The Adult Outcomes of Children with Learning Disabilities Margaret Bruck McGill Montreal Children's Hospital Learning Center Montreal, Quebec
The results of four follow-up studies of learning-disabled children are reviewed. A comparison of results among the studies and of analyses of individual variation within the studies suggest some factors associated with long-term outcomes. It is argued that in many cases learning disabilities are not a lifelong handicapping condition, especially if adequate treatment is provided during childhood.
Although there has been a proliferation of clinical, psychological, and educational research on children with learning disabilities (LD), our knowledge of the natural history and long-term effects of the syndrome beyond childhood is limited. Some studies indicate that learning disabilities are a lifelong condition; adults with childhood diagnoses of LD continue to exhibit reading, spelling, and math problems (Bruck 1985; Frauenheim and Heckerl 1983; Muehl and Forell 1974). Less is known about the long term consequences of childhood learning disabilities. Despite common beliefs that learning disabilities may signal serious long-term adjustment problems in the areas of educational and occupational achievements as well as social and emotional functioning, there are few well controlled studies to test these hypotheses. The purpose of this paper is to review and integrate results of four
Preparation of this manuscript was supported by the National Health and Welfare Research and Development Program through a project grant and through a National Health Research Scholar Award to Margaret Bruck. I would like to thank Dr. Barton Childs and Dr. Otfried Spreen for their comments on an earlier draft of this manuscript.
252
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
253
follow-up studies of adults with childhood diagnoses of learning disabilities that provide data relevant to these hypotheses. The four studies were selected because they overcome the following major criticisms of studies on the long-term effects of childhood learning disabilities (see Bruck 1985; Herjanic and Penick 1972; Schonhaut and Satz 1984 for recent reviews). First, samples sizes were adequate. Second, appropriate control groups were tested. Third, at the time of follow-up, the subjects had finished their high school careers, thus permitting the examination of specific areas of adult adjustment and functioning. Fourth, the studies are broad in scope; comprehensive data on educational and occupational achievements were collected, and, in two studies, social and emotional adjustments were examined. Thus, the results can be used to test several popular hypotheses about the relationship among learning disabilities, juvenile delinquency, and unemployment (e.g., Mauser 1974; Schenk et al. 1980). Furthermore, these data will suggest whether the frequently noted social and emotional problems of LD children (cf. Ceci 1986 for reviews) are specific to the childhood years or whether they persist into adulthood. Fifth, the subjects met traditional exclusionary definitions of learning disabilities. That is, their learning problems could not be attributed to such factors as neurological impairment, mental retardation, social, or cultural deprivation. These subject selection criteria allow more direct interpretations of associations between childhood disabilities and adult outcomes than is found in other follow-up studies in which subjects had low or borderline IQs, were from lower class backgrounds, had a number of psychiatric symptoms, or were neurologically impaired. While it is possible that children sharing such symptoms may also have a specific learning disability, adult outcomes might be associated with these symptoms (e.g., low social status or neurological impairment) rather than with a learning disability as such. Sixth, in each of the studies, there was a wide range of variability in terms of subjects' childhood learning disabilities. Studies that examine only children with the most severe disabilities cannot be generalized to the LD population, which includes individuals with varying degrees of impairment. Finally, because the characteristics of the subjects were well specified in these studies, it is possible to draw some conclusions about the characteristics of adults with the poorest outcomes and to identify some predictors of adult adjustment.
Method The settings, subject selection procedures, and follow-up methods of each of the four studies are summarized. The first study by Rawson (1968) examined the adult outcomes of
254
MXNDS AT WORK
LD boys who attended a private elementary school for normal boys and girls. Because this school was ideologically aligned with the principles of the progressive education movement, it showed a commitment to meet the individual needs of its pupils. Thus, not only were a number of LD children admitted to the school, but these children were also provided with daily individualized remedial programs. According to the subject selection criteria, the children exhibited no neurological or emotional problems and had above average IQ scores (average of 130). The children demonstrated severe and moderate childhood disabilities. All had failed to learn to read by grade 2 or 3. Most of the parents were professionals and business people; they were very involved in the operation of the school and in the education of their children. At the time of follow-up (1964-65), the subjects' average age was 33 years. Their occupational and educational status was compared to that of 20 classmates with no history of academic difficulties. The second study, conducted by a group of researchers at Johns Hopkins University (Childs et al. 1982; Finucci, Gottfredson, and Childs 1985; Gottfredson, Finucci, and Childs 1984), evaluated the educational and occupational outcomes of 579 boys who attended a residential college preparatory school for LD adolescents. In addition to the academic program, all students were required to take a remedial reading and writing class. All subjects had long histories of school failure, despite normal IQ's (average 118), and no neurological, sensory, or overriding emotional disorders. On the basis of standardized test scores, it was estimated that 75 percent of the sample was severely disabled when they entered the school at age 14. The families were white-collar professionals. Strong parental support and understanding is suggested by the fact that these children were sent to this special school. At follow-up (1970), questionnaires requesting information about educational and occupational achievements were sent to the LD subjects. The average age of the respondents was 33. Their responses were compared to those of 612 men, of similar age and socioeconomic status, who had attended a private college preparatory day school. The third study by Spreen (1981, 1982, 1984, personal communciation) included a sample of 8-12-year-old boys and girls who were assessed at a neuropsychology clinic. Subjects were classified into three groups on the basis of childhood neurological symptoms. The group of subjects with learning difficulties with no neurological indications of brain dysfunction are of interest for the present discussion. These subjects were of average IQ (mean 99), although subjects with IQ scores as low as 70 were included in the sample. On the average, they were from middle class backgrounds, although 25 percent were from lower class backgrounds. The children received treatment for learning disabilities that was available through the school system, tutors, or summer
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
255
schools. Nevertheless, follow-up interviews suggested that despite early diagnosis, these children did not receive "intensive or even adequate special t r e a t m e n t . . , the most common complaints of both patients and clients centered around the lack of educational intervention" (Spreen 1981). Forty-three subjects were first followed-up during the early 1970s at mean age of 18.5 years, and 27 were recently reassessed at mean age of 25 years. Their educational and occupational achievements, as well as their social and emotional functioning, were compared to a control group of normally achieving students matched on the basis of sex, age, and social status. The fourth study conducted by Bruck (1985, 1986) included 101 boys and girls w h o were assessed between the ages of five to ten years at the learning-disabilities clinic. Upon initial diagnosis, they had at least normal IQs (average 103), showed no primary behavioral or emotional disturbance, and were free of neurolgical, sensory, or physical disabilities. All children had primary problems learning to read and 75 percent had difficulties with mathematics. Based on clinicians' judgments of the severity of childhood learning disabilities, it was estimated that 43 percent were severely disabled, 31 percent were moderately disabled, and 27 percent were mildly disabled. The children were primarily from middle class backgrounds, with only 6 percent of the sample from lower class backgrounds. The children received substantial help for their learning disabilities throughout their compulsory schooling. Some of this help came directly through the clinic. This included post-assessment counselling for parents and teachers which involved explaining the nature of the child's problem, finding tutors, and providing assistance in educational planning. Half of the sample received an individualized clinic remedial program as a supplement to their normal school activities. This program was designed to give each child the necessary support and skills to achieve positive experiences in both academic and nonacademic settings. In addition, during the elementary school years, 45 percent of the sample received special school assistance for their learning disabilities and 59 percent had private tutoring. In high school, 25 percent received special school services, and 55 percent received private tutoring. On the average each subject received 4.5 years of special assistance for learning problems. At follow-up (the late 1970s), the average age was 21; approximately half of the subjects were late adolescents (17-21 years) and half were young adults (22-29 years). Their literacy skills, educational and occupational achievements, as well as their social and emotional functioning, were compared to two control groups. The first was a group of non-learning-disabled peers matched to the LD subjects on the basis of sex, age, and social status. Comparisons with local and national norms indicated that this control group's academic and occupational achieve-
256
MrNDs A T WORK
merits were atypically high. Thus a group of non-learning-disabled siblings matched with the LD subjects for sex, age, and social status was used to further assess educational and occupational achievements.
Results
The results are presented in four sections: literacy skills, academic outcomes, occupational outcomes, and social and emotional adjustments. Levels of Literacy Skills
Only one (Bruck 1985) of the four studies objectively assessed the literacy skills of LD adults. Compared to a college control group, LD subjects performed more poorly on standardized tests of reading and comprehension, reading speed, single word decoding, spelling, and math. Reading comprehension was the most intact skill; major deficits were most apparent for single word decoding and spelling skills. Although these data indicate that learning disabilities persist into adulthood, the interpretation must be tempered by two qualifications. First, subjects compensated for basic word recognition deficits to achieve adequate levels of comprehension. Second, both the Hopkins group and Bruck reported that by most standards the subjects were literate; Bruck estimated that approximately 6 percent of the sample might be considered illiterate (test scores below the grade 6 level). Thus subjects had the basic reading, writing, and math skills to meet the literacy demands of higher educational and skilled occupational environments. Academic Outcomes It is a common belief that LD children are at risk for high school drop-out. Only Spreen's results supported this hypothesis; 62 percent of his LD subjects (vs 2 percent of the control subjects) dropped out of high school. The drop-out rates for the other three studies were low: 0 percent (Rawson), 6 percent (Hopkins)) and 10 percent (Bruck). Except in Spreen's study, most LD subjects entered college after finishing high school; these rates were 100 percent (Rawson), 81 percent (Hopkins), 58 percent (Bruck), and 12 percent (Spreen). The percentage of college students who had dropped out of undergraduate programs was 20 percent (Rawson), 31 percent (Bruck) and 44 percent (Hopkins). llt is not clear from the Hopkins' data whether this percentage of the sample dropped out of school or whether they simply did not graduate from high school (Childs, personal communication).
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
257
Bruck and Rawson found similar patterns and rates of educational achievement for the LD and control groups, while Spreen and the Hopkins group reported poorer achievements for the LD than for the control group. It seems that in the case of the Hopkins' study, group differences reflected the exceptionally high achievement of a select control group rather than the poor functioning of the LD subjects. 2 Although a large proportion of LD children do enter a college program, their educational experiences are not similar to those of nondisabled students. According to Bruck, the LD college students worked extremely hard to function in a university program. Reading complex prose, writing technical reports, and taking timed exams, presented particular difficulties. A high proportion of LD college students required additional years to complete their college or university program (31 percent in the Bruck study and 33 percent in the Hopkins study) either because they had taken reduced academic loads or had taken time off during their college years. The associations of educational level to severity of childhood learning disability, childhood IQ, and family socioeconomic status (SES) were explored in some studies. The Hopkins results indicated that those adolescents with the most severe reading and spelling disabilities were least likely to complete a college program. However, Bruck found that while the most disabled children were least likely to enter a college program, once in college, their chance of success was equal to that of those with milder disabilities. In general, although the associations between childhood learning disabilities and educational achievement were statistically significant, they were weak. For example, Bruck found that there were many severely disabled subjects who did enter a college program, and the Hopkins data indicated that there were many severely disabled subjects who did complete their college education. Bruck (unpublished data) found that childhood IQ followed by family SES were the best predictors of educational outcomes. Further analyses revealed that these associations were strongest for those subjects at the extremes of each scale. For example, all subjects from lower class backgrounds either dropped out of high school or did not continue their education beyond high school. Subjects who were successful in college programs had the highest childhood IQs. In summary, in three studies the LD subjects were moderately suc2The Hopkins group argues that these differences are psychologically real, given the backgrounds and expectations of the LD boys' parents. However, there are no data on the parents' expectations to support this contention. Given the fact that these boys" learning problems were severe enough to require special school placement, it seems more likely that despite the equivalent backgrounds of the LD and control subjects" families, the LD parents would have very different expectations than the parents of the normal boys.
258
MXNDS AT WORK
cessful in terms of their educational achievements (see Balow and Blomquist 1965; Preston and Yarrington 1967, Robinson and Smith 1962 for similar findings). The high rate of school drop-out in Spreen's study may reflect specific characteristics of his subjects. In contrast to other samples, a greater proportion of Spreen's subjects were from lower SES backgrounds, had lower IQs, and received few adequate services for learning disabilities. It is possible that the presence of these factors compound the consequences of learning disabilities. Conversely, high levels of family support and understanding, adequate intelligence levels, and adequate intervention programs may dilute the association between childhood learning disabilities and educational achievement (see Childs et al. 1982, for further discussion). Occupational Achievements
At follow-up, the unemployment rates for each study were: 0 percent (Rawson), 6 percent (Hopkins) 11 percent (Bruck) and 46 percent (Spreen). There were no differences between the unemployment rates for LD and control subjects, except in the Spreen study. In terms of occupational status, the patterns of results are quite consistent across three studies. For example, in the Hopkins' study, comparisons with the census data indicated that the LD subjects were in prestigious occupations; they were seldom found in the lower groups that employ 67 percent of all adult white men. However, the Hopkins group found that control men were more likely to be professionals (doctors, lawyers) and to hold positions requiring high levels of education and literacy skills, while the LD men were more likely to be in managerial (presidents, vice-presidents) or sales positions that required good communication skills. Occupational achievements were lower in Spreen's study; 56 percent of the subjects were unskilled laborers compared to an average rate of 10 percent in the three other studies. The high unemployment rates and low occupational achievements found in Spreen's study probably reflect the poor academic achievement of his sample rather than the long term effects of childhood learning disabilities. In Bruck's study, for example, both unemployed LD and control subjects were most likely to be high school drop-outs. In general these data indicate that LD adults are gainfully employed, and many are represented in prestigious occupations. Although they may not become lawyers or doctors, they achieve their success through other channels (see also Preston and Yarrington 1967; Robinson and Smith 1962). Social and Emotional Functioning
Three issues about the social and emotional adjustments of LD adults will be addressed. First, are specific learning disabilities precur-
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
259
sors of antisocial behaviors such as juvenile delinquency, drug and alcohol abuse? Second, do the emotional and social problems frequently noted in LD children persist into adulthood, or are these problems specific to the childhood years? Third, what are the characteristics of LD adults most at risk for poor social and emotional adjustments? Only the data from the Bruck and Spreen studies are relevant to these issues. Regarding the first issue, both Spreen (1981) and Bruck concluded that childhood learning disabilities were not precursors of asocial behaviors. Bruck reported that there were no differences between LD and control groups in the rates of drug or alcohol abuse. In the Bruck and Spreen studies, control and LD subjects did not differ in terms of the number of delinquent acts or encounters with the police. Bruck found no differences in the rates of incarceration; however, Spreen reported that the LD group received somewhat more frequent and severe penalties, despite no differences in the number of offenses committed. It is possible that the more severe penalties reflect the LD subjects' more inappropriate or less persuasive behavior in court. This interpretation is consistent with reports that LD children do not present a good first impression and do not ingratiate themselves with their listeners (Donahue and Bryan 1984; Perlmutter and Bryan 1984). The findings of no association between asocial behavior and learning disabilities are clearly inconsistent with other studies that are characterized by their retrospective and correlational nature (see Spreen 1981 for a review). However, by virtue of their design, retrospective studies do not indicate whether LD children are more at risk for becoming delinquent than a nondisabled group; they merely show that delinquent or incarcerated samples include many subjects with poor learning skills or histories. In contrast, prospective studies such as Spreen's and Bruck's demonstrate that LD children who are followed into late adolescence and early adulthood commit delinquent offenses, but these are similar in frequency and in quality to those reported by subjects who did not experience academic problems. In terms of the second issue, there are some data to indicate that like LD children, LD adults may be at risk for social and emotional problems. Based on parents' ratings of their children's social and emotional adjustments at age 18, Spreen (Peter and Spreen 1979) found that LD subjects differed from controls on only one of six dimensions of problem behaviors--asocial acting out. However, on all self-rating scales, these LD subjects perceived themselves to be less socially and emotionally adjusted than control subjects. Based on clinicians' ratings of interview data, Bruck found that LD and control subjects showed similar adjustments in the areas of family relationships and interactions with the opposite sex. However, LD subjects suffered more severe peer relationship problems with members of the same sex; 13 percent of the LD subjects were lonely with few social contacts (no control
260
MINDS AT WORK
subjects showed this pattern). They also suffered higher rates of poor psychological adjustment; they showed specific difficulties in controlling their tempers and in dealing with frustrations in a constructive way. While the learning disabled and control groups were differentiated on statistical grounds, the clinical significance of these findings is not clear. It is possible that between-group differences simply reflect that LD adults are less happy or less socially involved than control subjects; on the other hand, they may indicate that LD adults are unhappy or socially isolated. Bruck suggests that only a small proportion of LD adults showed serious adjustment problems. Only 5 percent of the LD sample had received recent counseling for social and emotional problems. Approximately 15 percent of the LD subjects would have been appropriate referrals for family counseling, social skills training, or individual psychotherapy; but none were disturbed sufficiently to require hospitalization. Both Bruck and Spreen identified some characteristics of LD adults most likely to show adjustment problems. Bruck found that adults with adjustment problems also showed similar problems as children. It should be noted, however, that many showed improvement from childhood to follow-up; over 50 percent of subjects with poor childhood adjustment showed good adjustment at follow-up. In both studies, LD female subjects showed the poorest adjustments, and, according to Bruck's data, females were more likely to show persistent childhood problems. In the case of the Bruck stud~ sex differences were specific to social and emotional functioning only; females and males did not differ in terms of severity of childhood learning disabilities, literacy skills, occupational or educational achievements. The finding that LD females were at risk for poor adjustment not only characterizes samples of LD adults but characterizes some samples of LD children (see Bruck 1986 for a review). The poorer adjustments of female subjects may reflect differences in society's reactions to failure in males and females. Females who fail may be more rejected by peers and less accepted by adults than boys who fail. These interactions may be the precursors of social and emotional problems of LD females. Bruck reported several other characteristics that were associated with poor adult adjustments. Age at follow-up was related to psychological adjustment; poor adjustments were characteristic of the younger LD group (17-20-year-olds) but not of the older group (21-29year-olds). Educational success was associated with improved peer relationships. This may reflect greater opportunities for peer interaction in educational than in occupational milieus. However, severity of childhood learning disability was not related to family, peer, or psychological adjustments at follow-up. To summarize, like LD children, LD adults were more likely to
ADULT OUTCOMES OF CHILDREN WITH L~ARNING DISABILITIES
261
show adjustment problems than normal peers. The finding that severity of learning disabilities was not related to measures of social and emotional adjustment suggests that, at least in the adult population, social and emotional difficulties are widely distributed and are not confined to the more severe cases of learning disabilities. However, it should also be noted that the rates of reported problems were low, and at least in Bruck's study, most of the subjects were well-adjusted.
Conclusions
Several conclusions can be drawn from these four studies concerning the natural history and long term consequences of childhood learning disabilities. First, learning disabilities are a lifelong disorder; they persist into adulthood. Results of interview and standardized test data indicate that LD adults continued to experience some difficulty in the areas of literacy skills. However, most subjects had sufficient skills to function in academic and skilled occupational environments. While these data document the persistence of learning disabilities, little is known about the underlying nature of LD adults' literacy problems. Future studies should examine the cognitive processes that LD adults use for reading and spelling and should identify the compensatory mechanisms that LD adults may employ for reading and spelling. The data also indicate that the majority of adults with childhood diagnoses of learning disabilities are well-adjusted and successful in major areas of their adult lives. For example, in all but the Spreen study, LD subjects were not at risk for high school drop-out. The majority of LD subjects entered a university program, where despite their persistent problems, many were academically successful. LD adults were not at risk for unemployment or underemployment. They were represented in a wide range of occupations, and few held unskilled jobs. The poor occupational outcomes of Spreen's subjects highlight the importance of high school graduation as a prerequisite for successful educational and occupational achievements. LD students who are provided with an adequate and stimulating educational program will have acquired the motivation and skills to continue their education and to become gainfully employed. The data on social and emotional functioning indicate that there is no association between childhood learning disabilities and socially deviant behaviors such as delinquency, alcohol abuse, or drug abuse. Although the data show that childhood learning disabilities are not precursors of extreme forms of adult psychopathology, it seems that some LD adults, especially females, experience social and emotional problems. More refined measures must be used in future studies to indicate the clinical significance of these poorer adjustments.
262
MINDS AT WORK
A comparison of the results across the four studies along with the results of analyses within specific studies suggest some factors that are associated with positive adult outcomes. One factor is severity of childhood disability. Some analyses indicate that the most severely disabled children were most at risk for poor outcomes in terms of educational, occupational, and literacy skills. Other analyses indicated that socioeconomic status and IQ accounted for individual differences in outcomes within and across studies. Thus, not surprisingl3~ subjects from lower socioeconomic backgrounds or of marginally average IQs had the poorest outcomes, and studies that included high proportions of such subjects yielded the least optimistic outcomes. While the poorer outcomes found in such studies may not be associated with the childhood syndrome of learning disabilities, but with other secondary factors (e.g., IQ), more research is required to determine whether the presence of these factors act additively or synergistically with learning disabilities. The results suggest that the most important antecedents of positive outcome are early identification accompanied by adequate intervention. In addition to their direct educational benefits, it also appears that these antecedent conditions may have emotional and motivational benefits that contribute to positive outcomes. In Bruck's study, for example, early identification was accompanied by counselling for parents and teachers to explain the nature of the child's problems. These adults" subsequent understanding and support may have decreased many of the pressures on these children and led to more positive attitudes toward education, learning, and their own self-concepts. It was this dimension of intervention, rather than dimensions of IQ, socioeconomic status, and severity of childhood disability, that most differentiated Spreen's samples from those examined in the Bruck, Rawson, and Hopkins studies. It is possible that lack of adequate services for Spreen's subjects was an antecedent of high school drop-out which preceded poor employment opportunities. By emphasizing the importance of early identification paired with intervention for positive adult outcomes, one also emphasizes the fact that childhood learning disabilities are not a trivial condition that can be left untreated. On the other hand, it is suggested that with adequate treatment, learning disabilities are not necessarily a lifelong handicapping condition. With increased awareness of this condition and with increases in the delivery of services for these children, it is hoped that in the future the prognoses of adult outcomes will become more positive. Refeycnces Balow, B. J. and Blomquist, M. 1965. Young adults 10-15 years after a severe reading disability. Elementary School Journal 66:44-48.
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
263
Bruck, M. 1985. The adult functioning of children with specific learning disabilities: A follow-up study. In I. Siegel (ed.). Advances in Applied Developmental Psychology. New Jersey: Ablex. Bruck, M. 1986. Social and emotional adjustments of learning disabled children: A review of the issues. In S. Ceci (ed.). Handbook of Cognitive, Social and Neuropsychological Aspects of Learning Disabilities. Vol. 1 Hillsdale, N.J.: Erlbaum. Ceci, S. 1986. Handbook of Cognitive, Social and Neuropsychological Aspects of Learning Disabilities. Hillsdale, N.J.: Erlbaum. Childs, B., Finucci, J. M., Pulver, A. E., and Tielsch, J. 1982. The natural history of specific reading disability. Educational outcomes. Unpublished manuscript. Department of Pediatrics, Johns Hopkins University. Donahue, M., and Bryan, T. 1984. Communicative skills and peer relations of learning disabled adolescents. Topics in Language Disorders 1:10-21. Finucci, J., Gottfredson, L. S., and Childs, B. 1985. A follow-up study of dyslexic boys. Annals of Dyslexia 35:117-136. Frauenheim, J. G., and Heckerl, J. R. 1983. A longitudinal study of psychological and achievement test performance in severe dyslexic adults. Journal of Learning Disabilities 16:339-347. Gottfredson, L. S., Finucci, J. M., and Childs, B. 1984. Explaining the adult careers of dyslexic boys. Variations in critical skills for high level jobs. Journal of Vocational Behavior 24:355-373. Herjanic, B., and Penick, E. 1972. Adult outcomes of disabled child readers. Journal of Special Education 6:397-410. Mauser, A. J. 1974. Learning disabilities and delinquent youth. Academic Therapy 9:389402. Muehl, S., and Forell, E. 1974. A follow-up study of disabled readers: Variables related to high school reading performance. Reading Research Quarterly 9:110-123. Peter, B. M., and Spreen, O. 1979. Behavior rating and personal adjustment scales of neurologically and learning handicapped children during adolescence and early adulthood: Results of a follow-up study. Journal of Clinical Neuropsychology 1:75-92. Perlmutter, B. F., and Bryan, J. 1984. First impressions, ingratiation, and the learning disabled child. Journal of Learning Disabilities 17:156-161. Preston, R. C., and Yarrington, D. J. 1967. Status of 50 retarded readers eight years after reading clinic diagnosis. Journal of Reading 11:122-129. Rawson, M. 1968. Developmental Language Disability. Baltimore: Johns Hopkins Press. Robinson, H. M., and Smith, H. 1962. Reading clinic clients--ten years after. Elementary School Journal 63:22-27. Schenck, B. J., Fitzsimmons, J., Bullard, P. C., Taylor, H. G., and Satz, P. 1980. A prevention model for children at risk for reading failure. In R. Knights and D. J. Bakker (eds.). Treatment of Hyperactive and Learning Disordered Children. Baltimore: University Park Press. Schonhaut, S. and Satz, P. 1984. Prognosis of the learning disabled child: A review of follow-up studies. In M. Rutter (ed.). Behavioral Syndromes and Brain Dysfunction in Childhood. New York: Guilford Press. Spreen, O. 1981. The relationship between learning disability, neurological impairment and delinquency: Results of a follow-up study. Journal of Nervous and Mental Diseases 169:791-802. Spreen, O. 1982. Adult outcomes of reading disorders. In R. N. Malatesha and P. G. Aaron (eds.). Reading Disorders, Varieties and Treatments. New York: Academic Press. Spreen, O. 1984. A prognostic view from middle childhood. In M. Levine and P. Satz (eds.). Middle Childhood: Developmental variations and dysfunction. New York: Academic Press.
The results of four follow-up studies of learning-disabled children are reviewed. A comparison of results among the studies and of analyses of individual variation within the studies suggest some factors associated with long-term outcomes. It is argued that in many cases learning disabilities are not a lifelong handicapping condition, especially if adequate treatment is provided during childhood.
Although there has been a proliferation of clinical, psychological, and educational research on children with learning disabilities (LD), our knowledge of the natural history and long-term effects of the syndrome beyond childhood is limited. Some studies indicate that learning disabilities are a lifelong condition; adults with childhood diagnoses of LD continue to exhibit reading, spelling, and math problems (Bruck 1985; Frauenheim and Heckerl 1983; Muehl and Forell 1974). Less is known about the long term consequences of childhood learning disabilities. Despite common beliefs that learning disabilities may signal serious long-term adjustment problems in the areas of educational and occupational achievements as well as social and emotional functioning, there are few well controlled studies to test these hypotheses. The purpose of this paper is to review and integrate results of four
Preparation of this manuscript was supported by the National Health and Welfare Research and Development Program through a project grant and through a National Health Research Scholar Award to Margaret Bruck. I would like to thank Dr. Barton Childs and Dr. Otfried Spreen for their comments on an earlier draft of this manuscript.
252
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
253
follow-up studies of adults with childhood diagnoses of learning disabilities that provide data relevant to these hypotheses. The four studies were selected because they overcome the following major criticisms of studies on the long-term effects of childhood learning disabilities (see Bruck 1985; Herjanic and Penick 1972; Schonhaut and Satz 1984 for recent reviews). First, samples sizes were adequate. Second, appropriate control groups were tested. Third, at the time of follow-up, the subjects had finished their high school careers, thus permitting the examination of specific areas of adult adjustment and functioning. Fourth, the studies are broad in scope; comprehensive data on educational and occupational achievements were collected, and, in two studies, social and emotional adjustments were examined. Thus, the results can be used to test several popular hypotheses about the relationship among learning disabilities, juvenile delinquency, and unemployment (e.g., Mauser 1974; Schenk et al. 1980). Furthermore, these data will suggest whether the frequently noted social and emotional problems of LD children (cf. Ceci 1986 for reviews) are specific to the childhood years or whether they persist into adulthood. Fifth, the subjects met traditional exclusionary definitions of learning disabilities. That is, their learning problems could not be attributed to such factors as neurological impairment, mental retardation, social, or cultural deprivation. These subject selection criteria allow more direct interpretations of associations between childhood disabilities and adult outcomes than is found in other follow-up studies in which subjects had low or borderline IQs, were from lower class backgrounds, had a number of psychiatric symptoms, or were neurologically impaired. While it is possible that children sharing such symptoms may also have a specific learning disability, adult outcomes might be associated with these symptoms (e.g., low social status or neurological impairment) rather than with a learning disability as such. Sixth, in each of the studies, there was a wide range of variability in terms of subjects' childhood learning disabilities. Studies that examine only children with the most severe disabilities cannot be generalized to the LD population, which includes individuals with varying degrees of impairment. Finally, because the characteristics of the subjects were well specified in these studies, it is possible to draw some conclusions about the characteristics of adults with the poorest outcomes and to identify some predictors of adult adjustment.
Method The settings, subject selection procedures, and follow-up methods of each of the four studies are summarized. The first study by Rawson (1968) examined the adult outcomes of
254
MXNDS AT WORK
LD boys who attended a private elementary school for normal boys and girls. Because this school was ideologically aligned with the principles of the progressive education movement, it showed a commitment to meet the individual needs of its pupils. Thus, not only were a number of LD children admitted to the school, but these children were also provided with daily individualized remedial programs. According to the subject selection criteria, the children exhibited no neurological or emotional problems and had above average IQ scores (average of 130). The children demonstrated severe and moderate childhood disabilities. All had failed to learn to read by grade 2 or 3. Most of the parents were professionals and business people; they were very involved in the operation of the school and in the education of their children. At the time of follow-up (1964-65), the subjects' average age was 33 years. Their occupational and educational status was compared to that of 20 classmates with no history of academic difficulties. The second study, conducted by a group of researchers at Johns Hopkins University (Childs et al. 1982; Finucci, Gottfredson, and Childs 1985; Gottfredson, Finucci, and Childs 1984), evaluated the educational and occupational outcomes of 579 boys who attended a residential college preparatory school for LD adolescents. In addition to the academic program, all students were required to take a remedial reading and writing class. All subjects had long histories of school failure, despite normal IQ's (average 118), and no neurological, sensory, or overriding emotional disorders. On the basis of standardized test scores, it was estimated that 75 percent of the sample was severely disabled when they entered the school at age 14. The families were white-collar professionals. Strong parental support and understanding is suggested by the fact that these children were sent to this special school. At follow-up (1970), questionnaires requesting information about educational and occupational achievements were sent to the LD subjects. The average age of the respondents was 33. Their responses were compared to those of 612 men, of similar age and socioeconomic status, who had attended a private college preparatory day school. The third study by Spreen (1981, 1982, 1984, personal communciation) included a sample of 8-12-year-old boys and girls who were assessed at a neuropsychology clinic. Subjects were classified into three groups on the basis of childhood neurological symptoms. The group of subjects with learning difficulties with no neurological indications of brain dysfunction are of interest for the present discussion. These subjects were of average IQ (mean 99), although subjects with IQ scores as low as 70 were included in the sample. On the average, they were from middle class backgrounds, although 25 percent were from lower class backgrounds. The children received treatment for learning disabilities that was available through the school system, tutors, or summer
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
255
schools. Nevertheless, follow-up interviews suggested that despite early diagnosis, these children did not receive "intensive or even adequate special t r e a t m e n t . . , the most common complaints of both patients and clients centered around the lack of educational intervention" (Spreen 1981). Forty-three subjects were first followed-up during the early 1970s at mean age of 18.5 years, and 27 were recently reassessed at mean age of 25 years. Their educational and occupational achievements, as well as their social and emotional functioning, were compared to a control group of normally achieving students matched on the basis of sex, age, and social status. The fourth study conducted by Bruck (1985, 1986) included 101 boys and girls w h o were assessed between the ages of five to ten years at the learning-disabilities clinic. Upon initial diagnosis, they had at least normal IQs (average 103), showed no primary behavioral or emotional disturbance, and were free of neurolgical, sensory, or physical disabilities. All children had primary problems learning to read and 75 percent had difficulties with mathematics. Based on clinicians' judgments of the severity of childhood learning disabilities, it was estimated that 43 percent were severely disabled, 31 percent were moderately disabled, and 27 percent were mildly disabled. The children were primarily from middle class backgrounds, with only 6 percent of the sample from lower class backgrounds. The children received substantial help for their learning disabilities throughout their compulsory schooling. Some of this help came directly through the clinic. This included post-assessment counselling for parents and teachers which involved explaining the nature of the child's problem, finding tutors, and providing assistance in educational planning. Half of the sample received an individualized clinic remedial program as a supplement to their normal school activities. This program was designed to give each child the necessary support and skills to achieve positive experiences in both academic and nonacademic settings. In addition, during the elementary school years, 45 percent of the sample received special school assistance for their learning disabilities and 59 percent had private tutoring. In high school, 25 percent received special school services, and 55 percent received private tutoring. On the average each subject received 4.5 years of special assistance for learning problems. At follow-up (the late 1970s), the average age was 21; approximately half of the subjects were late adolescents (17-21 years) and half were young adults (22-29 years). Their literacy skills, educational and occupational achievements, as well as their social and emotional functioning, were compared to two control groups. The first was a group of non-learning-disabled peers matched to the LD subjects on the basis of sex, age, and social status. Comparisons with local and national norms indicated that this control group's academic and occupational achieve-
256
MrNDs A T WORK
merits were atypically high. Thus a group of non-learning-disabled siblings matched with the LD subjects for sex, age, and social status was used to further assess educational and occupational achievements.
Results
The results are presented in four sections: literacy skills, academic outcomes, occupational outcomes, and social and emotional adjustments. Levels of Literacy Skills
Only one (Bruck 1985) of the four studies objectively assessed the literacy skills of LD adults. Compared to a college control group, LD subjects performed more poorly on standardized tests of reading and comprehension, reading speed, single word decoding, spelling, and math. Reading comprehension was the most intact skill; major deficits were most apparent for single word decoding and spelling skills. Although these data indicate that learning disabilities persist into adulthood, the interpretation must be tempered by two qualifications. First, subjects compensated for basic word recognition deficits to achieve adequate levels of comprehension. Second, both the Hopkins group and Bruck reported that by most standards the subjects were literate; Bruck estimated that approximately 6 percent of the sample might be considered illiterate (test scores below the grade 6 level). Thus subjects had the basic reading, writing, and math skills to meet the literacy demands of higher educational and skilled occupational environments. Academic Outcomes It is a common belief that LD children are at risk for high school drop-out. Only Spreen's results supported this hypothesis; 62 percent of his LD subjects (vs 2 percent of the control subjects) dropped out of high school. The drop-out rates for the other three studies were low: 0 percent (Rawson), 6 percent (Hopkins)) and 10 percent (Bruck). Except in Spreen's study, most LD subjects entered college after finishing high school; these rates were 100 percent (Rawson), 81 percent (Hopkins), 58 percent (Bruck), and 12 percent (Spreen). The percentage of college students who had dropped out of undergraduate programs was 20 percent (Rawson), 31 percent (Bruck) and 44 percent (Hopkins). llt is not clear from the Hopkins' data whether this percentage of the sample dropped out of school or whether they simply did not graduate from high school (Childs, personal communication).
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
257
Bruck and Rawson found similar patterns and rates of educational achievement for the LD and control groups, while Spreen and the Hopkins group reported poorer achievements for the LD than for the control group. It seems that in the case of the Hopkins' study, group differences reflected the exceptionally high achievement of a select control group rather than the poor functioning of the LD subjects. 2 Although a large proportion of LD children do enter a college program, their educational experiences are not similar to those of nondisabled students. According to Bruck, the LD college students worked extremely hard to function in a university program. Reading complex prose, writing technical reports, and taking timed exams, presented particular difficulties. A high proportion of LD college students required additional years to complete their college or university program (31 percent in the Bruck study and 33 percent in the Hopkins study) either because they had taken reduced academic loads or had taken time off during their college years. The associations of educational level to severity of childhood learning disability, childhood IQ, and family socioeconomic status (SES) were explored in some studies. The Hopkins results indicated that those adolescents with the most severe reading and spelling disabilities were least likely to complete a college program. However, Bruck found that while the most disabled children were least likely to enter a college program, once in college, their chance of success was equal to that of those with milder disabilities. In general, although the associations between childhood learning disabilities and educational achievement were statistically significant, they were weak. For example, Bruck found that there were many severely disabled subjects who did enter a college program, and the Hopkins data indicated that there were many severely disabled subjects who did complete their college education. Bruck (unpublished data) found that childhood IQ followed by family SES were the best predictors of educational outcomes. Further analyses revealed that these associations were strongest for those subjects at the extremes of each scale. For example, all subjects from lower class backgrounds either dropped out of high school or did not continue their education beyond high school. Subjects who were successful in college programs had the highest childhood IQs. In summary, in three studies the LD subjects were moderately suc2The Hopkins group argues that these differences are psychologically real, given the backgrounds and expectations of the LD boys' parents. However, there are no data on the parents' expectations to support this contention. Given the fact that these boys" learning problems were severe enough to require special school placement, it seems more likely that despite the equivalent backgrounds of the LD and control subjects" families, the LD parents would have very different expectations than the parents of the normal boys.
258
MXNDS AT WORK
cessful in terms of their educational achievements (see Balow and Blomquist 1965; Preston and Yarrington 1967, Robinson and Smith 1962 for similar findings). The high rate of school drop-out in Spreen's study may reflect specific characteristics of his subjects. In contrast to other samples, a greater proportion of Spreen's subjects were from lower SES backgrounds, had lower IQs, and received few adequate services for learning disabilities. It is possible that the presence of these factors compound the consequences of learning disabilities. Conversely, high levels of family support and understanding, adequate intelligence levels, and adequate intervention programs may dilute the association between childhood learning disabilities and educational achievement (see Childs et al. 1982, for further discussion). Occupational Achievements
At follow-up, the unemployment rates for each study were: 0 percent (Rawson), 6 percent (Hopkins) 11 percent (Bruck) and 46 percent (Spreen). There were no differences between the unemployment rates for LD and control subjects, except in the Spreen study. In terms of occupational status, the patterns of results are quite consistent across three studies. For example, in the Hopkins' study, comparisons with the census data indicated that the LD subjects were in prestigious occupations; they were seldom found in the lower groups that employ 67 percent of all adult white men. However, the Hopkins group found that control men were more likely to be professionals (doctors, lawyers) and to hold positions requiring high levels of education and literacy skills, while the LD men were more likely to be in managerial (presidents, vice-presidents) or sales positions that required good communication skills. Occupational achievements were lower in Spreen's study; 56 percent of the subjects were unskilled laborers compared to an average rate of 10 percent in the three other studies. The high unemployment rates and low occupational achievements found in Spreen's study probably reflect the poor academic achievement of his sample rather than the long term effects of childhood learning disabilities. In Bruck's study, for example, both unemployed LD and control subjects were most likely to be high school drop-outs. In general these data indicate that LD adults are gainfully employed, and many are represented in prestigious occupations. Although they may not become lawyers or doctors, they achieve their success through other channels (see also Preston and Yarrington 1967; Robinson and Smith 1962). Social and Emotional Functioning
Three issues about the social and emotional adjustments of LD adults will be addressed. First, are specific learning disabilities precur-
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
259
sors of antisocial behaviors such as juvenile delinquency, drug and alcohol abuse? Second, do the emotional and social problems frequently noted in LD children persist into adulthood, or are these problems specific to the childhood years? Third, what are the characteristics of LD adults most at risk for poor social and emotional adjustments? Only the data from the Bruck and Spreen studies are relevant to these issues. Regarding the first issue, both Spreen (1981) and Bruck concluded that childhood learning disabilities were not precursors of asocial behaviors. Bruck reported that there were no differences between LD and control groups in the rates of drug or alcohol abuse. In the Bruck and Spreen studies, control and LD subjects did not differ in terms of the number of delinquent acts or encounters with the police. Bruck found no differences in the rates of incarceration; however, Spreen reported that the LD group received somewhat more frequent and severe penalties, despite no differences in the number of offenses committed. It is possible that the more severe penalties reflect the LD subjects' more inappropriate or less persuasive behavior in court. This interpretation is consistent with reports that LD children do not present a good first impression and do not ingratiate themselves with their listeners (Donahue and Bryan 1984; Perlmutter and Bryan 1984). The findings of no association between asocial behavior and learning disabilities are clearly inconsistent with other studies that are characterized by their retrospective and correlational nature (see Spreen 1981 for a review). However, by virtue of their design, retrospective studies do not indicate whether LD children are more at risk for becoming delinquent than a nondisabled group; they merely show that delinquent or incarcerated samples include many subjects with poor learning skills or histories. In contrast, prospective studies such as Spreen's and Bruck's demonstrate that LD children who are followed into late adolescence and early adulthood commit delinquent offenses, but these are similar in frequency and in quality to those reported by subjects who did not experience academic problems. In terms of the second issue, there are some data to indicate that like LD children, LD adults may be at risk for social and emotional problems. Based on parents' ratings of their children's social and emotional adjustments at age 18, Spreen (Peter and Spreen 1979) found that LD subjects differed from controls on only one of six dimensions of problem behaviors--asocial acting out. However, on all self-rating scales, these LD subjects perceived themselves to be less socially and emotionally adjusted than control subjects. Based on clinicians' ratings of interview data, Bruck found that LD and control subjects showed similar adjustments in the areas of family relationships and interactions with the opposite sex. However, LD subjects suffered more severe peer relationship problems with members of the same sex; 13 percent of the LD subjects were lonely with few social contacts (no control
260
MINDS AT WORK
subjects showed this pattern). They also suffered higher rates of poor psychological adjustment; they showed specific difficulties in controlling their tempers and in dealing with frustrations in a constructive way. While the learning disabled and control groups were differentiated on statistical grounds, the clinical significance of these findings is not clear. It is possible that between-group differences simply reflect that LD adults are less happy or less socially involved than control subjects; on the other hand, they may indicate that LD adults are unhappy or socially isolated. Bruck suggests that only a small proportion of LD adults showed serious adjustment problems. Only 5 percent of the LD sample had received recent counseling for social and emotional problems. Approximately 15 percent of the LD subjects would have been appropriate referrals for family counseling, social skills training, or individual psychotherapy; but none were disturbed sufficiently to require hospitalization. Both Bruck and Spreen identified some characteristics of LD adults most likely to show adjustment problems. Bruck found that adults with adjustment problems also showed similar problems as children. It should be noted, however, that many showed improvement from childhood to follow-up; over 50 percent of subjects with poor childhood adjustment showed good adjustment at follow-up. In both studies, LD female subjects showed the poorest adjustments, and, according to Bruck's data, females were more likely to show persistent childhood problems. In the case of the Bruck stud~ sex differences were specific to social and emotional functioning only; females and males did not differ in terms of severity of childhood learning disabilities, literacy skills, occupational or educational achievements. The finding that LD females were at risk for poor adjustment not only characterizes samples of LD adults but characterizes some samples of LD children (see Bruck 1986 for a review). The poorer adjustments of female subjects may reflect differences in society's reactions to failure in males and females. Females who fail may be more rejected by peers and less accepted by adults than boys who fail. These interactions may be the precursors of social and emotional problems of LD females. Bruck reported several other characteristics that were associated with poor adult adjustments. Age at follow-up was related to psychological adjustment; poor adjustments were characteristic of the younger LD group (17-20-year-olds) but not of the older group (21-29year-olds). Educational success was associated with improved peer relationships. This may reflect greater opportunities for peer interaction in educational than in occupational milieus. However, severity of childhood learning disability was not related to family, peer, or psychological adjustments at follow-up. To summarize, like LD children, LD adults were more likely to
ADULT OUTCOMES OF CHILDREN WITH L~ARNING DISABILITIES
261
show adjustment problems than normal peers. The finding that severity of learning disabilities was not related to measures of social and emotional adjustment suggests that, at least in the adult population, social and emotional difficulties are widely distributed and are not confined to the more severe cases of learning disabilities. However, it should also be noted that the rates of reported problems were low, and at least in Bruck's study, most of the subjects were well-adjusted.
Conclusions
Several conclusions can be drawn from these four studies concerning the natural history and long term consequences of childhood learning disabilities. First, learning disabilities are a lifelong disorder; they persist into adulthood. Results of interview and standardized test data indicate that LD adults continued to experience some difficulty in the areas of literacy skills. However, most subjects had sufficient skills to function in academic and skilled occupational environments. While these data document the persistence of learning disabilities, little is known about the underlying nature of LD adults' literacy problems. Future studies should examine the cognitive processes that LD adults use for reading and spelling and should identify the compensatory mechanisms that LD adults may employ for reading and spelling. The data also indicate that the majority of adults with childhood diagnoses of learning disabilities are well-adjusted and successful in major areas of their adult lives. For example, in all but the Spreen study, LD subjects were not at risk for high school drop-out. The majority of LD subjects entered a university program, where despite their persistent problems, many were academically successful. LD adults were not at risk for unemployment or underemployment. They were represented in a wide range of occupations, and few held unskilled jobs. The poor occupational outcomes of Spreen's subjects highlight the importance of high school graduation as a prerequisite for successful educational and occupational achievements. LD students who are provided with an adequate and stimulating educational program will have acquired the motivation and skills to continue their education and to become gainfully employed. The data on social and emotional functioning indicate that there is no association between childhood learning disabilities and socially deviant behaviors such as delinquency, alcohol abuse, or drug abuse. Although the data show that childhood learning disabilities are not precursors of extreme forms of adult psychopathology, it seems that some LD adults, especially females, experience social and emotional problems. More refined measures must be used in future studies to indicate the clinical significance of these poorer adjustments.
262
MINDS AT WORK
A comparison of the results across the four studies along with the results of analyses within specific studies suggest some factors that are associated with positive adult outcomes. One factor is severity of childhood disability. Some analyses indicate that the most severely disabled children were most at risk for poor outcomes in terms of educational, occupational, and literacy skills. Other analyses indicated that socioeconomic status and IQ accounted for individual differences in outcomes within and across studies. Thus, not surprisingl3~ subjects from lower socioeconomic backgrounds or of marginally average IQs had the poorest outcomes, and studies that included high proportions of such subjects yielded the least optimistic outcomes. While the poorer outcomes found in such studies may not be associated with the childhood syndrome of learning disabilities, but with other secondary factors (e.g., IQ), more research is required to determine whether the presence of these factors act additively or synergistically with learning disabilities. The results suggest that the most important antecedents of positive outcome are early identification accompanied by adequate intervention. In addition to their direct educational benefits, it also appears that these antecedent conditions may have emotional and motivational benefits that contribute to positive outcomes. In Bruck's study, for example, early identification was accompanied by counselling for parents and teachers to explain the nature of the child's problems. These adults" subsequent understanding and support may have decreased many of the pressures on these children and led to more positive attitudes toward education, learning, and their own self-concepts. It was this dimension of intervention, rather than dimensions of IQ, socioeconomic status, and severity of childhood disability, that most differentiated Spreen's samples from those examined in the Bruck, Rawson, and Hopkins studies. It is possible that lack of adequate services for Spreen's subjects was an antecedent of high school drop-out which preceded poor employment opportunities. By emphasizing the importance of early identification paired with intervention for positive adult outcomes, one also emphasizes the fact that childhood learning disabilities are not a trivial condition that can be left untreated. On the other hand, it is suggested that with adequate treatment, learning disabilities are not necessarily a lifelong handicapping condition. With increased awareness of this condition and with increases in the delivery of services for these children, it is hoped that in the future the prognoses of adult outcomes will become more positive. Refeycnces Balow, B. J. and Blomquist, M. 1965. Young adults 10-15 years after a severe reading disability. Elementary School Journal 66:44-48.
ADULT OUTCOMES OF CHILDREN WITH LEARNING DISABILITIES
263
Bruck, M. 1985. The adult functioning of children with specific learning disabilities: A follow-up study. In I. Siegel (ed.). Advances in Applied Developmental Psychology. New Jersey: Ablex. Bruck, M. 1986. Social and emotional adjustments of learning disabled children: A review of the issues. In S. Ceci (ed.). Handbook of Cognitive, Social and Neuropsychological Aspects of Learning Disabilities. Vol. 1 Hillsdale, N.J.: Erlbaum. Ceci, S. 1986. Handbook of Cognitive, Social and Neuropsychological Aspects of Learning Disabilities. Hillsdale, N.J.: Erlbaum. Childs, B., Finucci, J. M., Pulver, A. E., and Tielsch, J. 1982. The natural history of specific reading disability. Educational outcomes. Unpublished manuscript. Department of Pediatrics, Johns Hopkins University. Donahue, M., and Bryan, T. 1984. Communicative skills and peer relations of learning disabled adolescents. Topics in Language Disorders 1:10-21. Finucci, J., Gottfredson, L. S., and Childs, B. 1985. A follow-up study of dyslexic boys. Annals of Dyslexia 35:117-136. Frauenheim, J. G., and Heckerl, J. R. 1983. A longitudinal study of psychological and achievement test performance in severe dyslexic adults. Journal of Learning Disabilities 16:339-347. Gottfredson, L. S., Finucci, J. M., and Childs, B. 1984. Explaining the adult careers of dyslexic boys. Variations in critical skills for high level jobs. Journal of Vocational Behavior 24:355-373. Herjanic, B., and Penick, E. 1972. Adult outcomes of disabled child readers. Journal of Special Education 6:397-410. Mauser, A. J. 1974. Learning disabilities and delinquent youth. Academic Therapy 9:389402. Muehl, S., and Forell, E. 1974. A follow-up study of disabled readers: Variables related to high school reading performance. Reading Research Quarterly 9:110-123. Peter, B. M., and Spreen, O. 1979. Behavior rating and personal adjustment scales of neurologically and learning handicapped children during adolescence and early adulthood: Results of a follow-up study. Journal of Clinical Neuropsychology 1:75-92. Perlmutter, B. F., and Bryan, J. 1984. First impressions, ingratiation, and the learning disabled child. Journal of Learning Disabilities 17:156-161. Preston, R. C., and Yarrington, D. J. 1967. Status of 50 retarded readers eight years after reading clinic diagnosis. Journal of Reading 11:122-129. Rawson, M. 1968. Developmental Language Disability. Baltimore: Johns Hopkins Press. Robinson, H. M., and Smith, H. 1962. Reading clinic clients--ten years after. Elementary School Journal 63:22-27. Schenck, B. J., Fitzsimmons, J., Bullard, P. C., Taylor, H. G., and Satz, P. 1980. A prevention model for children at risk for reading failure. In R. Knights and D. J. Bakker (eds.). Treatment of Hyperactive and Learning Disordered Children. Baltimore: University Park Press. Schonhaut, S. and Satz, P. 1984. Prognosis of the learning disabled child: A review of follow-up studies. In M. Rutter (ed.). Behavioral Syndromes and Brain Dysfunction in Childhood. New York: Guilford Press. Spreen, O. 1981. The relationship between learning disability, neurological impairment and delinquency: Results of a follow-up study. Journal of Nervous and Mental Diseases 169:791-802. Spreen, O. 1982. Adult outcomes of reading disorders. In R. N. Malatesha and P. G. Aaron (eds.). Reading Disorders, Varieties and Treatments. New York: Academic Press. Spreen, O. 1984. A prognostic view from middle childhood. In M. Levine and P. Satz (eds.). Middle Childhood: Developmental variations and dysfunction. New York: Academic Press.
