Gender Bias and Special Education Referrals Kristen G. Anderson Hampshire EducationalCollaborative North Hampton,Massachusetts
Referrals for special education determine who receives special education services. The most common referring agents are regular education teachers. Today, there is a large discrepancy between the number of females to males found in learning disabilities programs. However, studies have found the number of female and male students with learning disabilities in schools is more proportional than enrollment in special education suggests. The position of this paper is that gender bias among referring agents is a major factor in the unequal distribution of males and females in learning disabilities programs. The learning disabled population has shown the largest growth among exceptionalities since the passage of Public Law 94-142, Education for All Handicapped Children Act in 1975 (Learner 1993). The diagnosis of learning disabled has become one of "the most c o m m o n n e u r o b e h a v i o r a l d i s o r d e r s " i d e n t i f i e d among children (Shaywitz et al. 1990, p. 998) Our school systems have had to move rapidly to provide a range of services for this growing population. With such a large population of individuals being served by our educational system, research in learning disabilities has become of great importance. One of the most interesting facets of this population becomes quickly apparent when looking at its demographics--the p o p u l a t i o n is p r e d o m i n a n t l y male. The US D e p a r t m e n t of Education reported 72% of the learning disabled population as Annals of Dyslexia,Vol.47, 1997 Copyright©1997by The InternationalDyslexiaAssociation ISSN 0736-9387
151
152
DrAcNosrzc Issues
male, 28% female (Lerner 1993). Other estimates range as high as fifteen to one, males to females, in learning disabilities programs (Vogel 1990). In 1992, the state of Iowa reported their population of learning disabled to be approximately 70% male, 30% female (Kavale and Reese 1992). There have been numerous causes attributed to the preponderance of boys in special education programs. These causes center around the issues of: a possible gender-based referral bias, the effects of classroom behavior on the implementation of referrals, poor criteria in place for the definition of learning disabilities in team decisions, and the role of genetics in the development of learning disabilities. Flynn and Rahbar (1994) attempted to determine whether a proportionally higher number of boys than girls experienced reading failure. They analyzed a sample of 708 children participating in a longitudinal study of reading development in the Midwestern United States. The sample was analyzed using their scores on group achievement tests, teacher referrals to learning disabilities programs, and Chapter One programs in both the first and third grades. Students' reading failure was determined by their scores on the group achievement test. The researchers found a ratio of 1.4:1 (boys:girls) in grade one and 1.3:1 (boys:girls) in grade three in groups determined to be "Severely Learning Disabled" (Flynn and Rahbar 1994, p. 68). However the ratio of teacher-based referrals to learning disabled program, in both grades, was 2:1 (boys:girls). Shaywitz et al. (1990) compared the prevalence of reading disabilities in research-identified and school-identified groups. They studied a group of 215 girls and 199 boys from kindergarten through third grade. The authors found there was no significant difference in the amount of research identified, reading disabled girls versus boys. However, significantly more boys than girls were identified as reading disabled through the school's identification criteria. Wadsworth et. al. (1992) investigated the ratio of reading disabled individuals to their siblings, using data from five different studies of learning disabilities. They found there were differences in the ratio of reading disabled males to reading disabled females within each study, based on how samples were selected. "Within the five individual samples of reading-disabled children included in the present analysis, an excess of male probands occurred only in the referred or clinic sample. In the two twin studies in which subjects were ascertained systematically using research criteria, gender ratios of approximately 1:1 were obtained" (p. 1235).
GENDER BIAS AND SPECIALEDUCATION
153
Mellard and Byrne (1993) studied the status of California's community college programs for learning disabled students. They were interested in the referral process, eligibility outcomes, and services of the 103 community college learning disabled programs. In the area of gender, the referrals and populations served showed alternating differences, year to year, of the proportion of males to females. There was a slightly higher percentage of males served in the learning disabled programs (approximately 4%) than females. These results vary greatly from the ratios found in the K-12 system, 3:1 or higher. The authors noted that the referral process for postsecondary students was significantly different from the K-12 process because the students referred themselves to the programs. The authors felt the differences in the number of females to males served, between self-referral and teacherreferral systems, could be linked to bias. In a study of ADD, Berr~ Shaywitz, and Shaywitz (1985) investigated the difference among individuals with attention-deficit disorder and attention-deficit disorder with hyperactivity. They reported the ratio of males to females diagnosed with ADD range from 4:1 to 6:1. The report's purpose was to analyze the differences among males and females with ADD to determine possible factors pertinent to the difference in diagnosing this disorder. The authors found significant differences among females and males in the areas of age of referral, intellectual dysfunction, school behavior, peer relations, and group interaction. They concluded, "Girls with ADD may represent an underidentified and underserved group of children that is at significant risk for long-term academic, social, and emotional difficulties. Clinical samples were unavoidably subject to biases in referral, and unfortunately, many prevalence studies have depended on parent or teacher identification, without careful attention to those specific behavioral features specific to case inclusion, also allowing for biases in ascertainment" (Berry, Shaywitz, and Shaywitz 1985, p. 808). States have identified those individuals who can refer students to special education. For example, the Commonwealth of Massachusetts, in Chapter 766, outlined six individuals who could refer such students: a school official, including the classroom teacher; social workers; parents; a judicial officer; a family physician; or the child's custodian (Massachusetts Department of Education, Chapter 766 Regulations). The most common referral agent, however, is the regular education teacher. In a study of Iowa's learning disabled population, researchers found 74% of pupils were referred to special education services by a teacher (Kavale and Reese 1992). Ysseldyke et al. (1983) found
154
DIAGNOSTIC ISSUES
teacher referrals carried heavy weight in the team's decision to place a student in special education. Of those students referred to special education, 73% were d e t e r m i n e d eligible for special education services. If teachers are the most c o m m o n referral agent, what factors influence their decision to refer? Often student disruption of the classroom is a primary cause. In Gregory (1977), teachers were asked to make recommendations for referral on hypothetical case studies of children, including information as to the children's genders. One h u n d r e d and forty teachers were surveyed. Boys were more likely to be referred for special education services than girls with an identical profile, except in the area of reading disability (Vogel 1990). Ysseldyke et al. (1983) found teachers tended to refer students w h o bothered them in the classroom. The definition of bothersome behavior varied from teacher to teacher. Berry, Shaywitz, and Shaywitz (1985) concluded, " . . . selected referral may represent an important factor in determining the sex ratio, i.e., boys with attentional problems and hyperactivity are more likely than girls to be referred for evaluation because they exhibit more disruptive behaviors that are troublesome to adults" (p. 801). Mirkin (1982) studied the effects of behavior on referrals to special education. She compared teacher referrals to a systematic referral system based on weekly academic outcomes. The author found teacher referrals were 80% male, while the measurement system referrals were only 65% male. Mirkin d e t e r m i n e d that teachers were more likely to base their decision for referral on the disruptive behavior of the student rather than on academic difficulties. Because more male students displayed disruptive behaviors in the classroom, they were more likely to be referred to services for the learning disabled. This remained consistent even if female students had as significant academic impairments as their male counterparts (Vogel 1990). Shaywitz et al. (1990) also identified behavior as a pivotal factor in school's identification of students as reading disabled, " . . . children who met objective criteria for RD but w h o were not identified by their schools as reading disabled were significantly more likely to be rated by their teachers as having no behavioral problems" (p. 1001). If classroom behavior is so pivotal to the decision to refer students to special needs services, w h y is gender a factor? What differences are there between the behavior of females and males in the classroom that makes males more noticeable? Naiden (1976), in her discussion of the disparate make-up of Seattle's reading disabled classes, outlined the referral process
GENDER BIAS AND SPECIALEDUCATION
155
and the possible effects of bias. "A potential LLD student is first identified and referred by the classroom teacher. . . . There are several possibilities, as yet unverified, for why classroom teachers decide boys are learning disability students four times as often as girls, when in reality there are three boys to every two girls among disabled readers" (p. 441). Naiden identifies differential classroom behavior as being a potential cause for the differential make up of these classes. "The foremost reason is undesirable behavior. The low achieving boy shows his frustration in more overt ways than the low achieving g i r l . . . The low achieving girl does not become a behavior problem as often as the low achieving boy. Teachers will readily find a reason to seek help for a disruptive student" (p. 442). Sadker and Sadker (1994) i n v e s t i g a t e d the differences among boys and girls in the areas of classroom behavior, grades and standardized testing, socialization and self-esteem, and long-term success in education. The authors see these gender differences in the educational setting as socialized by a culture continuing to value the contributions of males over females. They see this cultural phenomenon as being demonstrated in our classrooms every day. At the surface level, girls seem to be doing better than boys in the elementary grades: better grades, fewer punishments. "Quieter and conforming, they are the elementary school's ideal students. 'If it ain't broke, don't fix it' is the schools operating principle as girls' good behavior frees the teacher to work with the more difficult-to-manage boys. The result is that girls receive less time, less help, and fewer challenges. Reinforced for passivity, their independence and self-esteem suffer--girls are penalized for doing what they should and lose ground as they go through school. In contrast, boys get reinforced for breaking the rules; they are rewarded for grabbing more than their fair share of the teacher's time and attention" (Sadker and Sadker 1994, p. 44). Through girls "good" behavior, they lose out on teacher's attention. This quieter behavior, the authors feel, is one reason girls in need of special education are overlooked by teachers. They have not disrupted the classroom, so they go unnoticed. Leinhardt, Seewald, and Zigmond (1982) examined learning disabilities classrooms. Their findings of girls' behavior in a learning disabilities classroom were similar to Sadker and Sadker's. The authors found girls got more rewards and waited more in the classroom. They attributed this to, " . . . 'girls play school'" better and, rather than clearly going off task, they wait
15 6
DIAGNOSTICISSLIES
or sharpen pencils or "look busy." It may also be that it takes them longer to get materials and they are unable to get the teacher's attention, so they wait longer" (p. 839). This research seems to indicate girls behave differently in the classroom, and this conforming behavior leads them to be "passive" or "waiting" (Sadker and Sadker 1994; Leinhardt, Seewald, and Zigrnond 1982). Shaywitz et al. (1990) also found differences in classroom behavior for boys and girls, both disabled and nondisabled. "We found significant gender differences between normal boys and girls in the school setting. In general, teachers rate boys as significantly more active, more inattentive, and less dexterous and as having more problems in behavior, language, and academics than their female peers. Yet, despite teacher reports of difficulties in the classroom, the measured overall ability and achievement of boys is comparable to girls" (p.1001). In their comparison of research-identified versus school-identified reading disabled students, the influence of classroom behavior is apparent. The students targeted by the school were more likely to be described as having "an excess of activity and behavior problems" (p. 1001). Smart, Sanson, and Prior (1996) studied Australian students from Grades 2 through to Grade 4. They investigated the temporal and causal connections between reading disorders and behavior disorders. Among their sample, they did not find any significant differences in reading, spelling, and intelligence. However, they found that, "Sex differences were noted in the association, with two-thirds of reading-disabled boys also having behavior problems and two thirds of reading-disabled girls having no behavior problems" (p. 363). This finding clearly supports the hypothesis that learning disabled boys and girls behave markedly differently in the classroom. Further compounding this issue is a lack of clear criteria for the designation learning disabled. Many authors have criticized the lack of a clear definition of learning disabilities as being a factor in the differential make up of learning disabilities programs. Ysseldyke et al. (1983) reviewed information gathered in five years of research at the Minnesota Institute. In research performed at the institute on learning disabilities guidelines, there was much disagreement among practitioners and researchers on the criteria for learning disabilities designations. This lack of conformity led to difficulties in determining the incidence, prevalence, and populations served in learning disabilities programs. The authors cited research in which they examined the
GENDER BIAS AND SPECIALEDUCATION
157
criteria used for learning disabilities referrals. In the professional literature, they found 42 definitions of learning disabilities. After s e l e c t i n g 17 of t h e s e d e f i n i t i o n s to use as a classification system, they found that, "more than 80% of normal students" and "over 75% of low achieving students" met criteria for learning disabilities on one or more definitions. "Yet only 75% of the students labeled learning disabled met at least one set of criteria for being considered LD" (Ysseldyke et al. 1983, p. 79). Berry, Shaywitz, and Shaywitz (1985) examined gender differences among a sample of the ADD population. They found significant gender differences in referrals for special education services. Their recommendation for school districts was to define "uniform criteria" to use in the classification of girls with ADD (p. 808). Leinhardt, Seewald, and Zigmond (1982) found "a lack of uniformity" in placement criteria for students in learning disabilities programs. The authors stated, "Although variation by individual is expected and valued, variation that is associated with group membership is not" (p. 838). If the criteria used for identifying who is learning disabled is unclear, the chance for bias in determination is less easily controlled. Another researcher, Vogel (1990), questions the standard by which learning disabled students are compared. Schools and many states have criteria in place to label a student learning disabled. These definitions are often garnered from available research on the intellectual and behavioral characteristics of the learning disabled. Vogel states, "There are several possible explanations for the underidentification of females with learning disabilities. The first relates to a mismatch between the child's problems and the screening or referring agent's expectations of a learning disabled child. This mismatch is a consequence of the scarcity of research regarding the behavioral manifestations and clinical profiles of females with learning disabilities" (p. 50) Because there is so little research on females with learning disabilities, the generalization of research done on males with learning disabilities may be in question. Research on referral standards, referral patterns, and classroom behavior makes a compelling argument that gender has a significant role in the placement of students in special education programs. Another school of thought, commonly held by practitioners, is that biology and genetics are central to the gender distribution in learning disabilities programs. Nass (1993) reviewed the literature on the hormonal sex differences, maturation rate differences, genetic factors, and the in-
158
DIAGNOSTIC ISSUES
fluences of perinatal brain injury on the development of learning difficulties. Nass believes that males are at a higher risk, or cannot compensate as well as females, for learning disabilities at a biological level. This suggests the genetic factor may be in both males and females. However, females may be able to overcome this factor because of a sex-based propensity for language proficiency. These factors may lead to a larger number of males who display learning disabilities symptoms and therefore are referred to programs for learning disabled students. Wolf and Gow (1985/86) studied reading acquisition in children. They identified learning style differences in reading acquisition between males and females. Females were more proficient in the rate of linguistic processing and basic word recognition. Male proficiency was in semantic-based relationships. The authors hypothesized that the differential make-up of the reading disabled population could be related to these differences in reading acquisition. Boys reading development could be hindered by their reliance on semantic strengths, creating "an early arrested strategy that is inappropriate to reading acquisition" (Wolf and Gow 1985/86, p. 102). Pennington (1988) believed females were "protected" by environmental a n d / o r sex related factors that make them stronger in the areas in language acquisition and proficiency. These factors could help negate any genetic proclivities toward learning disabilities. " . . . even if females acquire the specific gene that puts them at risk for dyslexia, they may be better able to compensate and, therefore, are less likely to express the gene" (as cited by Vogel 1990, p. 50). Theories surrounding a biological basis for the larger population of males with learning disabilities need to be more fully investigated. Nass (1993) cautioned that the majority of these theories are based on speculation and that caution needs to be exercised in examining the results. Information from other developmental disabilities, such as fragile X syndrome are often generalized to learning disabilities. The findings on other developmental disorders may not be applicable to learning disabilities. A clear connection has yet to be made. Often research on learning disabilities includes only individuals already in special education programs, a large proportion of w h o m are predominantly male (Vogel 1990). Further research on the causes of learning disabilities and its effects on both males and females needs to occur. Flynn and Rahbar (1994) report that there is a strong belief in the theory of a sex-based, genetic etiology for learning disabilities. The authors feel this
GENDER BIAS AND SPECIAL EDUCATION
159
belief is reflected in the higher number of boys in learning disabilities programs, and the dominance of research on learning disabilities using male samples. It is possible this belief influences a teacher's decision to refer males, more often than females, to learning disabilities programs. Shaywitz et al. (1990) hypothesized that the increased numbers of males to females identified as reading disabled in studies was a function of sampling bias, based on the use of samples identified by referrals to services. They felt that the influence of referral bias significantly effected the populations often studied in research. "Our data indicate that school-identified samples are almost unavoidably subject to a referral bias and that reports in the literature of an increased prevalence of RD in boys may reflect this bias in ascertainment. Caution is suggested in generalizing from research studies based on school-identified populations" (p. 1002). On the basis of the research cited above, we can make some tentative hypotheses regarding gender and special education referrals: 1. 2. 3.
4.
5.
Teacher referrals are often affected by the gender of the student referred (Flynn and Rahbar 1994; Vogel 1990; Naiden 1976). Teacher referrals are heavily influenced by the student's classroom behavior (Yesseldyke et al. 1983; Berry, Shaywitz, and Shaywitz 1985; Naiden 1976). Males and females tend to display different behavior in the classroom. Males often display more hyperactive or disruptive behavior in the classroom (Smart, Sanson, and Prior 1996; Sadker and Sadker 1994; Shaywitz et al. 1990; Berry, Shaywitz, and Shaywitz 1985; Leinhardt, Seewald, and Zigmond 1982; Naiden 1976). Standards regarding the placement of individuals in learning disabled programs lack clear definition and are often based on male norms (Vogel 1990; Yesseldyke et al. 1993; Berry, Shaywitz, and Shaywitz 1985; Leinhardt, Seewald, and Zigmond 1982). There is a strong belief in a sex-based, genetic, or biological etiology of learning disabilities which is reflected in the make up of learning disability programs (Flynn and Rahbar 1994; Nass 1994; Vogel 1990; Wolf and Gow 1985/86).
There is compelling evidence pointing to gender bias as a central factor in the referral of learning disabled students to spe-
160
DIAGNOSTIC ISSUES
cial educational services. If this is the case, what can be done to minimize this effect? Flynn and Rahbar (1994) suggest more careful monitoring of girl's performance in the classroom. Vogel (1990) feels further research on learning disabilities, its etiology and potential profile differences of learning disabled males and learning disabled females, needs to occur. Sadker and Sadker (1994) indicate that better teacher education in gender issues and systemic change is necessary to minimize the damage done to individuals in our educational structure. As an example, a recent text on learning disabilities used for teacher preparation devotes only one paragraph to gender differences in the make-up of learning disabled programs (Lerner 1993). As teachers, we are obligated to meet the needs of our students to the best of our abilities. In the times of huge student/ teacher ratios in our regular education classroom, especially in urban public schools, it is often difficult to identify students at risk. Often, as practitioners, our decision to refer is based on the student most easily identified--the disruptive student, the active student, or the truant student. To better assess the students who need special services, both male and female, we need to look beyond overt behavior in the classroom. As stated by Flynn and Rahbar (1994), more careful monitoring of student performance needs to occur. I would propose the following suggestions to help monitor student performance. (1) Use teaching strategies that allow more direct contact with students to identify those at risk for learning difficulties. For example, small, cooperative group work, where the teacher can move from group to group, can help identify those team members who require more assistance to complete tasks. (2) Require one-to-one meetings with students to identify student needs. For example, if there seems to be a discrepancy between a student's academic performance and his or her cognitive abilities, a learning disabilities assessment should be conducted. (3) Increase awareness of those students who "quietly" fail. It is important to remember students can express frustration with tasks in many different ways. Acting out is only one example. (4) Use professional resources. These include other teachers and the special education staff. If a student is having problems in class, are these problems prevalent in all of his or her classes? Do the causes of these problems become apparent? Can the special education staff identify possible warning signs? Along with increased awareness of our individual student's needs, it is necessary for all of us in the field of education to
GENDER BIAS AND SPECIAL EDUCATION
161
push for further research on the needs of learning disabled girls and w o m e n . Further research, education, and cultural awareness pertaining to gender must occur to minimize its effects on the success of individuals in education and society. Practitioners need to recognize that gender does affect the decisions w e make in the classroom. As more and more research becomes available, we will be able to identify those factors influencing our decisions regarding services, create better teacher preparation programs, and develop better methods for serving students' needs, both male and female. Regardless of the basis of bias, we need to work to inhibit its influences on our behavior. We need to e x a m i n e closely the progress of all students in the classroom, whether male or female, to make decisions that best suit their individual needs.
References Berry, C., Shaywitz, S., and Shaywitz, B. 1985. Girls with attention deficit disorder: A silent minority? A report on behavioral and cognitive characteristics. Pediatrics 76(5):801-09. Flynn, J., and Rahbar, M. 1994. Prevalence of reading failure in boys compared with girls. Psychology in the Schools 31:66-71. Kavale, K., and Reese, J. 1992. The character of learning disabilities: An Iowa profile. Learning Disabilities Quarterly 15:74-94. Leinhardt, G., Seewald, A., and Zigmond, N. 1982. Sex and race differences in learning disabilities classrooms. Journal of Educational Psychology 74(6):83543. Lerner, J. 1993. Learning Disabilities: Theories, Diagnosis & Teaching Strategies. Boston: Houghton Mifflin. Massachusetts Department of Education 1991. 766 Regulations. Boston: Massachusetts Department of Education. Mellard, D., and Byrne, M. 1993. Learning disabilities referrals, eligibility outcomes, and services in community colleges: A four year summary. Learning Disabilities Quarterly 16:199-217. Naiden, N. 1976. Ratio of boys to girls among disabled readers. The Reading Teacher Feb. 1976, 439-42 Nass, R. 1993. Sex differences in learning abilities and disabilities. Annals of Dyslexia 43:61-77. Pennington, B. 1988. November. Genotype and phenotype analysis of familial dyslexia. Paper presented at the 39th conference of The Orton Dyslexia Society. Tampa, FL. As cited in Vogel 1990. Sadker, M., and Sadker, D. 1994. Failing at Fairness: How Our Schools Cheat Girls. New York: Touchstone. Shaywitz, S., Shaywitz, B., Fletcher, J., and Escobar, M. 1990. Prevalence of reading disability in boys and girls: Results of the Connecticut longitudinal study. Journal of the American Medical Association 264(8):998-1002. Smart, D., Sanson, A., and Prior, M. 1996. Connections between reading disability and behavior problems: Testing temporal and causal hypotheses. Journal of Abnormal Child Psychology24(3):363--83.
162
DIAGNOSTICISSUES
Vogel, S. 1990. Gender differences in intelligence, language, visual-motor abilities, and accademic achievement in students with learning disabilities: A review of the literature. Journalof Learning Disabilities23:44-51. Wadsworth, S., DeFries, J., Stevenson, J., Gilger, J., and Pennington, B. 1992. Gender ratios among reading-disabled children and their siblings as a function of parental impairment. Journal of Child Psychiatry 33:1229-39. Wolf, M., and Gow, D. 1985/86. A longitudinal investigation of gender differences in language and reading development. First Language 6:81-110. Ysseldyke, J., Thurlow, M., Graden, J., Wesson, K., Algozzine, B., and Deno, S. 1983. Generalizations from five years of research on assessment and decision making: The University of Minnesota Institute. Exceptional Children Quarterly 4(1): 75-93.
A d d r e s s c o r r e s p o n d e n c e to: Kristen G. A n d e r s o n , 3099 K i r k l e v i n g t o n Dr., #102, Lexington, KY 40517.