and (c) with no human or computer assistance (NA). In both studies students with LD demonstrated higher holistic scores in the SR than in the NA condition. The authors postulated that the technology was responsible for this finding in two ways. First, the dictated words are automatically spelled correctly for the students. Second, the freedom from over-involvement with spelling and checking spelling allows college students with LD to attend to more important concerns, such as content creation, organization, and effective use of language. However, in Higgins and Zvi's study (1996), LD students not only received AT training, they also received one-on-one tutoring. Thus, the improvement in their writing could not be only attributable to AT. Higgins and Zvi also explored emotional factors, such as anxiety, that affect the performance of students with LD/WD. This issue will be presented again in the future research section.
Characteristics and Error Patterns
Seven articles in the literature from 1990 to 2000 explored characteristics and error patterns in college students with LD/WD; six of the seven articles are research reports. Researchers in the area of college students with LD/WD share the concern that there is a lack of empirical research investigating the written language skills and error patterns of college students with LD/WD and that there is little empirical evidence to support their instruction. Researchers have mostly compared the writing performance of college students with LD with that of non-LD writers, and some focused on LD writers themselves and the nature and features of their writing difficulties. Researchers employed different error analysis systems to identify and analyze errors made by college writers with LD/WD, which makes it difficult to compare findings. The literature review yielded two categories of errors and characteristics, although different researchers use different names for each category: (a) mechanical errors (i.e., spelling, punctuation, and capitalization), and (b) content problems (e.g., planning, organization, and coherence). For a review of studies on this topic before 1990, please refer to Scott (1991). The following section will discuss the findings in each category; in addition, research findings on gender differences in college students with LD/WD will be presented.
Spelling, punctuation, and capitalization. One error pattern that could distinguish LD writers from non-LD writers is the significant number of spelling, punctuation, and capitalization errors they make (Gregg & Hoy, 1990; Gregg, Hoy, & Sabal, 1988; Leuenberger & Morrist 1990). Different researchers gave different names for this type of errors, such as conventions (Morris-Friehe & Leuenberger, 1992), mechanics error (Gregg & Hoy, 1990), residual errors (Gregg, Hoy, Alexander, & Hayes, 1991), and misspelled subtypes (Leuenberger & Morris, 1990). This error pattern has been observed and supported by clinical observations and findings from both qualitative and quantitative studies (e.g., Gregg et al., 1991; Smith, 1993). Different explanations have been given to account for this error pattern. For example, according to Gregg et al. (1991), writers with LD may suffer certain cognitive processing deficits that affect their ability to acquire or produce rules to present ideas in appropriate sentences and to use appropriate punctuation. Such mechanical errors indicate a syntax disorder rather than a lack of instruction, and this error pattern can serve as a diagnostic tool to identify students with writing disability (Duques, 1989).
Coherence errors. Textual coherence is a critical aspect of a written product. It can be achieved through different means, one of which is cohesive referencing (Gregg & Hoy, 1990). Referential cohesion, with the proper use of different types of pronouns, helps the writer to communicate effectively with the intended audience. Students with LD/WD are believed to frequently confuse readers with an unclear writer-audience relationship (Gregg & Hoy, 1990), which affects the reader's overall impression of the quality of the student's writing. Improper cohesive pronoun use indicates difficulty with linguistic devices, a sense of audience, and the actual cognitive demands of the text (Litowitz, 1981).
Gregg and Hoy (1990) compared frequency and accuracy of pronoun references in a narrative essay in three groups of college students--normally achieving, learning disabled, and underprepared. The results demonstrated significant group difference only in the frequency of nominate and objective pronouns among the eight different types of pronouns. Students with LD and normally achieving students showed similar patterns and underprepared students had the highest frequency of nominate and objective pronouns. Students with LD were also found to have better comprehension and to produce more coherent text structure than underprepared students, as measured by a holistic coherence scale (a 5-point instrument, 5 indicating a full coherent essay based on prescribed criteria). However, the result of the holistic scoring of essays using a quality scale w a scale provides a total score from the sum of four subscores in the areas of mechanics, organization, development, and style--revealed that college students with LD differed significantly from normally achieving writers. Gregg and Hoy (1990) attributed this finding to the punctuation and spelling errors of LD writers, which affected the rater's holistic judgment of the writing product.
Error patterns and gender difference. Among the studies of characteristics and error patterns in college students with LD/WD, only two examined gender differences. Plata et al. (1995) found that female college students with LD, in comparison with their male counterparts, achieved the lowest mean score in a Junior Level Essay Test, a writing test designed to determine if the students were ready to continue their studies at advanced levels and to graduate. The Junior Level Essay Test was measured by a holistic assessment score composed of six levels of competence. It was also found that more male LD students than female LD students requested alternative tests. Male LD students took the same test more often than the female LD group, which Plata et al. believed was due to a lack of male confidence or overconfidence in the female group.
In studying the error patterns of college students with LD/WD, Leuenberger and Morris (1990), using both quantitative and qualitative spelling error classification systems, found that male college students with LD demonstrated a significantly higher total error rate and percentage than their female counterparts. In addition, male students with LD/WD made significantly more phonetic, spelling, and lexical errors (homophone, association, and morphological errors, such as "their" for "there," "woman" for "women," "use" for "used," or other omission and addition errors). The finding that male students with LD made significantly more spelling errors was supported by clinical observations and other empirical studies (e.g., Johnson & Blalock, 1987). Researchers attributed such findings and observations to differential academic and social expectations for men and women and different cognitive profiles. However, the authors cautioned against overgeneralization of gender effects in examining students with LD since the error analysis systems used in different studies vary from each other. Furthermore, Leuenberger and Morris (1990) suggested that future researchers specify the age and gender of their participants so that their studies can be more informative and more effectively compared.
Instructional Strategies and Approaches
As noted earlier, student performance in written language in postsecondary schools is of great concern to both the students themselves and their instructors. Indeed, Vogel and Moran (1982) regard clear and precise writing as important as a bachelor's degree itself. Researchers and practitioners alike have been actively involved in the study of instructional support and strategies to help college writers with disabilities. The current literature review found that articles on instructional support took up the biggest portion of the literature on college students with LD/WD, with 22 out of the total 38 journal articles, book chapters, on-line resources, and professional presentations.
A careful study of the 22 articles revealed four major categories of instructional support to college students with LD/WD: (a) instructional strategies involved in the different writing stages, (b) professional and peer tutorial support, (c) whole language approach to writing, and (d) other accommodations and support.
Instructional strategies in the various writing stages.
Regarded as an integrated part of language by many researchers, writing is an "interactive process that sanctions the writer's rethinking, rewriting, and recreating" (Stracher, 1993, p. 69). It has been generally agreed that writing involves three main stages: prewriting (rehearsal), writing (drafting), and rewriting (revision) or editing (e.g., Gersten, Baker, & Edwards, 1999; Gould, 1991; Neff, 1994; Pardes & Rich, 1996; Sills, 1995). In the prewriting stage, student writers generate thoughts and ideas and develop plans for what they are going to write based on the instructor's requirements, the intended audience, and the goal of writing.
College students with LD/WD tend to experience tremendous difficulties in the prewriting stage (Gould, 1991), often complaining that they do not have anything to put on the paper and that they do not know how to develop their ideas (Gould, 1991; Neff, 1994). Researchers and practitioners have employed various instructional strategies to help college students with LD/WD to overcome the first obstacle on their way to completing a written assignment. For example, in the writing course presented by Pardes and Rich (1996), the teacher first modeled a self-instructional strategy to show students how to find important information from reading materials and put the information on note and quote cards. After a series of class discussions about the notes and quotes, students tried to integrate them into a meaningful and cohesive piece of writing. In this article, the researchers take the amount of time needed to finish a written assignment as an indicator in student essay evaluation. This seems to contradict ADA and most educators, who espouse longer writing time for students with LD/WD. Thus, the effectiveness and reasonableness of using time limits to evaluate the writing products of students with LD/WD might be challenged by both educators and students.
In a meta-analysis of research-based instructional approaches to teaching writing to students with LD, Gersten et al. (1999) found that plan of action is a frequently used strategy in developing plans for handling the various subtasks involved in writing longer pieces. One type of plan of action is the "Planning Think Sheet," which prompts the writer with questions related to the audience, purpose, background knowledge, and ways to group and organize ideas (Englert, Raphael, & Anderson, 1992, as cited in Gersten et al., 1999). The authors believe that this plan of action has an advantage over traditional teacher-centered instructional methods in that it allows the student and the teacher a chance to discuss the plan. This strategy is referred to as dialogues or directed conversations by Gould (1991) and writing conferences by Neff (1994) respectively.
Free writing has been used as a traditional instructional support with the purpose of recording for later use the important information obtained from lectures, readings, and other sources. However, according to Neff (1994), free writing is not an effective strategy and can even be devastating for college students with LD/WD. Writing in this unstructured manner might be their weakest point, since it does not provide explicit ideas about what to write. That is, when these students do not know where the writing is heading, the writing exercise might be of little benefit for them. Thus, according to Neff and Gould, a more encouraging and effective way to help college writers with LD to plan their writing is through directed conversations or writing conferences.
Through different examples, Neff (1994) and Gould (1991) described the advantages and significance of directed conversations/writing conferences. Neff believed that the reason why college students with LD/WD find it overwhelmingly difficult to generate ideas and plan their writing tasks is that they are not aware of the great amount of information they already possess as well as their difficulty in retrieving such information. The role of the instructor in the writing conference is to "unlock trapped information" by asking guiding questions related to the writing topic and helping students discover their ideas and identify a specific topic on which they already have sufficient information (Gould, 1991; Neff, 1994). Some researchers indicated that some college students with LD/WD have superior oral language communication skills (Kupersmith, 1990; Sills, 1995). These students can tape their ideas first and then transcribe them into writing themselves or by the assistance of another person (Sills, 1995). This way, students can concentrate on idea generation and organization without the worry of the mechanical aspects of writing, such as spelling, capitalization, and punctuation.
Gould (1991) also suggested that in the writing conference, teachers model topic selection and initial planning through a think-aloud strategy. This process of verbal self-instruction and brainstorming helps students to clarify general and specific points, decide to include and exclude, and organize their ideas in a coherent manner--all of which are the basis of the critical thinking process in which all writers should engage (Gould, 1991; Sills, 1995).
The second stage, writing (drafting), is the actual writing stage where students put their ideas down on paper. Here, instructors need to reteach the strategies used in the prewriting stage, as writers need to stop to think and rehearse what they learned in the prewriting stage and plan and organize what they will say next (Gould, 1991). Based on the current literature review, research related to this stage is sparse. Many of the articles addressing strategies in different stages of writing do not include the drafting stage, jumping from the prewriting to the revising stage.
Gould (1991) and Kupersmith (1990) are among the few who have provided specific strategies for college students with LD/WD. Specifically, Kupersmith recommended a "talk-write" technique that encourages students to write as they talk. This approach capitalizes on students' superior spoken language abilities while allowing students with "dysgraphia" to adapt their oral language skills to written language.
Guild (1991) suggested a similar verbal-rehearsal technique, which may release students with writing difficulties from the concern of making mechanical errors. In addition, Gould recommended using concrete and physical cues for students experiencing writing difficulties. For example, students can use colored paper, stickers, etc., to remind themselves that the writing is not completed and that they do not have to worry too much about the mistakes. Gould also believed that students could benefit from dictating their work to a tape recorder, freeing themselves from the mechanical impediments of writing. For students with severe writing disabilities, it is suggested that another person transcribe the tape and read the text back to the student while also asking leading questions to elicit ideas for changes and revisions (Gould, 1991).
The final stage of revision and editing is generally regarded as critical in the writing process, as it involves the final polishing of the writing product. Most researchers agree that students with LD/WD should begin their final writing stage with explicit purposes and plans as they reread their texts (e.g., Gould, 1991; Sills, 1995). This allows students to deal with one specific category of problems at a time, thus achieving more efficiency and effectiveness. The revision plans usually range from content and coherence checks to mechanical and grammar examination, with the content and coherence check occurring first. In real life, however, college students with LD/WD are more likely to focus on mechanical changes and neglect content and coherence revisions (Gould, 1991).
According to Gould (1991), college students with LD/WD tend to assume that others know what they mean in their writing, when in fact their incoherent writing often confuses readers. One technique to help students to clarify their audience, purpose, and meaning is to hold group revision conferences. Here the writer reads the material out loud, and peers ask questions about the unclear and ambiguous sections to help the writer clarify his/her ideas in the revision section. Peers are encouraged to provide positive comments on the good parts of the writing, which is very crucial to writers with disabilities, who are often not sure of what is good and what needs improvement (Gould, 1991).
Another proofreading tool available for college students with LD/WD is the SCAN checklist. SCAN helps to improve clarity and coherence during the revision stage (Graham & MacArthur, 1988, as cited in Sills, 1995). SCAN is an acronym for the following steps:
S = does it make Sense?
C = is it Connected?
A = can I Add more?
N = do I Note errors? (p. 69)
With the revision of the context and meaning of the written material accomplished, the mechanical errors of spelling, punctuation, and capitalization become the primary focus of the next step in the editing and proofreading stage. Self-check and peer-check are encouraged with the assistance of spell and grammar checks on the computer and mnemonic devices such as COPs (Capital letters, Overall appearance, Punctuation, and Spelling) (Schumaker, Deshler, Nolan, Clark, Alley, & Warner, 1981, as cited in Neff, 1994). According to Neff (1994), however, it is not enough to depend on the computer or other devices to produce error-free text. Having a good reader go through the text can help a student with LD/WD to identify possible grammatical or punctuation errors. In addition, it is suggested that teachers always be available even when the students are self-checking through computers.
Professional and peer tutorial service. Professional and peer tutors have been actively involved in helping college students with LD/WD (Ives, 1990; Manganelle, 1994; Pardes& Rich, 1996; Stracher, 1993). Professional tutors are usually reading and/or writing specialists, whereas peer tutors are typically upper-level graduate students enrolled in education and special education programs/teacher training programs. One characteristic of tutoring service is that the tutor, either professional or peer, helps college students with LD/WD in an individual manner with particular strategies, based on their specific learning difficulties and strategies (Pardes & Rich, 1996; Stracher, 1993).
In the case studies presented by Manganello (1994), after, analyzing the student's learning problems, a learning specialist, the professional tutor, developed a cognitive strategy to help the student to organize and write a longer text. The strategy involved identifying the subgoals of the writing task, and planning and writing the text. The learning specialist and the student worked together to set up a list of subgoals and then accomplished each in a logical manner.
The three case studies provided by Stracher (1993) give readers a comprehensive and detailed picture of how a peer tutor analyzed students' learning difficulties and helped them with modeling techniques and high-level leading questions. In the first study, the graduate-level tutor modeled reading a paragraph in the text, asking herself questions and paraphrasing the sentences. The student replicated the procedure and wrote down the important information for later use. The tutor and the student also went through each stage of the writing process with the tutor guiding the student using high-level questions to elicit reflective thoughts on the writing (Stracher, 1993). Stracher put forward an important issue that few other researchers have touched, that is, with IQ and available supports remaining the same, to what extent does devotion, motivation, perseverance, tenacity, and passion for a subject affect the writing performance of students with LD/WD, and what should educators do to cultivate these qualities that students with LD/WD often lack of? Further studies are needed to explore this issue.
Tutorial service has been considered an effective way to help college students with LD/WD to improve their writing skills (Ives, 1990; Manganello, 1994; Martin 1991; Stracher, 1993). Before they start to work with these students, however, tutors must be trained on the special features of writing disabilities, the special needs of students with such disabilities, and the corresponding instructional strategies. According to Stracher (1993) and Manganello (1994), tutors should be patient and empathetic because students with whom they work have experienced many frustrations and failures in their academic and social lives. Nevertheless, tutors are reminded they are not to do the work for the student with LD/WD, no matter how much they want to help, because "learning takes place when students are actively involved" (Ives, 1990, p. 9).
Whole language approach to writing. Whole language instruction finds its theoretical support from Vygosky's zone of proximity and scaffolding strategies. The core notion that whole language teachers and philosophers support is that "language, whether it is oral or written, cannot be divided into discrete subskills for instruction because the act of segmenting and focusing on a target `subskill' changes the linguistic process" (Keefe & Keefe, 1993, p. 172). Writing teachers following the whole language philosophy provide students with more time and opportunities to write. This is believed to be crucial for students with LD/WD, who tend to spend more time in writing than their average-level peers (Bardine, 1997). In whole language writing classes, the topics are mostly self-selected, meaningful, and authentic, which motivates LD students to write for their own purposes. Moreover, the whole language teacher helps create a supportive and collaborative environment through choice, ownership, self- and peer evaluation, and peer collaboration, which helps develop LD students' self-regulation and self-confidence (Bardine, 1997; Gaskins, 1995; Keefe & Keefe, 1993).
Based on the whole language philosophy, the Landmark Method of teaching writing to college students with LD/WD was created and used by the Landmark Institute (Bardine, 1997; Gaskins, 1995). Using the Landmark Method, teachers teach reading, writing, listening, and speaking in an integrated manner. Students at the Landmark Institute attend speech and writing classes at the same time, taking advantage of their superior oral communication skills and integrating them into their writing (Bardine, 1997; Kupersmith, 1990; Sills, 1995). Students are also instructed to reflect on their writing process and texts through peer collaborations and evaluation, a portfolio system, and the process approach to writing. Teaching to the students' strengths rather than remediating their deficient subskills is another characteristic of the Landmark Method. As a result, teachers teach through various modalities (such as visual and auditory), and adapt to the students' preferred learning modality to achieve the best results. Finally, teachers are encouraged to be patient and empathetic, and provide positive feedback (Bardine, 1997; Gaskins, 1995).
As indicated in the Landmark Method, peer collaboration and evaluation play important roles in enhancing the writing performance of college students with LD/WD. In line with this, Ammer (1998) proposed a peer evaluation model, in which classmates review the quality of the writing produced by students with learning disabilities. This model is believed to provide another audience that can help improve the writing skills of students with LD/WD. Ammer (1998) listed the advantages of the model, factors that teachers should take into account in developing and integrating the model, and the basic steps necessary to carry out the model. In addition, he pointed out that peer evaluators could serve as learning coaches to students with writing difficulties. Due to similar learning experiences, interests, and communication skills, the peer work not only provides opportunities for a student to improve his or her writing by seeking out collaboration and receiving constructive correction of errors, but also allows the student to get redirective advice about a piece of writing "without the stigmatism of failure that frequently accompanies such assistance directly from a teacher" (p. 269). Another advantage of peer evaluation is that it helps strengthen social skills, which students with learning disabilities often lack.
The whole language approach espouses collaborative learning and evaluation, focusing on writing as a process rather than a product. In line with this belief, the reflective portfolio is another strategy that seems to enjoy support among whole language instructors. As a performance-based authentic assessment form, portfolios are regarded as supportive of and compatible with the needs of children with LD, and are also emerging as an assessment option at the college level (Hansen, 1998). According to Hansen (1998),
A portfolio is a multidimensional collection of student's work assembled in an organized fashion and represents the cognitive, affective or psychomotor dimensions of learning. It is a body of work that is selected both by the student and by the teacher and reflected on by students largely through writing. (p. 307)
Reflective writings constitute a major portion of a portfolio and allow a student's skills in writing to be examined over time. Moreover, portfolios may provide writing teachers with an opportunity to integrate writing into the curriculum, allowing students to write for an authentic audience and on meaningful topics, so that they "write to learn" rather than "learn to write" (Hansen, 1998). Portfolios also provide a record of growth over time and allow teachers to recognize the different writing styles of their students. More research needs to be done on how college teachers can use portfolios to teach and evaluate the writing of students with LD/WD.
Other accommodations and support. In addition to the preceding writing strategies and instructional approaches used to help college students with LD/WD, classroom modifications and accommodations, or bypass strategies as Gaskins (1995) calls them, are also very necessary and important for this population (Gaskins, 1995; Ives, 1990). According to the Auxiliary Aids and Services for Postsecondary Students with Handicaps: Higher Education's Obligations Under Section 504, college students with LD should be accommodated with such auxiliary aids and services as taped texts, note takers, interpreters, words processors, and speech synthesis and recognition computer systems (Houston, 1994).
Besides the auxiliary aids and services mentioned above, Martin (1991) provided a comprehensive list of classroom accommodations to help college writers with LD/WD, ranging from advice on specific facets of classroom life (such as lectures, assignment directions and requirements, class discussions and activities, responses to papers, grading, tests and in-class writing) to advocating general principles (such as being patient, flexible, and empathetic). Above all, extra time should be allowed for assignments, especially in-class writing or quizzes, as students with LD/WD tend to become anxious under timed situations (Corrigan, 1997; Gaskins, 1995; Martin, 1991). Depending on their specific difficulty, students could be allowed to take written tests orally because they can dictate a more coherent text than they can produce writing by hand. In addition, written assignments can be broken down into manageable parts, or the number required can be cut. This does not mean, however, that the student's workload is reduced; rather, he or she needs much more time and effort in order to reach the same level of excellence as their peers (Corrigan, 1997). Finally, Martin (1991) believed that college students with LD/WD have long suffered failure and humiliation with regard to their writing, and therefore require extensive positive feedback, patience, understanding, and encouragement from instructors and peers.
DISCUSSION AND DIRECTIONS FOR FUTURE RESEARCH
Funded by the PEEL grant through the University of Arizona, the authors conducted a comprehensive search of the literature on college students with writing disabilities. This article has synthesized the literature of the last decade from the learning disabilities and composition/writing fields that deals with the writing issues of college students with learning disabilities. The remainder of this article will address research issues related to college students with LD/WD and provide directions for future research.
First, researchers tend to use the term "learning disability" homogeneously to describe college students with different specific learning disabilities such as reading, writing, listening, and math. In some of the quantitative studies, although the researchers mentioned that their sample was composed of students with specific learning disabilities in language and/or abstract concept formation, when they presented their results, they neglected the heterogeneity of the sample and failed to discuss whether or not the different nature of the learning difficulties had contributed to the research results.
Second, when Hughes and Smith (1990) published their synthesis of the literature on the cognitive and academic performance of college students with LD a decade ago, they pointed out that the study sample might not be representative of this student population since different measurement instruments and identification procedures of students with LD were used. Unfortunately, this is still the case 10 years later. For example, different scales and scores have been used to evaluate the writing performance of college students with LD/WD, such as holistic assessment of essays with multiple levels of performance and standard scores of W-J and TOAL (Morris-Friehe & Leuenberger, 1992; Plata et al., 1995). To ensure reliability and compatibility across studies, constant standards are necessary, either with regard to the sample selection or the instruments used to measure writing performance. Also, as mentioned, the same group of researchers has carried out all the studies in certain areas of learning disabilities (e.g., the effectiveness of AT). Thus, more studies are warranted in different contexts conducted by different researchers using similar or different research methods and procedures so that research results can be more accurately interpreted and compared, ultimately leading to more diverse understandings of a topic.
Third, AT has been found effective in some studies to help students with spelling, grammar, and the initial steps of writing, such as outlining ideas. More advanced AT such as speech recognition and synthesis technologies can dictate or read out the writer's text, taking advantage of the superior oral communication skills of many college writers with LD/WD and freeing them from mechanical concerns in the process of content development. However, AT cannot replace human help in either the mechanical or the content aspects of writing since it cannot detect contextual errors and has limited capabilities in improving the content and organization of writing. Future education software developers should direct their efforts towards improving the content, organization, and coherence of writing. Furthermore, more empirical research needs to be conducted to determine which AT interventions are most effective for certain types of writing difficulties (Hughes & Smith, 1990), and even what the effectiveness is of AT relative to a particular type of writing (such as narrative, expressive, or analytical writing).
Fourth, due to increasing numbers of students with learning disabilities attending postsecondary institutions, researchers, writing instructors, and writing specialists have been actively developing and implementing university support programs for students with specific writing disabilities. However, college writing is not limited to writing labs and composition classes. It is a demanding task that occurs in every single academic course. Thus, it is essential that instructors and faculty members across academic disciplines be aware of the special needs of students with LD/WD and the instructional strategies and approaches that can accommodate their needs, and adjust their methods of assessing the writing needs and performance of these students accordingly (Scott, 1991). To that end, comprehensive training should be provided to faculty members from different academic areas regarding the nature and features of writing disabilities as well as effective instructional strategies. Such training can take the form of workshops or handouts and pamphlets, which can be prepared and presented by the staff of the campus LD support program (Scott, 1991). With such training, hopefully faculty members would not make the legal and practical mistakes described by Houston (1994), who pointed out that some faculty in her college, out of ignorance about the nature of learning disabilities and related legal requirements, were questioning the fairness of accommodations to a student with LD. The special training programs described by Scott (1991) should serve as a model to educate and change the mindset of professionals who are closely involved in students' academic work and who have an immense impact on their psychosocial well being.
Fifth, 10 years ago, Hughes and Smith (1990) expressed their disappointment over the lack of empirical studies of the effectiveness of a particular instructional strategy, saying that "it is surprising that more applied research has not been carried out in the more than 20 years during which professionals have begun to focus on the needs of college students with learning disabilities" (p. 77). Unfortunately, a decade later, the current review of the relevant literature reveals exactly the same situation. That is, among the 21 articles on instructional support for college students with LD/WD published in the last 10 years, no empirical study has investigated the effectiveness of the various instructional strategies and approaches. The 21 articles reviewed here, as did those of 10 years ago, mainly provide lists of "what to dos" and "how to dos," recommendations, and the outcome of a general program or a case study. Thus, empirical studies are desperately needed about LD/WD in postsecondary schools, especially in the area of instructional support. Several specific research areas are of particular importance. For example, a dispute has been going on between researchers, LD practitioners, and theorists about the relative effectiveness of two approaches to writing instruction, the traditional skills-centered approach and the process approach. The former emphasizes solid subskills, to be acquired by students so that they can write better. The teacher evaluates the writing product to determine whether it is acceptable. The latter approach stresses the process of writing, proposing that the subskills are not to be taught discretely in designated instructional sessions, but be integrated within the context of the writing process. In addition, the evaluation of the writing focuses on student effort and engagement in the writing and on student progress over time (Gersten et al., 1999). Empirical studies should be conducted to evaluate the effectiveness of the two approaches for the classroom writing of college students with LD/WD.
Empirical research should also examine the instructional strategies used to improve the writing content of college students with LD/WD. Gersten et al. (1999) noted that more articles are published on the mechanical problems of student writing, such as capitalization, punctuation, and spelling, than on higher-order concerns, such as organization and coherence of the content. Gersten et al. (1999) recommended that content receive priority since it capitalizes on the strengths of many individuals with LD.
Sixth, while some researchers kept emphasizing the need for intense individual diagnosis of students so that appropriate individualized strategies may be created, little attention has been paid to the needs of college writing teachers, who often are overwhelmed with several classes of 30 or more students. Many practitioners are now using the instructional strategies promoted by the authors summarized in the effective instructional approaches and methods section of this article (peer reviews, portfolios for assessment, collaborative learning, student-centered classrooms, the process approach to writing, and explicit teaching of subtasks within that process)--yet no studies have assessed how well these approaches work with students with LD/WD, especially college students. How can the call for an integrative, collaborative learning environment be balanced with so many individualized needs assessments and instructional interventions? Can teachers truly engage in this level of support, given teaching loads?
Further, how does the increasing popular concept of universal design fit into this discussion? Universal design encourages constructing classrooms so that all students, regardless of disability, learning style, cultural and socioeconomic factors, can engage equally in learning environments. Can writing classrooms be universally designed? Or will educators always be working to "compensate" for students with LD/WD in these situations? Does more than just how writing is taught need to change? Does what gets taught need to be reconsidered as well? That is, is writing as it is currently conceived the only valid way of processing and expressing university level knowledge?
Seventh, it remains eminently necessary to study the cognitive process and difficulties of students with LD/WD in order to more effectively help them improve their writing. For example, Gregg et al. (1991) pointed out that "referencing" is a precursor to "disorders" in that it establishes an empirical base for a descriptive discussion of the relationship between a writer's sense of audience and textual structures (specifically, "textual referencing," i.e., the ability to separate the roles of writer and audience in order to make clear references to noun antecedents so that readers can follow what a writer is referring to as they make a point). Knowing what cognitive difficulties underlie sentence-level errors such as pronoun references is vital for postsecondary-level instructors to understand so that they will be able to distinguish between a student's problems understanding audience and possible reading or oral language difficulties. Without this fine-tuned knowledge, successful interventions cannot be made in terms of either referring students to appropriate support services or creating effective instructional interventions.
Gregg et al. wrote in 1991 that empirical research has not kept up with practitioners' needs in this area. Based on the current literature review, it still has not. More research is needed to identify the cognitive processes involved in sentence and text production, so that individual diagnoses can be done more accurately and instructional interventions can focus on the real issues underlying written errors. Gregg et al. address individual diagnosis as the best way to accurately diagnose and assist students; the implications for large college classrooms now need to be pursued.
Eighth, although collaborative learning is regarded as essential in student learning (Ives, 1990; Pardes & Rich, 1996; Stratcher, 1993), there are gaps in our knowledge of how effectively students with LD/WD collaborate with average writers. In the peer evaluation model described earlier, students with LD/WD were in the position of being tutored, helped, and evaluated, with the unfortunate implication that they are not able to evaluate their peers' writing due to their own writing difficulties. Their role in the collaborative learning situation is a passive one; in fact, the process is not collaborative at all, since the student with writing difficulties mostly receives comments and feedback from the peer evaluator. What is more, there is no evidence in the model to show that the student with LD/WD is contributing to the other person's writing. In future practice and research, it is essential that the true collaborative role of the college student with LD/WD be recognized, enhanced, and expanded. It is when students with LD/WD feel that they are contributing to the peer evaluation process that they become more confident writers.
Next, in a more general sense, one practical goal of writing labs and writing instruction is to help students build up writing skills that they can use in their chosen academic areas. Thus, future practice and research needs to be directed at the transfer of writing skills and writing strategies learned in writing classes to other content areas, and thus ultimately to the overall academic improvement and achievement of these students.
Lastly, most of the researchers cited here have been focusing on help and assistance for students with LD/WD from external sources, LD specialists, AT, and faculty. Very few studies have examined the internal factors that directly or indirectly affect the writing performance of students with LD/WD, such as their fears and related anxieties, their motivation, perseverance, and internal locus of control (Higgins & Zvi, 1996). While external assistance is important to these students, emotional support may be all the more essential. Consequently, more research effort should be made in this area.
List of Journals Searched for the Literature Review
1. Academic Therapy Quarterly (LC4001 A25)
2. ADE Bulletin (PE1001 A3)
3. Adult Education Quarterly (LC5201 A33)
4. Adult Learning (LC5201 L53)
5. American Educational Research Journal (L11 .A66)
6. Annals of Dyslexia (J496 A5 078)
7. Assessing Writing (PE1404 A86)
8. B.C. Journal of Special Education Learning (LC 3961 B35)
9. Cognition and Instruction (LB1050 C3416)
10. College Composition and Communication (PE 1001 C6)
11. College English (PE1 C6) (88-89)
12. College Teaching (L11 I4)
13. Community/Junior College Research Quarterly (LB2328 .C685)
14. Community College Review (LB2328 C69)
15. Computer and Composition (LB1028.5 C6)
16. The Computing Teacher (LB1028 5A 1C65)
17. Educational Research (L16 E38)
18. English Journal (PE1 E5))
19. English Language Teaching Journal (PE1128.A2 ES)
20. Exceptional Children (LC3951 J6)
21. Higher Education Review (L11 C77)
22. Higher Research Quarterly (L11 E11)
23. Intervention in School and Clinic (LC4001 A25)
24. Journal of Adolescent & Adult Literacy (LB1050 J6)
25. Journal of Basic Writing (PE1404 J68)
26. Journal of Developmental Education (LB2331 J68)
27. Journal of Education (L11 J5)
28. Journal of Educational Psychology (L11 J8)
29. Journal of Educational Research (l11 j75)
30. Journal of Higher Education (L11 J78)
31. Journal of Learning Disabilities (LC3951 J64)
32. Journal of Psycho-Educational Assessment (LB 1051 J68) (1990-1993)
33. Journal of Research and Development in Education (LB1028.A1 J69)
34. Language Learning (P1 L33)
35. Journal of Special Education (LC4001 J6)
36. Journal of Special Education Technology (LC4023 J68)
37. Journal of Teaching Writing (PE1404 J69)
38. LD Forum
39. Learning Disabilities; a Multidisciplinary Journal
40. Learning Disabilities Research & Practice (LC4704 L443)
41. Learning Disabilities Focus (LC4704 L43)
42. Learning Disability Quarterly (LC4704 L44)
43. Learning Disabilities Research (LC4704 L442)
44. Psychology in the Schools (LB1101.P75)
45. Reading Improvement (LB1362 R4)
46. Reading and Writing (BF456.R2 R338)
47. Research in the Teaching of English (PE1001 R47)
48. Research & Teaching in Developmental Education (LB2343.4.R48)
49. Teachers & Writers (LB1576 T347)
50. Reading Research Quarterly (LB1950 R42)
51. Reading Teacher (Lb1050.5.J6)
52. Remedial and Special Education (LC3950 E17)
53. Research in Higher Education (LB2300 R4)
54. Review of Educational Research (L11 .R35)
55. School Psychology Review (LB1051 S373)
56. Scientific American (T1 S5)
57. Studies in Higher Education (LB2300 S8) (1990-1992)
58. Teacher Education and Special Education (LC4019 T42)
59. Teaching English in the Two-Year College (PE 1065 T4)
60. Teaching English Topics in Language Disorders (RC423 T7)
61. Teaching Exceptional Children (LC3950 T4)
62. TESOL Quarterly (PE1128 T2)
63. Topics in Learning and Learning Disabilities (LC4704 T6)
64. The Writing Center Journal (PE1404 W7)
65. The Writing Instructor (PE1128 A2 W7) (1990-1993)
66. Writing Program Administrator (PE1404 W18)
67. Written Communication (PE 1404 W 75)
Table 1 Four Categories of Articles Article Author(s)'s Name and Year of Publication Number of Categories Articles Overview of the Hunt-Bert, M., & Rankin, J. L. (1994) 6 available De La Paz, S. (1999) assistive Meyer, A., Pisha, B., & Rose, D. (1991) technology Raskind, M. H., & Higgins, E. L. (1998) Raskind, M. H., & Scott, N. G. (1993) Speziale, M. (1993) Empirical Collins, T. (1990) 4 studies of the Higgins, E., & Raskind M. H. (1995) effectiveness Higgins, E. L., & Zvi, J. C. (1996) of assistive Raskind M. H., & Higgins, E. (1995) technology Characteristics Gregg, N., & Hoy, C. (1990) 6 and error Gregg, N., Hoy, C., McAlexander, P., patterns & Hayes, C. (1991) Leuenberger, J., & Morris, M. (1990) Morris-Friehe, M., & Leuenberger, J. (1992) Plata, M., Zelhart, P., & House, G. (1995) Smith, J. O. (1993) Instructional Ammer, J. J. (1998) 22 methods Bardine, B. (1997) Corrigan, J. R. (1997) El-Hindi, A. E. (1997) Gaskins, J. C. (1995) Gersten, R., Baker, S., & Edwards, L. (1999) Gould, B. W. (1991) Graham, S., Harris, K. R., MacArthur, C., & Schwartz, S. (1991a) Graham, S., Harris, K. R., MacArthur, C. A., & Schwartz, S. (1991b) Hansen, B. (1998) Houston, L. S. (1994) Isaacson, S., & Gleason, M. M. (1997) Ives, N. R. (1990) Keefe, C. H., & Keefe, D. R. (1993) Mangenello, R. E. (1994) Martin, J. L. (1991) Neff, J. (1994) Pardes, J. R., & Rich, R. Z. (1996) Perin, D. (1990) Scott, S. S. (1991) Sills, C. K. (1995) Stracher, D. A. (1993) Table 2 Types of Articles Type of Articles Name of Authors and Year Published Number of Articles Research Report 1. Assistive Technology 10 Collins, T. (1990) Higgins, E., & Raskind M. H. (1995) Higgins, E. L., & Zvi, J. C. (1996) Raskind M. H., & Higgins, E. (1995) 2. Characteristics and Error Patterns Gregg, N., & Hoy, C. (1990) Gregg, N., Hoy, C., McAlexander, P., & Hayes, C. (1991) Leuenberger, J., & Morris, M. (1990) Morris-Friehe, M., & Leuenberger, J. (1992) Plata, M., Zelhart, P., & House, G (1995) Smith, J. O. (1993) Literature Overview of Assistive Technology 2 Review De La Paz, S. (1999) Raskind, M. H., & Higgins, E. L. (1998) Program Instruction Methods 2 Description Graham, S., Harris, K. R., MacArthur, C., & Schwartz, S. (1991a) Pardes, J. R., & Rich, R. Z. (1996) Informational 1. Overview of Assistive Technology 24 Hunt-Bert, M., & Rankin, J. L. (1994) Raskind, M. H., & Scott, N. G. (1993) 2. Instruction Methods Ammer, J. J. (1998) Bardine, B. (1997) Corrigan, J. R. (1997) El-Hindi, A. E. (1997) Gaskins, J. C. (1995) Gersten, R., Baker, S., & Edwards, L. (1999) Gould, B. W. (1991) Graham, S., Harris, K.R., MacArthur, C., & Schwartz, S. (1991b) Hansen, B. (1998) Houston, L. S. (1994) Isaacson, S., & Gleason, M. M. (1997) Ives, N. R. (1990) Keefe, C. H., & Keefe, D. R. (1993) Manganello, R. E. (1994) Martin, J. L. (1991) Meyer, A., Pisha, B., & Rose, D. (1991) Neff, J. (1994) Perin, D. (1990) Scott, S. S. (1991) Sills, C. K. (1995) Speziale, M. (1993) Stracher, D. A. (1993)
Ammer, J. J. (1998). Peer evaluation model for enhancing writing performance of students with learning disabilities. Reading & Writing Quarterly, 14, 263-283.
Americans with Disabilities Act of 1990, P.L. 101-336, 42 U.S.C.A. 12, 101-12, 213 (West Supp. 1991).
Bardine, B. (1997). Working with learning disabled writers: Some perspectives. Research to practice (Report No. 039-0200-0019). Kent, OH: Ohio Literacy Resource Center. (ERIC Document Reproduction Service No. ED406554)
Collins, T. (1990). The impact of microcomputer word processing on the performance of learning disabled students in a required first-year writing course. Computers and Composition, 8, 49-67.
Communication Skillbuilders. (1988). Cue Write [Computer program]. Tucson, AZ: Author.
Corrigan, J. R. (1997). Teaching writing to dyslexic students: A guide for the composition instructor. Teaching English in the Two-Year College, 24, 205-211.
De La Paz, D. S. (1999). Composing via dictation and speech recognition systems: Compensatory technology for students with learning disabilities. Learning Disability Quarterly, 22, 173-182.
Don Johnston Developmental Equipment, Inc. (1992). CoWriter [Computer program]. Wauconda, IL: Author.
Duques, S. L. (1989). Grammatical deficiencies in writing: An investigation of learning disabled college students. Reading and Writing: An Interdisciplinary Journal, 1, 309-325.
Gajar, A. (1989). A computer analysis of written variables and a comparison of compositions written by university students with and without learning disabilities. Journal of Learning Disabilities, 22, 125-130.
Ganschow, L. (1984). Analysis of written language of a learning disabled college student and instructional implications. Annals of Dyslexia, 43, 271-277.
Gaskins, J. C. (1995). Teaching writing to students with learning disabilities: The landmark method. Teaching English in the Two-Year College, 24, 116-122.
Gersten, R., Baker, S., & Edwards, L. (1999). Teaching expressive writing to students With learning disabilities. ERIC/OSEP Digest # 590.
Gould, B. W. (1991). Curricular strategies for written expression. In A. M. Bain, L. L. Bailet, & L. C. Moats (Eds.), Written language disorders theory into practice (pp. 129-164). Austin, TX: Pro-Ed.
Graham, S., Harris, K. R., MacArthur, C., & Schwartz, S. (1991a). Writing and writing instruction for students with learning disabilities: Review of a research program. Learning Disability Quarterly, 14, 89-114.
Graham, S., Harris, K. R., MacArthur, C., & Schwartz, S. (1991b). Writing instruction. In B.Y.L. Wong (Ed.), Learning about learning disabilities (pp. 309-343). San Diego: Academic Press, Inc.
Gregg, N. (1983). College learning disabled writers: Error patterns and instructional alternatives. Journal of Learning Disabilities, 6, 334-338
Gregg, N., Hoy, C., & Sabal, R. (1988). Spelling error patterns of normal, learning-disabled, and underprepared college writers. Journal of Psycho-Educational Assessment, 6, 14-23.
Gregg, N., & Hoy, C. (1990). Referencing: The cohesive use of pronouns in the written narrative of college underprepared writers, nondisabled writers, and writers with learning disabilities. Journal of Learning Disabilities, 23, 557-563.
Gregg, N., Hoy, C., Mcalexander, P., & Hayes, C. (1991). Written sentence production error patterns of college writers with learning disabilities. Reading and Writing: An Interdisciplinary Journal, 3, 169-185.
Hallenbeck, M. J. (1996). The cognitive strategy in writing: Welcome relief for adolescents with learning disabilities. Learning Disabilities Research & Practice, 11, 107-119
Hammill, D. D., Brown, V. L., Larsen, S. C., & Wiederholt, J. L. (1980). Test of Adolescent Language (TOAL). Austin, TX: Pro-Ed. Hansen, B. (1998). Using reflective portfolios as a tool to teach writing to students with learning disabilities: A project for pre-service teachers. Reading & Writing Quarterly, 14, 307-408.
Higgins, E. L., & Raskind, M. H. (1995). Compensatory effectiveness of speech recognition on the written composition performance of postsecondary students with learning disabilities. Learning Disability Quarterly, 18, 159-174.
Higgins, E. L., & Zvi, J. C. (1996). Assistive technology for postsecondary students with learning disabilities: From research to practice. Annals of Dyslexia, 45, 123-142.
Houston, L. S. (1994). Accommodations for learning differences in the English classroom. Teaching English in the Two-Year College, 21, 297-301
Hughes, C. A., & Smith, J. O. (1990). Cognitive and academic performance of college students with learning disabilities: A synthesis of the literature. Learning Disability Quarterly, 13, 66-79.
Hunt-Bert, M., & Rankin, J. L. (1994). Ponder the possibilities: Computer-supported writing for struggling writers. Learning Disabilities Research & Practice, 9, 169-179.
Individuals with Disabilities Education Act of 1990, P.L. 101-476 (1993). Institute on Applied Technology. (1991). Write Away [Computer program]. Boston: Author.
Isaacson, S., & Gleason, M. M. (1997). Mechanical obstacles to writing: What can teachers do to help students with learning disabilities. [On-Line]. Available: http://www.ldonline.org/ld_indepth/ writing/isaacson_obstacles.html
Ives, N. R. (1990). The learning disabled student in the college writing class (Classroom teaching guide; position paper. Clearinghouse No. CS212722). Geneseo: State University of New York. (ERIC Document Reproduction Service No. ED329972)
Keefe, C. H., & Keefe, D. R. (1993). Instruction for students with LD: A whole language model. Intervention in School and Clinic, 28, 172-177.
Kupersmith, A. (1990). A three-tier approach for helping dysgraphic writers. In D. Perin (Ed.), Instructing students with learning disabilities: Guidelines for community college faculty and support staff. Conference proceedings (pp. 67-70). New York: City University of New York Institute for Research and Development in Occupational Education.
Johnson, D., & Blalock, J. (1987). Adults with learning disabilities and clinical studies. Orlando, FL: Grune and Stratton.
Leuenberger, J., & Morris, M. (1990). Analysis of spelling errors by learning disabled and normal college students. Learning Disabled Focus, 5, 103-118.
Litowitz, B. (1981). Developmental issues in written language. Topics in Language Disorders, 1, 71-89.
Manganello, R. E. (1994). Writing and the LD student. Journal of Teaching Writing, 12, 235-245.
Martin, J. L. (1991). Removing the stumbling blocks: 25 ways to help our learning disabled college writers. Teaching English in the Two-year College, 19, 283-289.
Martin, N., D'Arcy, P., Newton, B., & Parker, R. (1976). Writing and learning across the curriculum. London: Ward Lock Education for the Schools Council.
Meyer, A., Pisha, B., & Rose, D. (1991). Process and product in writing: Computer as enabler. In A. M. Bain, L. Y. Bailet, & L. C., Moats (Eds.), Written language disorders: Theory into practice (pp. 99-128). Austin, TX: Pro-Ed.
Morris-Friehe, M., & Leuenberger, J. (1992). Direct and indirect measures of writing for non-learning disabled and learning disabled college students. Reading and Writing: An Interdisciplinary Journal, 4, 281-296.
Neff, J. (1994). Learning disabilities and the writing center. In J. A. Mullin & R. Wallace (Eds.), Intersections: Theory-practice in the writing center (pp. 81-95). Urbana, IL: National Council of Teachers of English.
Norton, S. M. (1992). Postsecondary learning disabled students: Do their study habits differ from those of their nondisabled peers? Community/Junior College Quarterly, 16, 105-115.
Office of Special Education Programs. (1992). Fourteenth Annual Report to Congress on The Implementation of the Education of the Handicapped Act. Washington, DC: U.S. Government Printing Office.
O'Hearn, C. (1989). Recognizing the learning disabled college writer. College English, 51, 294-304.
Pardes, J. R., & Rich, R. Z. (1996). Teaching writing to college students with learning disabilities. Intervention in School and Clinic, 31, 297-302.
Perin, D. (1990). The writing of community college students: Comparison of students with and without self-reported learning disabilities. In D. Perin (Ed.), Instructing students with learning disabilities: Guidelines for community college faculty and support staff. Conference proceedings (pp. 71-75). New York: City University of New York Institute for Research and Development in Occupational Education.
Plata, M., Zelhart, P., & House, G. (1995). Comparative writing performance of college students with and without learning disabilities. Journal of Research and Development in Education, 29, 20-26.
Raskind, M. H., & Higgins, E. L. (1995). Effects of speech synthesis on the proofreading efficiency of postsecondary students with learning disabilities. Learning Disability Quarterly, 18, 141-157.
Raskind, M. H., & Higgins, E. L. (1998). Assistive technology for postsecondary students with learning disabilities: An overview. Journal of Learning Disabilities, 31, 27-40.
Raskind, M. H., & Scott, N. G. (1993). Technology for postsecondary students with learning disabilities. In S. A. Vogel & P. B. Adelman (Eds.), Success for college students with learning disabilities (pp. 240-279). New York: Springer-Verlag.
Scholastic Software. (1991a). Process Writer [Computer program]. New York: Author.
Scott, S. S. (1991). College writing labs: Are they meeting the needs of students with LD? Intervention in School and Clinic, 26, 170-174.
Sills, C. K. (1995). Success for learning disabled writers across the curriculum. College Teaching, 43, 66-71.
Smith, J. O. (1993). Self-reported written language difficulties of university students with learning disabilities. Journal of Postsecondary Education and Disability, 10, 1-10.
Speziale, M. (1993). When is APA not APA? When it stands for alternative presentations accepted. Preventing School Failure, 38, 16-17.
Speziale, M., & LaFrance, L. (1992). Multimedia and students with learning disabilities: The road to success. The Computer Teacher, 20, 31-34.
Stracher, D. A. (1993). Providing strategies for learning disabled college students; continuous assessment in reading, writing, and reasoning. Research & Teaching in Developmental Education, 10, 65-77.
Technology-Related Assistance for Individuals with Disabilities Act of 1988, P.L. 100-47, 29 U.S. S. 2201, 2202 (1988).
United States Department of Education. (1991). Auxiliary aids and services for postsecondary students with handicaps. Washington: General Printing Office.
Vogel, S., & Moran, M. R. (1982). Written language disorders in learning disabled college students: A preliminary report. In W. Cruickshank & J. Lerner (Eds.), Coming of age, Vol. 3, The best of ACLD (pp. 15-30). Syracuse, NY: Syracuse University Press.
Wechsler, D. (1981). Manual for the Wechsler Adult Intelligence Scale-Revised. San Antonio, TX: The Psychological Corporation.
White, W.J., Alley, G. R. Deshler, D. D., Shumaker, J. B., & Warner, M. M. (1982). Are there learning disabilities after high school? Exceptional Children, 29, 273-274.
Woodcock, W., & Johnson, M. (1977). Woodcock-John Psycho-Educational Battery Achievement Tests (WJ). Allen, TX: DLM Teaching Resources
Wordstar. (1990). Correct Grammar [Computer program]. Cameron Park, CA: Author.
Requests for reprints should be addressed to: Huijun Li, 5180 Hildring Dr. E., Apt. 254, Fort Worth, TX 76132.
HUIJUN LI, M.A., Ph.D. candidate, is research associate, University of Arizona.
CHRISTINE M. HAMEL, M.M., is writing skills coordinator, University of Arizona.