Part III Research Insights In universities and laboratories researchers seek clues that will help to define dyslexia, its manifestations, and components, and thus to address it more effectively. A number of current studies explore the roles of gender and family, some seek precursors and causal factors, while others compare performance in a variety of ways. Historically the focus in dyslexia studies has been primarily on the young child--preschool through the elementary grades. In the past decade researchers have undertaken studies dealing with older students and adults. While the number of such studies is not large, it is growing, as is awareness that dyslexia and other language learning disorders are life-long conditions and that, contingent on intelligence, experience, and motivation, they may remain undiscovered until later life. Of particular interest to those concerned with older students and adults is the issue of college achievement of dyslexics. Susan A. Vogel and Pamela B. Adelman provide some useful data in "Extrinsic and Intrinsic Factors in Graduation and Academic Failure among LD College Students." Matching LD students with nondisabled peers, the authors compared the groups with respect to high school preparation and performance, ACT and college performance, and failure and graduation rates. They found that oral language ability and motivation and attitude were the factors which differentiated the two groups, accounting for 60 percent of the variance in graduation rate. M. Jo. Worthy and Marcia Invernizzi, in "Spelling Errors of Normal and Disabled Students on Achievement Levels One through Four: Instructional Implications," report on kinds of spelling errors, using discriminant function analysis. They studied 72 disabled and 90 normal spellers. They conclude that disabled spellers who receive phonicsbased instruction achieve reasonable proficiency at lower-level spelling, but require a program emphasizing meaning, orthography, and derivations at Levels 2 and above. "Prediction of Dyslexia in Kindergarten Boys," reviews a recent 117
118
RESEARCH INSIGHTS
study by Nathlie A. Badian, Gloria B. McAnulty, Frank H. Duffy, and Heidelise Als. The researchers administered a number of neuropsychological and preacademic tests, as well as BEAM to 163 kindergarten boys. When these boys reached grade four, their reading test scores were compared with the earlier results in an attempt to identify precursors of dyslexia. Among other conclusions the authors reached, they find strong evidence of the importance of phonological processing ability as a dyslexia precursor. "Fine-Grained Auditory Discrimination and Performance on Tests of Receptive Vocabulary and Receptive Language," by Lois L. Elliott, Michael A. Hammer, and Margo E. Scholl, describes a study which included 175 children with normal school achievement and 189 children who demonstrated some degree of language-learning difficulty. The authors found that fine-grained auditory discrimination is a significant predictor of vocabulary and language performance of children in the age range of 8-11. The importance of such discrimination for younger children, particularly preschoolers, is discussed, as well as the instruction and environmental implications. Patricia A. Szeszulski and Franklin R. Manis sought to determine the extent to which dyslexics and their parents shared deficits in word recognition skills. "An Examination of Familial Resemblance Among Subgroups of Dyslexics" reports their findings based on a sample of 40 dyslexic children and their biologic parents and 105 normal readers. They discuss four patterns of deficit, concluding that phonological deftcit is that most commonly shared by dyslexics and their parents. Che Kan Leong, Donald R. Simmons, and MaryAnne IzattGambell present the results of two interrelated studies which emphasized learning words, learning about words, learning prose materials, and learning about prose materials in "The Effect of Systematic Training in Elaboration on Word Meaning and Prose Comprehension in Poor Readers." This study indicates the usefulness of direct explicit teaching which includes discussion and feedback as a means of enhancing the achievement of poor readers. J.C. De Fries, Sally J. Wadsworth, and Jacquelyn J. GiUis, in "Gender Differences in Cognitive Abilities of Reading-Disabled Twins," report the latest findings in the Colorado Twin Study. The authors studied samesex twin pairs with at least one reading disabled member in each pair, same-sex twin pairs with no history of reading disabilit)~ and oppositesex fraternal twin pairs with at least one reading-disabled member.
Extrinsic and Intrinsic Factors in Graduation and Academic Failure Among LD College Students Susan A. Vogel Northern Illinois University DeKalb, Illinois
Pamela B. AdeIman Barat College Lake Forest, Illinois
A group of 110 LD college students were compared to a random stratified sample (RSS) of 153 peers attending the same moderately selective college between 1980 and 1988. The LD students received comprehensive, highly coordinated support services for at least one semester. The groups were matched on gender, college experience, semester, and year of entry to the college. The LD and RSS groups were compared on high school preparation and performance, ACT and college performance, and graduation and academic failure rate. Although the LD students' high school records, ACT scores, and college performance were inferior to that of the RSS group, they graduated at the same rate and within the same time frame. Neither was there any significant difference in the academic failure rate. Closer examination of the LD graduates and academic failures" performance showed that in spite of the similarities in intellectual abilities, academic achievement, and aptitude-achievement discrepancy, two factors differentiated between the LD graduates and non-graduates: oral language abilities and motivation and attitude toward the teaching-learning process. These two factors accounted for 60 percent of the variance in graduation status.
This research was supported in part by grants from the Butz and Thorn River Foundations. Annals of Dyslexia, Vol. 40, 1990. Copyright © 1990 by The Orton Dyslexia Society ISSN 0736-9387
119
120
RESEAaCrI INsiGrrrs
This research reflects the increased awareness of the chronicity of learning disabilities, the question of long-term prognosis, and the need to determine intervention effectiveness at the postsecondary level. The study is part of a larger descriptive, eight-year follow-up study on educational and employment attainments of a group of selfreferred college students with diagnosed learning disabilities who received LD support services for at least one semester between 1980-88 at Barat College, a small, moderately selective private college that offered comprehensive, highly coordinated support service. This specific program is based on a four-pronged approach to assist students: 1. To use course support as needed (subject-matter tutoring); 2. To improve basic skills deficits through remediation; 3. To access and use appropriate accommodations (e.g., modified examination procedures); and 4. To develop and use compensatory strategies based on self knowledge, i.e., understanding the results of diagnostic testing, and how these relate to their strengths and deficits and learning styles. Several factors combined to fuel our interest in this study. First, LD children identified in the 1970s have grown up. Many (and in increasing numbers) are attending colleges and universities as a result of Section 504 and the increase in available services and accommodations for the handicapped in postsecondary settings. According to the American Council on Education's Freshman Norms for 1988 (Astin et al. 1988), six percent of all full-time/first-time college freshmen reported having at least one disability. This percent is more than twice the number of disabled freshmen a decade ago. Moreover, one percent of the total number of disabled freshmen self-identified as LD and this is a minimal estimate of the total number of LD college freshmen because many LD university students transfer from one PSI to another or take a part-time load and by design they were excluded from the data collection (Astin et al. 1988). The U.S. Department of Education (1987) reported in another study that 10.5 percent of all postsecondary students in the nation are disabled, approximately 1.3 million out of a total of 12.5 million students, including those enrolled in community colleges, professional schools, and doctoral programs (U.S. Dept. of Education, 1987). Of the 1.3 million disabled students, 12.2 percent (approximately 160,000) reported having a learning disability. Section 504 of the Rehabilitation Act of 1973 mandates equal access for handicapped students, including the learning disabled, to institutions of higher education. This law prohibits discrimination in admis-
LD GRADUATIONAND ACADEMIC FAILURE RATES
121
sions and in program accessibility to qu'alified handicapped students, and mandates that accommodations be made to ensure program access. Mandated accommodations require modifications in the ways teachers present information, ways students acquire knowledge, and the means by which faculty assess student learning. Commonly available accommodations include allowing a student to tape record lectures and providing extended time on examinations (Abrams and Abrams 1981; Ballard, Ramirez, and Zantal-Wiener 1987). The specific services are not mandated, but commonly provided are tutoring, priority registration, personal counseling, and notetakers (Vogel 1982; Vogel 1985a). Additional services available on campuses providing comprehensive services include assessment, one-on-one remediation with a certified LD specialist for one to three hours per week, special academic advising, advocacy, developmental courses and workshops in reading, written language, and study skills, and career counseling (See Adelman 1985; Adelman 1988; Adelman and Olufs 1986; Konrad 1984; Vogel 1982, 1987, 1990; Vogel and Adelman 1981). As a result of Section 504, increased pressure from advocates, lack of clarity on the definition of "qualified," perhaps fear of legal battles, dwindling college applicant pools, and the search for new student markets have all resulted in this increase in learning-disabled students attending college. Moreover, along with this significant increase in the last ten years in the number of students with learning disabilities attending colleges and universities, there is also an equal if not greater concern for the academic failure rate of these students. Even with such large numbers of students with learning disabilities enrolling in postsecondary institutions, very little is known about the graduation and failure rate of LD college students in various postsecondary settings. Only one study thus far has addressed this question. Bursuck et al. (1989) sent a survey to the 336 members of the LD Special Interest Group of the Association on Handicapped Student Service Programs in Postsecondary Education (AHSSPPE) (Bursuck, Rose and Cowen 1988). Members of this organization are directors or coordinators of services for handicapped students in two-year and four-year colleges or universities. Out of the total number of 197 .respondents, only 20 (or 10 percent) responded to the question regarding the percentage of students who graduated or completed a course of study. The average graduation or completion rate for those who responded was 30 percent. However, the authors neglected to describe the degree earned (two-year or four-year) or separate out those that reported graduation rate versus completion of a course of study, making interpretation of this information unclear. Neither do we have any substantial information about LD students' high school performance or college entrance examination
122
RESEARCh INSICnTS
scores. Cordoni and Harman (1985) studied the correlation of ACT scores and first semester college freshman gpa (grade point average) for a group of 107 LD college students. They reported that the Social Studies subtest had the highest correlation (r = .30) and accounted for 9 percent of the variance, while for English, Math, and the ACT composite the correlations were significant but low (r =.20, .21, and .23 respectively). In a critical analysis of this study, Vogel (1985b) compared the mean subtest and composite scores for Cordoni's group to a second LD college sample and a national LD sample and noted that though the hierarchies were different in each sample, the means were similar. Vogel (1985b) also questioned Cordoni's and Harman's use of first semester gpa as the criterion measure, since LD college freshmen entering with low ACT scores frequently are advised to take a part-time or minimum full-time load (12 credit hours) consisting of developmental courses, a beginning writing course, and perhaps one heavy reading course, resulting in an inflated gpa. On the other hand, LD freshmen entering with higher ACT scores and performing better on placement tests may be encouraged to take more typical college freshman courses and a full course load (15 credit hours), resulting in poorer grades for the stronger LD students. Vogel, therefore, recommended using college exit gpa as the criterion measure rather than first semester freshman gpa. The present study was, therefore, designed to determine: 1) How do LD college students compare to a random sample of nondisabled college peers on high school preparation and performance, college admissions testing, college performance, and graduation and failure rate? and 2) What are some of the extrinsic and intrinsic factors related to academic success and failure for LD college students?
The Method Subjects A group of 110 LD students was compared to a random stratified sample (RSS) of 153 peers attending the same college for at least one semester between 1980 and 1988. None of the RSS students had been referred for LD support services. All had English as their primary language and were accepted as degree candidates. Students in the LD group were matched with the RSS students on: 1) gender, 2) college experience (first-time freshmen or transfer students), and 3) semester and year-of-entry to the college. Females outnumbered males 2:1 because this school had been a women's college until 1982. Excluded from the LD sample were students with uncorrected visual or auditory
LD GRADUATIONAND ACADEMIC FAILURE RATES
123
impairments, known brain damage, or history of primary emotional disorders. The mean age of the RSS students at entrance to the college was 23 years and they were significantly older (p
118
RESEARCH INSIGHTS
study by Nathlie A. Badian, Gloria B. McAnulty, Frank H. Duffy, and Heidelise Als. The researchers administered a number of neuropsychological and preacademic tests, as well as BEAM to 163 kindergarten boys. When these boys reached grade four, their reading test scores were compared with the earlier results in an attempt to identify precursors of dyslexia. Among other conclusions the authors reached, they find strong evidence of the importance of phonological processing ability as a dyslexia precursor. "Fine-Grained Auditory Discrimination and Performance on Tests of Receptive Vocabulary and Receptive Language," by Lois L. Elliott, Michael A. Hammer, and Margo E. Scholl, describes a study which included 175 children with normal school achievement and 189 children who demonstrated some degree of language-learning difficulty. The authors found that fine-grained auditory discrimination is a significant predictor of vocabulary and language performance of children in the age range of 8-11. The importance of such discrimination for younger children, particularly preschoolers, is discussed, as well as the instruction and environmental implications. Patricia A. Szeszulski and Franklin R. Manis sought to determine the extent to which dyslexics and their parents shared deficits in word recognition skills. "An Examination of Familial Resemblance Among Subgroups of Dyslexics" reports their findings based on a sample of 40 dyslexic children and their biologic parents and 105 normal readers. They discuss four patterns of deficit, concluding that phonological deftcit is that most commonly shared by dyslexics and their parents. Che Kan Leong, Donald R. Simmons, and MaryAnne IzattGambell present the results of two interrelated studies which emphasized learning words, learning about words, learning prose materials, and learning about prose materials in "The Effect of Systematic Training in Elaboration on Word Meaning and Prose Comprehension in Poor Readers." This study indicates the usefulness of direct explicit teaching which includes discussion and feedback as a means of enhancing the achievement of poor readers. J.C. De Fries, Sally J. Wadsworth, and Jacquelyn J. GiUis, in "Gender Differences in Cognitive Abilities of Reading-Disabled Twins," report the latest findings in the Colorado Twin Study. The authors studied samesex twin pairs with at least one reading disabled member in each pair, same-sex twin pairs with no history of reading disabilit)~ and oppositesex fraternal twin pairs with at least one reading-disabled member.
Extrinsic and Intrinsic Factors in Graduation and Academic Failure Among LD College Students Susan A. Vogel Northern Illinois University DeKalb, Illinois
Pamela B. AdeIman Barat College Lake Forest, Illinois
A group of 110 LD college students were compared to a random stratified sample (RSS) of 153 peers attending the same moderately selective college between 1980 and 1988. The LD students received comprehensive, highly coordinated support services for at least one semester. The groups were matched on gender, college experience, semester, and year of entry to the college. The LD and RSS groups were compared on high school preparation and performance, ACT and college performance, and graduation and academic failure rate. Although the LD students' high school records, ACT scores, and college performance were inferior to that of the RSS group, they graduated at the same rate and within the same time frame. Neither was there any significant difference in the academic failure rate. Closer examination of the LD graduates and academic failures" performance showed that in spite of the similarities in intellectual abilities, academic achievement, and aptitude-achievement discrepancy, two factors differentiated between the LD graduates and non-graduates: oral language abilities and motivation and attitude toward the teaching-learning process. These two factors accounted for 60 percent of the variance in graduation status.
This research was supported in part by grants from the Butz and Thorn River Foundations. Annals of Dyslexia, Vol. 40, 1990. Copyright © 1990 by The Orton Dyslexia Society ISSN 0736-9387
119
120
RESEAaCrI INsiGrrrs
This research reflects the increased awareness of the chronicity of learning disabilities, the question of long-term prognosis, and the need to determine intervention effectiveness at the postsecondary level. The study is part of a larger descriptive, eight-year follow-up study on educational and employment attainments of a group of selfreferred college students with diagnosed learning disabilities who received LD support services for at least one semester between 1980-88 at Barat College, a small, moderately selective private college that offered comprehensive, highly coordinated support service. This specific program is based on a four-pronged approach to assist students: 1. To use course support as needed (subject-matter tutoring); 2. To improve basic skills deficits through remediation; 3. To access and use appropriate accommodations (e.g., modified examination procedures); and 4. To develop and use compensatory strategies based on self knowledge, i.e., understanding the results of diagnostic testing, and how these relate to their strengths and deficits and learning styles. Several factors combined to fuel our interest in this study. First, LD children identified in the 1970s have grown up. Many (and in increasing numbers) are attending colleges and universities as a result of Section 504 and the increase in available services and accommodations for the handicapped in postsecondary settings. According to the American Council on Education's Freshman Norms for 1988 (Astin et al. 1988), six percent of all full-time/first-time college freshmen reported having at least one disability. This percent is more than twice the number of disabled freshmen a decade ago. Moreover, one percent of the total number of disabled freshmen self-identified as LD and this is a minimal estimate of the total number of LD college freshmen because many LD university students transfer from one PSI to another or take a part-time load and by design they were excluded from the data collection (Astin et al. 1988). The U.S. Department of Education (1987) reported in another study that 10.5 percent of all postsecondary students in the nation are disabled, approximately 1.3 million out of a total of 12.5 million students, including those enrolled in community colleges, professional schools, and doctoral programs (U.S. Dept. of Education, 1987). Of the 1.3 million disabled students, 12.2 percent (approximately 160,000) reported having a learning disability. Section 504 of the Rehabilitation Act of 1973 mandates equal access for handicapped students, including the learning disabled, to institutions of higher education. This law prohibits discrimination in admis-
LD GRADUATIONAND ACADEMIC FAILURE RATES
121
sions and in program accessibility to qu'alified handicapped students, and mandates that accommodations be made to ensure program access. Mandated accommodations require modifications in the ways teachers present information, ways students acquire knowledge, and the means by which faculty assess student learning. Commonly available accommodations include allowing a student to tape record lectures and providing extended time on examinations (Abrams and Abrams 1981; Ballard, Ramirez, and Zantal-Wiener 1987). The specific services are not mandated, but commonly provided are tutoring, priority registration, personal counseling, and notetakers (Vogel 1982; Vogel 1985a). Additional services available on campuses providing comprehensive services include assessment, one-on-one remediation with a certified LD specialist for one to three hours per week, special academic advising, advocacy, developmental courses and workshops in reading, written language, and study skills, and career counseling (See Adelman 1985; Adelman 1988; Adelman and Olufs 1986; Konrad 1984; Vogel 1982, 1987, 1990; Vogel and Adelman 1981). As a result of Section 504, increased pressure from advocates, lack of clarity on the definition of "qualified," perhaps fear of legal battles, dwindling college applicant pools, and the search for new student markets have all resulted in this increase in learning-disabled students attending college. Moreover, along with this significant increase in the last ten years in the number of students with learning disabilities attending colleges and universities, there is also an equal if not greater concern for the academic failure rate of these students. Even with such large numbers of students with learning disabilities enrolling in postsecondary institutions, very little is known about the graduation and failure rate of LD college students in various postsecondary settings. Only one study thus far has addressed this question. Bursuck et al. (1989) sent a survey to the 336 members of the LD Special Interest Group of the Association on Handicapped Student Service Programs in Postsecondary Education (AHSSPPE) (Bursuck, Rose and Cowen 1988). Members of this organization are directors or coordinators of services for handicapped students in two-year and four-year colleges or universities. Out of the total number of 197 .respondents, only 20 (or 10 percent) responded to the question regarding the percentage of students who graduated or completed a course of study. The average graduation or completion rate for those who responded was 30 percent. However, the authors neglected to describe the degree earned (two-year or four-year) or separate out those that reported graduation rate versus completion of a course of study, making interpretation of this information unclear. Neither do we have any substantial information about LD students' high school performance or college entrance examination
122
RESEARCh INSICnTS
scores. Cordoni and Harman (1985) studied the correlation of ACT scores and first semester college freshman gpa (grade point average) for a group of 107 LD college students. They reported that the Social Studies subtest had the highest correlation (r = .30) and accounted for 9 percent of the variance, while for English, Math, and the ACT composite the correlations were significant but low (r =.20, .21, and .23 respectively). In a critical analysis of this study, Vogel (1985b) compared the mean subtest and composite scores for Cordoni's group to a second LD college sample and a national LD sample and noted that though the hierarchies were different in each sample, the means were similar. Vogel (1985b) also questioned Cordoni's and Harman's use of first semester gpa as the criterion measure, since LD college freshmen entering with low ACT scores frequently are advised to take a part-time or minimum full-time load (12 credit hours) consisting of developmental courses, a beginning writing course, and perhaps one heavy reading course, resulting in an inflated gpa. On the other hand, LD freshmen entering with higher ACT scores and performing better on placement tests may be encouraged to take more typical college freshman courses and a full course load (15 credit hours), resulting in poorer grades for the stronger LD students. Vogel, therefore, recommended using college exit gpa as the criterion measure rather than first semester freshman gpa. The present study was, therefore, designed to determine: 1) How do LD college students compare to a random sample of nondisabled college peers on high school preparation and performance, college admissions testing, college performance, and graduation and failure rate? and 2) What are some of the extrinsic and intrinsic factors related to academic success and failure for LD college students?
The Method Subjects A group of 110 LD students was compared to a random stratified sample (RSS) of 153 peers attending the same college for at least one semester between 1980 and 1988. None of the RSS students had been referred for LD support services. All had English as their primary language and were accepted as degree candidates. Students in the LD group were matched with the RSS students on: 1) gender, 2) college experience (first-time freshmen or transfer students), and 3) semester and year-of-entry to the college. Females outnumbered males 2:1 because this school had been a women's college until 1982. Excluded from the LD sample were students with uncorrected visual or auditory
LD GRADUATIONAND ACADEMIC FAILURE RATES
123
impairments, known brain damage, or history of primary emotional disorders. The mean age of the RSS students at entrance to the college was 23 years and they were significantly older (p
