Full opinion text
FINDINGS OF FACT, CONCLUSIONS OF LAW AND ORDER OPINION SOTOMAYOR, District Judge. INTRODUCTION This case, tried to the bench in 21 days of testimony accompanied by exhibits and briefs aggregating to more than 5000 pages, principally devolves to the meaning of a single wordi — substantially—as used in the Americans with Disabilities Act (the “ADA”), 42 U.S.C. §§ 12101-12213 (1995) and the Rehabilitation Act, 29 U.S.C. §§ 701-796 (1985) (“Section 504” or the “Rehabilitation Act”). Both Acts define a disability as “a physical or mental impairment that substantially limits one or more of’ an individual’s “major life activities.” 42 U.S.C. § 12102(2)(A) (1995 Supp.); 29 U.S.C. § 706(8)(B) (1996 Supp.) (emphasis added). Plaintiff claims she suffers from a learning disability that impairs her reading and her ability to be able to work as a lawyer. At issue in this case is whether plaintiff suffers from an impairment, and if so, whether it rises to the level of a substantial limitation cognizable under the ADA, thus entitling her to accommodations in taking New York State’s Bar Examination. She sues for injunctive and other relief under Titles II and III of the ADA, Section 504 of the Rehabilitation Act, the Due Process and Equal Protection Clauses of the Fourteenth Amendment, and 42 U.S.C. § 1983. The evidence at trial has convinced me that Marilyn Bartlett suffers from a learning deficit that evinces itself as a difficulty in reading with the speed, fluency and automaticity of an individual with her background and level of intellectual ability. Despite this impairment, plaintiff obtained a Ph.D. in Educational Administration and a law degree. By virtue of superior effort and not a small amount of courage, Marilyn Bartlett has been able to succeed academically and professionally despite the limitations her impairment has placed upon her. But this ease asks whether, in light of the confined language of the law, plaintiff is not merely impaired, but disabled. The term “substantially limited” is defined in 29 C.F.R. § 1630.2(j)(1)(ii) as: (ii) Significantly restricted as to the condition, manner or duration under which an individual can perform a particular major life activity as compared to the condition, manner, or duration under which the average person in the general population can perform that same major life activity, (emphasis added). Similarly, with respect to the major life activity of working, “substantially limited” is defined by 29 C.F.R. § 1630.2(j)(3)(i) to mean “significantly restricted in the ability to perform either a class of jobs or a broad range of jobs in various classes as compared to the average person having comparable training, skills and abilities” (emphasis added). Regulations such as the foregoing must be accorded substantial deference because they reflect and incorporate active Congressional intervention in their fashioning. See School Bd. of Nassau County v. Arline, 480 U.S. 273, 279, 107 S.Ct. 1123, 1126-27, 94 L.Ed.2d 307 (1987) (citing Consolidated Rail Corp. v. Darrone, 465 U.S. 624, 634-35, 104 S.Ct. 1248, 1254-55, 79 L.Ed.2d 568, & nn. 14-16 (1984)) (construing regulations adopted pursuant to the Rehabilitation Act). For those of us for whom words sing, sentences paint pictures, and paragraphs create panoramic views of the world, the inability to identify and process words with ease would be crippling. Plaintiff, an obviously intelligent, highly articulate individual reads slowly, haltingly, and laboriously. She simply does not read in the manner of an average person. I reject the basic premise of defendants’ experts that a learning disability in reading can be identified solely by a person’s inability to decode, i.e., identify words, as measured by standardized tests, and I accept instead the basic premise of plaintiffs experts that a learning disability in reading has to be identified in the context of an individual’s total processing difficulties. Having witnessed all of the trial testimony and having studied the thousands of pages of exhibits, affidavits and depositions, I conclude that plaintiff is not able to read in the same condition, manner or duration as an average reader when measured against “the average person having comparable training, skills and abilities.” 29 C.F.R. § 1630.2(j)(3)(i). For this reason, I find that plaintiff is substantially impaired under the law, and she is therefore entitled to receive reasonable accommodations in taking the New York State Bar Examination. For the reasons to be discussed, I deny plaintiffs equal protection, due process, and § 1983 claims. I award her injunctive relief, and compensatory, but not punitive, damages. BACKGROUND I. UNDISPUTED FACTS The following consists substantially of undisputed facts taken from the joint pretrial order submitted by the parties. The Court has added, where indicated, some additional facts to this section in order to clarify or complete the presentation set forth in the undisputed facts agreed to by the parties. A. Parties Plaintiff is a law school graduate who has met all the qualifications necessary to take the New York State Bar Examination. Defendants John Holt-Harris, Jr., Richard J. Bartlett, Laura Taylor Swain, Charles T. Beeching, Jr., Ira P. Sloane, and James T. Fuller, as Executive Secretary, are the members of the New York State Board of Law Examiners (the “Board”), and as such are responsible for the administration of the New York State Bar Examination. B. The Bar Examination The Board is authorized to conduct a written bar examination, twice a year, consisting of legal problems in both “adjective and substantive law.” (N.Y.Comp.Codes R. & Regs, tit. 22, § 520.7 (“22 NYCRR”)). The Bar Examination is given over two days and tests the candidates’ knowledge of legal principles and concepts that are relevant and important to the practice of law. The Board’s mandate is to test for minimal competence to practice law. One day is devoted to answering the New York portion of the test, created by the Board and consisting of 50 multiple-choice questions and six essay questions. Unless an accommodation of extra time is granted for a disability, the New York portion of the test must be completed within six hours: a three-hour session in the morning and a three-hour session in the afternoon. The second day, which may be taken in another state, is devoted to the 200 multiple-choice questions of the Multistate Bar Examination (“MBE”), created by the National Conference of Bar Examiners. The MBE normally takes six and one-half hours. If the candidate elects to take the MBE in New York, it is administered by the Board as part of the New York State Bar Examination. A combined score of 660 on the MBE and the New York portion of the test is needed to pass the Bar Examination. According to trial testimony, spelling errors in responding to questions are not penalized on the Bar Examination. The Court accepts plaintiffs contention, however, that difficulties in spelling affects the clarity of the presentation and detracts from the expression of concepts. Title 22 NYCRR § 220.13 authorizes the Board to adopt, amend or rescind rules it deems necessary and proper to enable it to discharge its duties. Title 22 NYCRR § 6000.4(a) permits applicants to apply for accommodations for the Bar Examination based upon a disability. It is the policy of the Board to provide accommodations in testing conditions to candidates with disabilities to the extent such accommodations are reasonable, consistent with the nature and purpose of the examination, and necessitated by the candidate’s disability. The Board has provided, inter alia, the following accommodations to applicants with disabilities: granted access to food and drink, provided a private room in which to take the examination and large print examinations, permitted up to double the amount of time over two days to take the examination, and approved use of a computer or amanuensis to record answers. If the MBE is taken in New York by a candidate to whom the Board has granted accommodations, the same accommodations apply to the MBE portion of the test. To request accommodations, an applicant completes a form enclosed with the application and returns it with supporting documentation to the Board. See 22 NYCRR § 6000.4(b). The supporting documentation must state the nature of the candidate’s disability, the requested accommodation, the causal relationship between the disability and the applicant’s ability to take the Bar Examination without the requested accommodations, and the reason the specific accommodation requested by the candidate is required. See 22 NYCRR § 6000.4(c). The Board’s rules also require applicants to provide documentation of the three most recent testing accommodations, if any, granted to the candidate by academic institutions, licensure authorities, or other test administrators. See 22 NYCRR § 6000.4(c). The Board has the discretion to require applicants to provide additional information relating to the disability and/or prior accommodations, and may also request that applicants submit to an examination by an expert designated by the Board in connection with an applicant’s request for testing accommodations. See 22 NYCRR § 6000.4(d). If a requested accommodation is denied, either in whole or in part, the Board’s notification must state the reason for the denial. The candidate may appeal the decision to the Board. See 22 NYCRR § 6000.4(e). The Board must notify the applicant of its determination no later than twenty days prior to the date of the examination for which the accommodations are requested. Title 22 NYCRR § 6000.4(f) defines the term “disability” as a “physical, neurological or learning disability” and the term “candidates with disabilities” as an “otherwise qualified candidate having such disabilities.” The Board in its discretion may delegate to its members, its Executive Secretary or Deputy Executive Secretary, all or any part of its duties and responsibilities in granting or denying accommodations, with the exception of the responsibilities relating to appeals. See 22 NYCRR § 6000.4(g). C. Plaintiffs Educational Background In 1970, plaintiff received a B.S.Ed. in Early Childhood Teacher Education from the State College at Worcester, Massachusetts. She graduated with a grade point average of 2.10. Plaintiff did not receive accommodations while at State College. Plaintiff thereafter took the Graduate Record Examination without accommodations. In 1976, plaintiff received a M.Ed. in Special Education, Educational Disturbances in Children, from Boston University. She graduated with a grade point average of 3.8. Plaintiff did not receive accommodations while at Boston University. In the Fall of 1976, plaintiff entered the Ph.D. program in Educational Administration at New York University. Plaintiff first requested and received accommodations for the 1977 Summer semester. Plaintiff had not been formally diagnosed with a learning disability prior to receiving these accommodations. The Court accepts the plaintiffs and Dr. Evan’s testimony that then-Ph.D. Program Director, Seymour Evans, who had knowledge of and experience with learning disabilities, recommended plaintiff for accommodations after he had worked with her and noted her reading difficulties. New York University did not request, and plaintiff did not submit, any documentation of a learning disability in support of her request for accommodations. The accommodations granted to plaintiff at New York University included unlimited time to complete final examinations, unlimited time to take the written comprehensive examinations, use of an electronic typewriter with correction capability to take examinations, and the use of a department secretary as an amanuensis. Plaintiff was not granted accommodations for her examinations in statistics and administration, courses taught in another department. Plaintiff fulfilled her Ph.D. foreign language requirement by reading a passage in German and answering questions on the passage for the head of the German department. Plaintiff received her Ph.D. in 1981. Plaintiff did not request accommodations for the Law School Aptitude Test, and she scored 32 out of a possible 48. Plaintiff entered Vermont Law School in 1988. Plaintiff did not request accommodations during her first year of law school. Plaintiffs grade point average during that first year was 2.09, with a class ranking of 155 out of 166 students. Plaintiff first requested and received accommodations during law school for the Fall 1989 examination period. Plaintiff received accommodations for the Spring 1990, Fall 1990 and Spring 1991 examination periods as well-. The law school accommodations included time-and-a-half to take examinations, the use of a yellow legal pad with a red left margin instead of the traditional “blue book,” and permission to circle the answers on multiple choice examinations instead of filling in a computer-scored answer sheet. Plaintiffs grade point averages after receipt of the accommodations were: Fall 1989 — 2.58; Spring 1990 — 2.50; Fall 1990 — 1.82; Spring 1991 — 2.90. Plaintiff graduated from Vermont Law School in May 1991, with a cumulative grade point average of 2.32, and a class standing of 143 out of 153 students. D. Plaintiffs Relevant Employment History Except for periods during which she was preparing for the bar examinations or moving from one job to another, plaintiff has been continuously employed since graduating from law school. Upon graduating from law school, plaintiff worked at a New York law firm until December 1992, when her firm dismissed her because she failed the Bar Examination for the third time. In January of 1993 and until June of 1993, she worked with a client of her former law firm on a special project until its completion. After a number of months of unemployment during which time she could not find work in the legal profession, in September of 1993, plaintiff became a director of a day care center in Brooklyn, New York. In July of 1994, plaintiff returned to her former profession of educational administration, and is currently employed as an Associate Professor of Educational Administration at Dowling College. She receives accommodations at work for her reading problems in the form of a full-time work-study student who assists her in reading and writing tasks. While working at the law firm, plaintiff predominantly self-accommodated her disability (e.g., dictating instead of writing reports, not billing for the additional time it took her to complete tasks), although she was given a computer before other associates because of her writing difficulties. E. Plaintiffs Bar Exam Applications Plaintiff took the Multistate Professional Responsibility Examination (“MPRE”) in 1991 and received accommodations, including extra time, for that examination. The MPRE is not administered by the Board. In June 1991, fewer than 45 days before the examination, plaintiff applied, and requested accommodations, for the July 1991 Bar Examination. Submitted with plaintiff’s application was a Psyehoeducational Evaluation from Philip M. Massad, Ph.D., a clinical psychologist, which indicated that he evaluated plaintiff on November 30, 1989 and December 7, 1989. In his evaluation, Dr. Massad concluded that plaintiff has “dyslexia characterized by a deficit in phonological processing (DSM-III-R 315.00).” (Pl.’s Ex. 20a, at 5, Massad’s Psy- ' choedueational Evaluation.) On July 1, 1991, James Fuller, the Executive Secretary to the Board, advised plaintiff that because she had missed the deadline for applying for accommodations, her request was denied. Fuller further indicated that the Board did not consider the materials she had submitted as current, and that the scores she earned in 1989 on the Woodcock test— the test utilized by the Board to screen reading disabled applicants- — did not qualify plaintiff for accommodations. Fuller based his conclusion on the fact that the Woodcock Word Attack and Word Identification scores on plaintiff’s test were above the 30th percentile. Dr. Frank R. Vellutino, a research psychologist retained by the Board to advise it on policies relating to learning disabled applicants, had previously indicated to the Board and Fuller that scores above the 30th percentile generally did not identify an applicant as having a significant reading disability. (Fuller Aff. ¶ 52.) Vellutino, however, did not review plaintiff’s application at this time. Plaintiff failed the July 1991 Bar Examination with a score of 563 (a passing score is 660). In November 1991, plaintiff applied for the February 1992 New York State Bar Examination. Plaintiff did not request accommodations for this test. Plaintiff took and failed the February 1992 Bar Examination with a score of 580. In June 1992, plaintiff applied for the July 1992 Bar Examination. The parties dispute whether plaintiff applied for accommodations for this test. Plaintiff claims she did, but the Board has no record of the request. Plaintiff was not accommodated for the test, which she took and failed with a score of 576. In January 1993, plaintiff applied for the February 1993 Bar Examination, again requesting accommodations for her learning disabilities. The accommodations sought by plaintiff were unlimited/extended time to take the test, and permission to tape record her essays and to circle her multiple choice answers in the test booklet. Submitted with plaintiffs request for accommodations was Dr. Massad’s 1989 Psyehoeducational Evaluation, previously submitted by plaintiff, and a November 20, 1992 letter from Dr. Massad to plaintiff reasserting the opinion he set forth in his 1989 Evaluation. Upon receipt of this application, Fuller referred the file to Dr. Vellutino. After evaluating the materials submitted to him, Dr. Vellutino recommended that plaintiffs request for accommodations be denied. Based on Dr. Massad’s 1989 evaluation and his 1992 letter, Dr. Vellutino concluded that there was “no compelling documentation” of a learning disability and that the reading test data did not support a diagnosis of dyslexia. By letter dated January 20, 1993, Fuller forwarded the documentation relating to plaintiff and Dr. Vellutino’s recommendation to the Board. The Board denied plaintiffs request for accommodations. In a letter dated January 26, 1993, Fuller advised plaintiff that the documentation she had submitted was insufficient to establish a basis for granting the accommodations requested. Plaintiff appealed the Board’s decision denying her accommodations in a letter received by the Board on February 17, 1993. Plaintiff did not submit any additional documentation concerning her learning disability with the appeal. By letter dated February 18, 1993, Fuller advised plaintiff that her appeal was untimely. Fuller also advised plaintiff that following consultation with an expert in the field, the Board had determined that the documentation that plaintiff had provided did not support the finding of a disability warranting accommodations. Plaintiff took and failed the February 1993 Bar Examination with a score of 615. In May 1993, plaintiff applied for the July 1993 Bar Examination, again requesting accommodations. On plaintiffs application, plaintiff identified her disability as “learning disabilities — DSM III-R 315.00.” Plaintiff obtained a new evaluation from a clinical psychologist, Dr. Richard F. Heath. Plaintiff requested the following accommodations: extra time, use of a word processor or permission to dictate essay responses, and leave to circle answers on the multiple choice questions examination sheet. Fuller referred this application to Dr. Vellutino. Dr. Vellutino again recommended that plaintiffs request for accommodations be denied, affirming his original opinion that plaintiff did not have a reading disability. By letter dated June 29,1993, the Board advised plaintiff that the test profiles she had provided did not support a diagnosis of dyslexia, and therefore, her request for accommodations was denied. By letter dated July 12, 1993 from Jo Anne Simon, Esq. to Fuller, plaintiff submitted her application for reconsideration. Plaintiff included the following with her appeal: an affidavit by Stephanie J. Wilbanks, Associate Dean for Academic Affairs at Vermont Law School, attesting to the fact that plaintiff was provided accommodations during her final two years at law school; Dr. Massad’s and Dr. Heath’s Evaluations; a copy of a letter from Paul A. Cullinan, Ph.D., Chair of the Educational Administration Department at New York University, stating that plaintiff had received accommodations at New York University; and a notice from the Pennsylvania Bar Examiners advising plaintiff that she had been granted accommodations for the July 1993 Pennsylvania Bar Examination. Dr.' Heath also submitted a letter to the Board, dated July 3, 1993, wherein he reaffirmed his earlier evaluation and recommendation for accommodations. Dr. Vellutino reviewed the file and again concluded that plaintiffs scores as reported by Dr. Massad and Dr. Heath supported his earlier opinion that there was insufficient documentation to support a finding of a learning disability. Fuller so notified plaintiff on July 19,1993. This litigation was commenced on July 20, 1993. Pursuant to a Stipulation dated July 26, 1993 and so ordered by this Court, the parties agreed that plaintiff would receive accommodations during the July Bar Examination pending the outcome of this litigation. The Board gave plaintiff time-and-a-half — a period of nine hours — for the New York portion of the test and the use of an amanuensis to read the test questions to plaintiff and record her responses. In addition, the Board allowed plaintiff to mark the answers to the multiple choice portion of the examination in the question book rather than on the computerized answer sheet. Plaintiff elected to take the MBE in Pennsylvania. Pursuant to the terms of the Stipulation, if plaintiff passed the July 1993 New York State Bar Examination with accommodations, the results were not to be certified to the Court of Appeals unless she was successful in this litigation. Despite accommodation, plaintiff failed the July 1993 Bar Examination with a score of 597. At trial, plaintiff claimed the accommodations granted to her for this test were inadequate because she had had insufficient time to rest between the New York and Pennsylvania Bar Examinations or to practice with her amanuensis, an accommodation she had never previously used. She also complained that the proctor placed in her room caused distracting noises during the test. F. Other Bar Examinations In July 1993, plaintiff took the Pennsylvania Bar Examination and MBE with accommodations. The Pennsylvania Bar Examiners allowed plaintiff to mark her answers directly in the question booklet, gave her a separate room to take the test, granted her time-and-a-half — the maximum allowable time — and authorized her to use an amanuensis. Plaintiff did not pass the Pennsylvania Bar Examination despite the accommodations. G. Overview of the Applications Submitted to the Board for Accommodations In February 1992, the Board administered the Bar Examination to 2,231 applicants. Among the applicants, 71 requested accommodations; 65 were granted, 4 were denied and 2 requested accommodations but either did not apply for the February 1992 Bar Examination or withdrew. Of the 71 applicants, 13 requested accommodations on the basis of a learning disability; 10 requests were granted and 3 were denied. In July 1992, the Board administered the Bar Examination to 7,436 applicants. Of the applicants, 152 requested accommodations; 127 were granted, 7 were denied, 10 did not apply for the July 1992 Bar Examination or withdrew, one applicant passed the previous Bar Examination on appeal, 6 applicants did not provide additional documentation requested, and one applicant changed location due to a medical reason. Of the 152 applicants, 26 requested accommodations on the basis of a learning disability or attention deficit disorder; 21 requests were granted and 5 were denied. In February 1993, the Board administered the Bar Examination to 2,202 applicants. Among the applicants, 102 requested accommodations; 88 were granted, 8 were denied, 1 did not qualify, 4 did not apply for the February 1993 Bar Examination or withdrew, and one applicant passed the previous Bar Examination on appeal. Of the 102 applicants, 19 requested accommodations on the basis of a learning disability or attention deficit disorder; 16 requests were granted and 3 were denied. In July 1993, the Board administered the Bar Examination to 7,373 applicants. Of the applicants, 181 requested accommodations; 155 requests were granted, 16 were denied and 10 applicants did not respond to a request for additional information. Of the 181 applicants, 51 requested accommodations of the basis of a learning disability or attention deficit disorder; 37 requests were granted and 14 were denied. II. ADDITIONAL FACTS Based on the testimony presented and the exhibits admitted during the bench trial, my additional factual findings pursuant to Fed. R.Civ.P. 52 are as follows: A. Plaintiffs Psychoeducational Evaluations The evaluations of plaintiff by her three psychologists, all of whom testified at trial, can be summarized as follows. 1. PHILLIP M. MASSAD, Ph.D. (Examination in December 1989) a) test results Wechsler Adult Intelligence Scale — Revised (“WAIS”) Verbal IQ: 126 Performance IQ: 109 FuH Scale IQ: 122 Mean = 100, Standard Deviation = 15 Verbal Scale Information 15 Digit Span 10 Similarities 16 Arithmetic 11 Vocabulary 15 Comprehension 16 Performance Scale Picture Completion 15 Picture Arrangement 11 Block Design 11 Object Assembly 10 Digit Symbol 10 Mean = 10, Standard Deviation = 3 Woodcock Reading Mastery Test — Revised, Form H Subtest Percentile Rank (by age) Word Attack 67th Word Identification 52nd Word Comprehension 98th Passage Comprehension 97th Overall Reading Cluster 90th Reading Comp. Cluster 97th Basic Skills Cluster 64th Wide Range Achievement Test — Revised (“WRAT”) Subtest Percentile Rank (by age) Spelling 34th Arithmetic 63rd Rey Osterreith Complex Figures Test Subtest Percentile Rank (by age) Immediate Recall 35th Delayed Recall 65th b) clinical observations Dr. Massad administered four tests for which he reported no scores: the Detroit Tests of Learning Aptitude, a test of visual memory with which he reported plaintiff “had difficulty”; the Bender Gestalt Visual Motor Test about which he made no comment; the Minnesota Multiphasie Personality Inventory from which he found no evidence of undue test anxiety or other psychopathology; and the Gray Test of Oral Reading— Revised (the “Gray”). (Dr. Massad’s Psychoeducational Evaluation is at Pl.’s Ex. 20a.) Of particular significance is the Gray, because it is a test on which the subject reads aloud from a passage and reading speed is measured. It is untimed, meaning the test taker is under no time constraints to complete the reading assignment. Questions are then asked to assess reading comprehension. Dr. Massad testified that he had plaintiff read aloud from the test passage to “get a feel” for plaintiffs reading rate. (Tr. at 222.) He did not score the result, and made no mention of it in his six-page Psychoeducational Evaluation because it is not normed for adults and because he did not remember that its findings were “significant or germane to the diagnoses or what I was trying to determine.” (Id.) “I didn’t see anything remarkable to report.” (Tr. at 224.) Dr. Massad administered no other test that evaluated plaintiff’s reading speed, nor did he draw any conclusions with regard to whether plaintiff was a slow reader. (Tr. at 206, 208.) Based on his examination, Dr. Massad testified at trial that “it is my professional opinion that the plaintiff has learning disabilities characterized by difficulties with automaticity, phonological processing, organizing and processing visual-spacial information, short term memory and sequential processing and will require accommodations on the New York State Bar Examination.” (Massad Aff. ¶ 79.) Dr. Massad, however, did not discuss plaintiffs automaticity problems in the evaluation he submitted to the Board. (Tr. at 233.) He defines automaticity as the “ability to not have to deliberate when decoding a word.” (Id.). Dr. Massad agrees with the definition of learning disabilities contained in the Diagnostic and Statistical Manual III-R. Based on this definition, Dr. Massad believes that learning disabilities are characterized and identified by “intraindividual” variability in test performance scores. (Massad Aff. ¶ 42.) He views plaintiffs disabilities as reflected in the variability exhibited by plaintiffs subtest scores on the Passage and Reading Comprehension WAIS subtests as compared to her Word Attack and Word Identification scores, and as between her verbal IQ and her spelling score on the WRAT. He further finds that plaintiff “had difficulty organizing and processing visual-spatial information” as evidenced by her score on the Block Design and Object Assembly subtests, and her Rey-Osterreith score. Dr. Massad also concludes that plaintiffs reading skills were below what would be expected of a subject with plaintiffs record of academic achievement and intelligence. (Massad Aff. ¶ 81.) 2. RICHARD F. HEATH, Ph.D. (Examination in May 1993) a) test results Woodcock Reading Mastery Test — Revised (The “Woodcock”) Form G % ile (age) Form H % ile (age) G&H % ile (age) G&H % ile (grade) * Visual-Auditory Learning 46 Letter Identification 18 Word Identification 37 52 45 35 Word Attack 28 50 37 26 Word Comprehension 76 90 84 73 Passage Comprehension 74 99 88 90 Readiness Cluster 28 Basic Skills Cluster 36 53 43 25 Reading Comp. Cluster 78 98 89 85 Total Reading Cluster 48 84 66 * The. Woodcock is normed up to grade 16.9, ie., college graduates. Plaintiffs percentile rank thus represents the proportion of college graduates in a demographically representative sample who scored below her on the test. b) clinical observations Dr. Heath, a clinician with an extensive background in diagnosing learning disabilities in adults, did not purport to diagnose plaintiff in his evaluation to the Board, but only to confirm Dr. Massad’s diagnosis and supply plaintiff with the Woodcock scores requested by the Board. (Pl.’s Ex. 16, Heath’s Psychoeducational Evaluation; Heath Aff. ¶ 52; Tr. at 505-06.) In his evaluation, however, Dr. Heath described plaintiff as a “dyslexic adult” in his evaluation. (Pl.’s Ex. 16, Heath’s Psychoeducational Evaluation, at 2.) Moreover, he noted in his evaluation that “Dr. Bartlett decoded words slowly and without automaticity; self-corrections were common.” (Id. at 1.) Further, in describing her reading tests, he noted that “[s]he read [ ] passages slowly, and she typically read the more complex passages two or three times in order to ascertain their meaning.” (Id. at 2.) In his trial affidavit, Dr. Heath described his observations more fully: 58. In administering the Woodcock to plaintiff, I observed several things which are relevant to and supportive of my opinion that the plaintiff has learning disability. First, I noticed that she had to make several attempts to sound out words which should have been second nature to her. She [sic] reading was full of hesitations, and self corrections. In other words, plaintiff will attempt to read a word such as “instigator” as “investigator.” Since she will hear that it sounds incorrect she will start over and often corrects her reading of the word after several attempts. On the Woodcock, this would be credited as a correct response, even though it took her three attempts to get it right and took more time than it would have taken a person who did not have to read in this fashion. 59. Second, I observed that she needs to use her finger to keep her place when reading a paragraph in the passage comprehension subtest. The paragraphs on this subtest are only three to five lines long and yet plaintiff has difficulty keeping her place when reading. 60. Third, I observed that plaintiff reads aloud in a hesitant manner, slowly and without automaticity. Automaticity is the phenomenon by which a person recognizes a printed word and is able to read it accurately, and immediately; in other words, automatically and without thinking. In particular, plaintiff had a great deal of difficulty reading polysyllabic words, vowels (especially diphthongs, digraphs and in ascertaining differences between long and short vowels), consonant blends and silent consonant conventions. 61. Fourth, I also observed that on the more complex reading passages, Dr. Bartlett typically read the passages over two or three times before she could respond to that test item. She uses contextual cues to facilitate her decoding. She reads very slowly. She will reread a phrase or sentence to make sure she gets it. You can often see her lips move or hear her read quietly to herself and when she does this, you can hear the mispronunciations. When she is faced with an unfamiliar polysyllabic word she is very slow to break down the word to different parts and she will mispronounce parts of the word. She is slow to synthesize the morphemes into a word. In his trial affidavit, Dr. Heath also opined that the results of the Woodcock test he administered were consistent with Dr. Mas-sad’s diagnosis: As I mentioned earlier, I observed the plaintiff needed to sound out words slowly and with repeated attempts. This pattern of word attack is indicative of someone whose decoding skills are not fully formed. Word attack skills are generally well formed by junior high school age. Plaintiffs scaled are on the Word Attack subtest form G was 91, 28th percentile with a grade equivalent of 4.7. Thus, in laymen’s terms plaintiff decodes pseudo-words at a fourth grade level. This is a strikingly different performance from what one would expect of a person whose Passage Comprehension score on the same form of the test (G) was 110 or the 74th percentile. (Heath Aff. ¶ 62.) Dr. Heath further described plaintiff as suffering from a “mild to moderate” reading disability. (Tr. at 507.) Dr. Heath utilizes the same diagnostic approach as Dr. Massad, viewing a learning disability as “intraindividual or intrinsic to the nature of the individual.” (Heath Aff. ¶ 47.) Dr. Heath uses a history, neuropsychological battery, intelligence tests and achievement tests to diagnose learning disability. He looks for variation between the verbal and performance IQ scores on the WAIS, discrepancies between timed and untimed subtests, and errors in subtests. (Heath Aff. ¶ 40.) In the evaluation he submitted to the Board, Dr. Heath identified the difference between plaintiffs Basic Skills Cluster score and Reading Comprehension Cluster score as consistent with dyslexia: [Bartlett’s] pattern of decoding errors, as well as the significant discrepancy between her basic reading skills (43rd percentile) and reading comprehension (89th percentile), are consistent with a language-based learning disability. (Pl.’s Ex. 16, Heath’s Psychoeducational Evaluation, at 2.) At trial, he also maintained that any discrepancy between IQ and achievement scores over 1.5 standard deviations was strong evidence of a learning disability. (Heath Aff. ¶ 71.) 3. ROSA A. HAGIN, Ph.D. (Examination in September 1994) a) test results Woodcock Reading Mastery Test — Revised Form G % ile (age) Form H % ile (age) Word Attack 50 50 Diagnostic Reading Test (“DBT”) * Form C (timed) Form A (untimed) Comprehension 50th % ile 98th % ile Speed 4th % ile (195 wpm) > 1st % ile (156 wpm) * The DRT is not age normed. The highest grade norm is college freshmen, and thus plaintiff’s score is ranked against that group. Wide Range Achievement Test % ile (age) Oral Reading 86 Spelling 45 Neuropsychological Battery Bender Gestalt Visual Motor Test: 2/9 figures recalled Phoenician Spelling Test (spelling nonsense words): 19/20 correct Trailmaking Test: Speed abnormally low, poor visual scanning Purdue Pegboard: Normal speed, but poor laterality Extension Test: Equivocal laterality Finger Gnosis: 4/7 errors, most on left; poor laterality Directionality 5/7 errors in directional orientation b) clinical observations Dr. Hagin has been among the nation’s leading researchers in the field of learning disability for more than two decades and is the author of many books and articles on the subject. She holds faculty appointments at Fordham University and in the Department of Psychiatry at New York University Medical Center, and supervises clinics at both institutions. She examined plaintiff in preparation for trial and served as her lead expert witness. Dr. Hagin opined, based on Drs. Massad’s and Heath’s evaluations and her own, that plaintiff “has a learning disability consistent with the National Joint Committee on Learning Disabilities definition.” (Pl.’s Ex. 93, Hagin’s Psychological Evaluation, at 3; see definition supra, note 7.) Dr. Hagin placed considerable emphasis on the DRT results, which she viewed as demonstrating plaintiffs slow rate of reading. She also based her opinion on: (a) the 17-point discrepancy between plaintiffs WAIS verbal and performance scores, which Dr. Hagin asserted occurs in less than 20% of the population; (b) neuropsychological findings suggesting “central nervous system dysfunction”; (e) variations of standard deviations or more among WRMT subtest scores and WAIS subtest scores; (d) an 18-point discrepancy between the verbal IQ score and the Word Attack scores, which Dr. Hagin asserted occurs in 5% of the population; (e) plaintiffs performance on a 53-word writing sample, which was “laborious” and contained five spelling errors, and (f) achievement test scores that, overall, contrast with plaintiffs superior cognitive abilities and academic achievement. (Pl.’s Ex. 93, Hagin’s Psychological Evaluation at 3.) Of central importance in her diagnosis is Dr. Hagin’s view that plaintiff has evolved a set of personal skills to compensate for her disability: She used several kinds of cues to assist her in responding to the tasks presented: slowing down the rate of response, verbal rehearsal of rote sequencing items, pointing cues to assist in keeping her place on visual text. (Pl.’s Ex. 93, at 2.) Dr. Hagin believed that plaintiffs earlier work as a school teacher where phonics were stressed allowed plaintiff to develop “self-accommodations” that account for her ability to spell better and to perform better on word identity and word attack tests than would be expected of a reading disabled person. According to Dr. Hagin, a learning disability’s diagnosis cannot be made “on the basis of any one score or any one test. It made [sic] based on a total picture.” (Hagin Aff. ¶ 110.). For this reason, she prefers the NJCLD definition of learning disability. Although many of plaintiffs achievement scores fell in the average range when compared with plaintiffs age group, Dr. Hagin’s judgment is that “one’s educational level and expectancy” and clinical judgment should be dispositive in identifying a learning disability rather than test scores based on age norms. “Clearly, graduation from law school denotes a high level of achievement and correlated expectancies.” (Hagin Aff. ¶ 123.) Dr. Hagin believes that Dr. Vellutino’s definition of dyslexia as solely a phonological decoding problem is too narrow. (Tr. at 698.) She views the reading task as more complex than simply identifying words. The reading process also requires understanding what text means. (Tr. at 695-96.) To Dr. Hagin, because the Woodcock tests relied upon by Dr. Yellutino do not test automaticity or reading rate, they are poor indicators of a decoding problem in individuals like plaintiff who function at higher cognitive levels. (Tr. at 699-703.) Dr. Hagin notes that the DRT is a “very easy test” — comparable to reading a passage in Reader’s Digest. Dr. Hagin expects a college-educated person to read DRT passages at the rate of 300 words-per-minute. Instead, plaintiff read at. 195 words-per-minute timed — the fourth percentile compared to college freshman, and 156 words-per-minute untimed — below the first percentile compared to college freshman. (Tr. at 435, 701, 1050-51, 1092.) According to Dr. Hagin, plaintiff should have been performing at the 50th percentile of college freshmen, and instead reads very slowly when compared to a college student. (Tr. at 1050-51.) For Dr. Hagin, the issue is not whether plaintiff can comprehend as she reads but the difficulty plaintiff has in the process of comprehending what she reads. (Tr. at 1076, 1632.) Dr. Hagin concludes that plaintiff does not read in the same condition, manner or duration of the average adult reader in that plaintiff does not read with the automaticity or speed of an average reader. (Tr. at 2494-98, 2545.) 4. Plaintiffs In-Court Demonstration As part of plaintiffs proof, Dr. Hagin administered an in-court demonstration of plaintiffs reading and writing ability. Plaintiff was asked to read a passage describing a criminal law hypothetical from a 1988 bar exam, a document selected at random by the Court from among the exhibits, consisting of 426 words. Plaintiff read haltingly and laboriously, whispering and sounding out some words more than once under her breath before she spoke them aloud. Plaintiff marked the right-hand side of the text with her right index finger, advancing it down the right margin and using her left hand to read across the line. (Tr. at 748.) She made one word identification error, reading the word “indicted” as “indicated.” Plaintiffs reading speed was approximately 40 words per minute. Plaintiff was also asked to write a 48-word passage as it was dictated to her. The specimen produced, Plaintiffs Exhibit 174, has six grammar and spelling errors (“families” for “family’s,” “prapar” for “prepare,” “Dave” for “David,” “brotha” for “brother,” “inaddvertently” for “inadvertently” and “omited” for “omitted”), and three words crossed out at the right margin which appear to have been written backwards or in “mirror writing.” It took plaintiff approximately 10 minutes to complete the task. Another specimen of plaintiffs handwritten work product admitted into evidence was her essay answers on the February 1993 New York Bar Examination, consisting of 38 single-spaced pages. (Pl.’s Ex. 185; Def.’s Ex. B.) Omitting what are commonly understood shorthand condensations of words (e.g., “managemt” for “management,”), I count 10 spelling errors. There are no examples of mirror writing, and the handwriting is generally legible. Plaintiff completed all six of the essay questions. The Court recognizes that the trial setting undoubtedly affected plaintiff’s performance in the courtroom demonstration. Therefore, the Court places limited value on the demonstration. The Court instead relies upon Dr. Hagin’s and Dr. Heath’s testimony of what they saw during their evaluation of the plaintiff and uses the demonstration only as illustrative of some of the phenomena Dr. Hagin and Dr. Heath described during their testimony. B. Defendants ’ Expert Opinions 1. FRANK R. VELLUTINO, Ph.D. At all relevant times, the Board employed a research psychologist expert in the field of learning disabilities, Dr. Frank R. Vellutino, to advise it on policies relating to the identification and accommodation of learning disabled applicants, and to screen applications for accommodations. Dr. Vellutino is a leading researcher in the field of learning disabilities and has published numerous books and articles on the subject. His primary experience is with children. He is a Professor in the departments of Linguistics, Psychology and Educational Psychology and Statistics at the State University of New York at Albany. He also supervises a clinic engaged in the identification and treatment of children with dyslexia. In its rules and regulations, the Board does not specify what tests, if any, applicants for accommodations should submit with then-applications. Dr. Vellutino prefers to receive scores from each of the Woodcock Reading Mastery Test — Revised Word Attack and Word Identification subtests in evaluating applicants claiming a reading disability. The Board advises applicants of Dr. Vellutino’s preferences if they call or write asking which test results they should submit. Even if an application does not provide results from the Woodcock test, Dr. Vellutino will examine the results from whatever tests are submitted and evaluate whether those test results contain a word identifieation/word attack component sufficient to support the clinician’s conclusions. The Woodcock Word Attack test requires a subject to sound out nonsense words and is thus a test of a person’s ability correctly to associate letter combinations with then-sounds, a task referred to as phonological decoding ability. A subject is presented with 45 separate words, beginning with simple one-syllable patterns {e.g., “ip” and “din”) and progressing to more complex combinations {e.g., “ceisminadolt” and “gnouthe”). The Woodcock Word Identification test requires a subject to identify 106 real words in isolation that range from simple (“is”) to difficult (“zymolysis”). Both tests are untimed, and the scores do not reflect incorrect tries that precede a correct answer. Dr. Vellutino discounts the significance of discrepancies in test scores as an identifier or discriminator for learning disabilities in an individual because even superior readers have discrepancies in scores. (Tr. at 1787-88 (noting that IQ/Achievement discrepancies are present in both good and poor readers); Vellutino Aff. ¶ 14 (stating that discrepancy reported in Dr. Bartlett’s scores is contrary to that found in reading disabled applicants because she has higher score in verbal skills than performance skills).) Similarly, Dr. Vellutino claims that research studies demonstrate that problems with spelling do not define a learning disabled person because “there are many good readers who are also poor spellers.” (Vellutino Aff. ¶ 14.) Neither do visual spatial organization problems, directional confusion or the like identify a reading disabled individual for Dr. Vellutino. (Tr. at 1173 (reporting that “in every piece of research we’ve done ... we get no differences between poor and normal readers” in performing these tasks); id. at 1200 (stating he does not believe that Dr. Hagin’s tests are “important diagnostic signs”).) Dr. Hagin concurs with Dr. Vellutino that a discrepancy in scores or difficulty in other visual or spatial functions do not identify a learning disabled individual, but she believes that the discrepancies and task malfunctions can signal the existence of a disability. (Hagin Aff. ¶¶ 116, 126.) In Dr. Bartlett’s case, as discussed, Dr. Hagin notes that plaintiff performed poorly on directional task tests; further, Dr. Hagin clinically observed the effect of such confusion upon plaintiff’s reading in plaintiffs use of her finger to track left to right reading, and in plaintiffs frequent skipping of a line when returning to the right side of the page. (Tr. at 748-19.) According to Dr. Vellutino, directional reading confusion exists in both learning disabled and nonlearning disabled children, and many adults retain vestiges of childhood coping mechanisms for reading difficulties. (Tr. at 1849-50.) Because the signals relied upon by Dr. Hagin are not, according to him, discriminators of learning disability, Dr. Vellutino believes that a diagnosis of dyslexia can only “be based exclusively on measures of reading ability, in particular measures of Word Identification and phonetic decoding or word analysis skills (ability to ‘sound out’ a word), deficiencies in which are characteristic of individuals with severe reading disability.” (Vellutino Aff. ¶ 33(a).) The only tests available which measure these functions are the Woodcock Reading Mastery Test — Revised or the Woodcock-Johnson Psychoeducational Test Battery. (Id. ¶ 33(e).) Dr. Vellutino prefers the Woodcock Reading Mastery Test over the Woodcock Johnson because the Mastery Test is more comprehensive. (Id.) Moreover, Dr. Vellutino believes the Word Attack subtest, is the most “direct measure of phonological dyslexia.” (Tr. at 1804.) Based on his view that a reading disability must affect an applicant’s ability to perform on reading function tests like word attack and word identification, Dr. Vellutino recommended to the Board that it automatically grant accommodations to applicants claiming reading disabilities if their scores on the Woodcock Word Attack and Word Identification tests are below the 30th percentile when age-normed or grade-normed. (Vellutino Aff. 1133(e); Tr. at 2058 (defining “significantly impaired in reading” as “deficiency in reading subskills”).) Dr. Vellutino recommended the 30th percentile cutoff on the basis of studies showing that the incidence of learning disability in the population is estimated at between 5% and 20% and thus, a 30% cutoff, according to him, would be reasonably certain to capture all disabled applicants. (Vellutino Aff. ¶ 32; see also Tr. at 1305-06 (describing his choice of a 30th percentile cut-off as arbitrary, but not irrational because the cut-off is overinclusive).) Dr. Vellutino admits that scores on the Woodcock, and hence his cut-off, do not distinguish reading disabled applicants who read slowly from purely slow readers. (Tr. at 2401-02.) Dr. Vellutino, however, will give applicants the benefit of doubt and has recommended accommodations for an applicant if either their Word Attack or Word Identification score is below 30% or 1 or 2 percentage points above, or other subset reading scores show marked deviations from the average range or are marginal. (See, e.g., Pl.’s Ex. 123-A68; Tr. at 2080-84, 2089, 2094, and 2123.) Plaintiff scored in the 29th percentile on the Word attack test given by Dr. Heath but Dr. Vellutino did not give her the benefit of the doubt or recommend accommodations for her because he considered, that one score an anomaly among other test scores that demonstrated above average, if not superior, reading functions. Moreover, he viewed that score as within an average range. (Tr. at 1303-05, 2118-19, 2167.) Dr. Vellutino also discounted Dr. Hagin’s characterization of plaintiff as a slow reader because he viewed plaintiffs performance at the rate of 195 words-per-minute on the DRT test as within average range. (Vellutino Aff. ¶ 62; Tr. at 1212-15.) In reaching this conclusion, Dr. Vellutino relied upon various studies of reading rates and extrapolated from them that plaintiffs DRT was within the normal range, despite the 4% untimed college norm and 1% timed college norm of the test. (Tr. at 1822.) Finally, Dr. Vellutino assumes that anyone who can score above the 30th percentile in Forms G and H of the Woodcock has sufficient automaticity to read most texts. (Tr. at 2405.) In short, Dr. Vellutino recommended against giving Dr. Bartlett accommodations because he has “rarely” seen clinical findings of a significant disability with such high test scores. (Tr. at 1314; but see Pl.’s Ex. 123-18; Tr. at 2161 (applicant with scores much like plaintiffs who Dr. Vellutino recommended for accommodation).) 2. DR. JACK M. FLETCHER Dr. Fletcher holds professorships at the University of Texas Medical School at Houston and the University of Houston. He is a psychologist and holds a diplómate in neuropsychology. He has published widely on dyslexia and neuropsychology, and devotes half of his time to clinical practice, principally, but not exclusively, with children. Earlier in his career, Dr. Fletcher wrote articles criticizing Dr. Vellutino’s approach to the diagnosis of learning disabilities. (Fletcher Aff. ¶9.) However, based on his own research and that of others, Dr. Fletcher has concluded that Vellutino’s approach is the only valid approach for identifying a learning disability. (Id. (“Over the years, Dr. Vellutino’s original hypotheses concerning the cognitive basis of reading disability have been shown to be correct. His early hypotheses presaged the now widely accepted understanding that reading disabilities have a linguistic basis and specifically reflect fundamental problems with the development of word decoding abilities that, in turn, reflect deficiencies in the acquisition of phonological awareness skills.”).) After examining plaintiff’s evaluation reports, he concluded that plaintiff was neither impaired nor disabled. (Fletcher Aff. ¶ 11.) He concurred fully with Dr. Vellutino’s evaluation of plaintiffs application for accommodations. (Fletcher Aff. ¶ 47.) C. Plaintiffs evidence of disability. 1. Psychometric Testing. The experts in this case disagreed on much, but none challenged the efficacy of psychometric testing per se. Plaintiffs experts use the same cluster of achievement tests as defendants’ experts to assess the presence of a reading disability. These tests have been standard in the psychology discipline for decades. The tests have gained acceptance in the field in part because statistical measures of their reliability are positive. Plaintiffs experts mention, as a general proposition, that test scores alone can not reliably identify reading disabled individuals, and they criticize Dr. Vellutino’s reliance on the Woodcock for identifying adults with a reading disability. I agree with plaintiffs experts. Plaintiffs experts have persuaded me that plaintiffs reading disability cannot be measured solely by psychometric testing. For example, no test measures automatieity directly. (Tr. 489, 503, 702.) A lack of automaticity in understanding words without undue attention to them is usually inferred from a combination of test scores and clinical observations. (Tr. at 701-02.) In this case, all three of plaintiffs experts noted plaintiffs stark lack of automatieity when she was required to read aloud. On the Woodcock tests themselves, plaintiff had to sound out the words repeatedly before coming to an answer. Plaintiffs lack of automatieity is further confirmed by her slow rate of reading compared to college freshmen on the DRT test. In that test, plaintiffs timed reading rate of 195 wpm compared to the 4% percentile of college freshmen. Finally, plaintiffs reading test data was not consistent across a wide range of reading-related skills. As noted by Dr. Heath and Dr. Hagin, plaintiffs high comprehension scores were incongruent with her relatively lower Word Attack and Word Identification scores. I find seriously infirm Drs. Vellutino and Fletcher’s presumption (albeit according to them rebuttable presumption) that a score above the 30th percentile on the Woodcock Word Attack and Word Identification sub-tests in all eases identifies the absence of a reading disability. As admitted by Dr. Vellutino, such a screening mechanism suffers from serious problems where an applicant’s other scores and clinical reports place him or her at or below the average on other reading skill indicators. (See, e.g., Tr. at 2107). Further, the best evidence for the Woodcock’s shortcomings comes from defendants’ experts and the scientific evidence upon which they rely. To support their testimony, both Drs. Vellutino and Fletcher relied principally on the studies of adult dyslexies conducted by Dr. Maggie Brack. (Tr. at 280, 1780.) Yet, Brack found the Woodcock subtests poor discriminators for a learning disability unless the subject’s reaction time was measured. (Defs.’ Ex. JJ at 444; Pl.’s Ex. 149 at 262 (“[I]t is the slowness of reading that is particularly characteristic of the deficient word recognition skills of adult Dyslexics”).) The Woodcock is an untimed measure of phonological decoding ability and does not score for reaction time. Further, both Dr. Fletcher and Dr. Vellutino do not credit clinical reports of lack of automaticity. Yet, Dr. Vellutino did acknowledge the Woodcock’s weakness with regard to discriminating for lack of automaticity. (Tr. at 2305.) A second criticism of the tests is that they are designed principally to test children and thus do not have enough items in the difficult range. Dr. Vellutino, in a recent research article, acknowledged the Woodcock has “severe limitations,” in that “there are far too few items at any given level to be certain of reliability at that level.” (Pl.’s Ex. 89 at 304 Vellutino, Scanlon & Tanzman, Components of Reading Ability (1994).) Further, although Form G and H are supposed to be equivalent tests for norms, Dr. Vellutino admits that in his clinical experience Form G is harder than Form H. (Tr. at 1955.) Dr. Fletcher does not use these tests in his research. Moreover, although Dr. Vellutino claims the 30th percentile cutoff is “over inclusive,” Brack reported that, using a test similar to the Woodcock, “one third of the subjects [adult Dyslexics] scored above the 30th percentile on the WRAT-R Level II.” (Defs.’ Ex. KK at 877.) In an earlier study of college-student dyslexics, the average score was at the 32nd percentile, with the range being from the 3rd to 81st. (Defs. Ex. JJ, Table 1, at 443.) Thus, despite Dr. Vellutino’s insistence that the 30th percentile cut-off is over generous in identifying reading disabled applicants, the studies he relies upon provide testing data that show reading disabled college students performing well above the 30th percentile. Brack reports that using a test similar to the Woodcock, “one third of the [adult dyslexics] subjects scored above the 30th percentile on the WRAT-R Level II.” (Def. Ex. KK at 877). In an earlier study of college-student dyslexics, the average score was at the 32nd percentile, with the range being from the 33rd to 81st percentiles. (Def. Ex. JJ, Table 1, at 443). Finally, I do not credit Dr. Vellutino’s attempt to equate Bartlett’s low DRT reading rate score with an average rate by extrapolation to other tests. This approach is seriously infirm in that it attempts to compare scores on different tests with different subject populations. As noted by Dr. Hagin, to be within the average range of college freshmen, plaintiff should have been performing at the 50th percentile of the DRT, and instead she reads at a very slow rate for the college student population which this test directly measures. (Tr. at 1050-51.) In short, I do not accept Dr. Vellutino and Dr. Fletcher’s conclusions that reading disabled individuals are incapable of having the test scores reflected by plaintiff. Plaintiffs experts have convinced me that a reading disability is not quantifiable merely in test scores. A learning disability is not measurable in the same way a blood disease can be measured in a serum test. By its very nature, diagnosing a learning disability requires clinical judgment. Clinicians need to examine a patient to ensure that low or disparate scores are not the result of low intelligence, or emotional or other social problems. Moreover, I accept the opinion of plaintiffs experts, based on the studies of Dr. Maggie Brack, that tests like the Woodcock are “poor discriminators” for adults. (Defs.’ Exh. JJ at 444.) Thus, as much as the Board would like to find an easy test discriminator for a reading disability in its applicants, such a test does not exist. Finally, I also do not accept the position of defendants’ experts that clinical judgments of a lack of automatieity must be rejected as subjective. Clearly, plaintiffs low, albeit within the average range, test scores on the Woodcock, combined with clinical observations of her manner of reading amply support a conclusion that she has an automaticity and a reading rate problem. 2. Discrepancy versus performance measures Central to this case has been the contention by plaintiffs experts that reading disability can be identified by