PUBLIC COUNSEL MARK ROSENBAUM (State Bar No. 59940) mrosenbaum@publiccounselorg ALISA HARTZ (State Bar No. 285141) altartz@pub1iccounsel.org AMANDA SAVAGE (State Bar No. 325996) asavage@publiccounse1.org 610 S. Ardmore Avenue, Los Angeles, CA 90005 Tel: (213) 385-2977 Fax: (213) 385-9089 SCHEPER KIM HARRIS LLP GREGORY A. ELLIS (State Bar No. 204478) gellis@scheperkim.com KATHERINE FARKAS (State Bar No. 234924) klarkas@scheperkim.com MICHAEL LAVETTER (State Bar No. 224423) m1avetter@scheperkim.com 800 West Sixth Street, 18th Floor Los Angeles, California 90017-2701 Tel: (213) 613-4655 Fax: (213)613-4656 Additional Counsef listed on next page Attorneys for Plaintiffs w, - (331-11.; I. ALAM Ft .4. OT PATTY DEC 10 2019 .1 OI: THE. SUPERIOR COURT .- .1 w. 11111 SUPERIOR COURT OF THE STATE OF CALIFORNIA COUNTY OF ALAMEDA KAWIKA SMITH. through his guardian ad litem LEILANI GLORIA D, through her guardian ad litem DIANA STEPHEN C., through his guardian ad Iitem, MARGARET ALEXANDRA VILLEGAS, an individual; CHINESE FOR AFFIRMATIVE ACTION, a nonpro?t organization; COLLEGE ACCESS PLAN, a nonpro?t organization; COLLEGE SEEKERS, a nonpro?t organization; COMMUNITY COALITION, a nonpro?t organization; DOLORES HUERTA FOUNDATION, a nonpro?t organization; and LITTLE MANILA RISING, a nonprofit organization, Plaintiffs, v. REGENTS OF THE UNIVERSITY OF JANET NAPOLITANO, in her of?cial capacity as President ofthe University of California; and DOES 1-100, Defendants. PLAINTIF COMPLAINT AND DEMAND FOR JURY TRIAL CASENO. RGI9046222 UNLIMITED JURISDICTION COMPLAINT FOR DECLARATORY AND INJUNCTIVE RELIEF 1. VIOLATION OF STATE EQUAL PROTECTION GUARANTEES (Cal. Const., art. I, 7(a) art. 1V, 2. DISCRIMINATION ON THE BASIS OF A PROTECTED CLASSIFICATION (Cal. Gov. Code 11135); 3. PROHIBITED DISCRIMINATION IN EDUCATION (Cal. Educ. Code 66270); 4. VIOLATION OF THE UNRUH ACT (Cal. Civ. Code 51); 5. VIOLATION OF THE CALIFORNIA DISABLED PERSONS ACT (Cal. Civ. Code 54 et seq.) 6. DECLARATORY RELIEF TRIAL 1 EQUAL JUSTICE SOCIETY Lisa Holder (State Bar No. 212628) 2 lisaholder@yahoo.com 634 S Spring Street, Suite 716, Los Angeles, CA 90014 3 Telephone: (213) 488-0868 4 MILLER ADVOCACY GROUP 5 MARCI LERNER MILLER (State Bar No. 162790) marci@milleradvocacy.com 6 1303 Avocado Ave, Suite 230, Newport Beach, CA 92660-7804 Telephone: (949) 706-9734 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 2 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 INTRODUCTION 1. 2 Public higher education is the gateway for all California students to build 3 successful careers, pursue their intellectual interests, and gain meaningful access to social 4 mobility. The California Constitution established the University of California (“UC” or the 5 “University”) as a public trust and envisioned the UC as the culmination of the State of California 6 (“State”)’s obligation to provide public education to its citizens. 1 As the State’s preeminent public 7 institution of higher education, the University of California has a legal obligation to provide equal 8 access to all qualified students, regardless of their backgrounds. 2. 9 But the UC has failed to carry out its promise: rather than fulfilling its vision as 10 “[a]n engine of opportunity for all Californians”2 and creating a level playing field in which all 11 students are evaluated based on individual merit, the UC requires all applicants to subject 12 themselves to SAT and ACT tests that are demonstrably discriminatory against the State’s least 13 privileged students, the very students who would most benefit from higher education. The 14 discriminatory effect of the use of these tests is well known to both the State and the UC. 15 Governor and ex officio Regent Gavin Newsom recently recognized that “use [of the SAT and 16 ACT] exacerbates the inequities for underrepresented students, given that performance on these 17 tests is highly correlated with race and parental income, and is not the best predictor for college 18 success.”3 UC Berkeley Chancellor Carol Christ similarly called for the UC to drop its test score 19 requirement because SAT and ACT scores “really contribute to the inequities of our system.” 4 20 1 Cal. Const. art. IX, § 9(a); see id. § 1. The State of California is the legal and political entity responsible for ensuring that all California public school students receive their individual and 22 fundamental right to an equal education, under the Equal Protection Clause of the California Constitution, Article I, section 7(a). 23 2 Univ. of Cal. Office of the Pres., Your University of California: A Report from President Mark 24 Yudof for Friends of UC (Sept. 2009), available at regents.universityofcalifornia.edu/regmeet/sept09/pres.pdf. 25 3 Governor Gavin Newsom, AB 751 Veto Message (Oct. 12, 2019), available at 26 https://www.gov.ca.gov/wp-content/uploads/2019/10/AB-751-Veto-Message.pdf. 21 4 Teresa Watanabe, Drop the SAT and ACT as a Requirement for Admission, Top UC Officials Say, L.A. Times (Nov. 23, 2019), https://www.latimes.com/california/story/2019-11-23/uc28 officials-recommend-dropping-sat-admission-requirement. 27 3 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 These comments are remarkably consistent with UC’s original 1962 decision not to use the SAT 2 because the “scores add little or nothing to the precision with which existing admissions 3 requirements are predictive of success in the University.” 5 They are also consistent with the results 4 of decades of research within the UC itself about the discriminatory effect and minimal predictive 5 value of these tests. These discriminatory tests irreparably taint UC’s ostensibly “holistic” 6 admissions process. The mere presence of the discriminatory metric of SAT and ACT scores in 7 the UC admissions process precludes admissions officers from according proper weight to 8 meaningful criteria, such as academic achievement and personal qualities, and requires them 9 instead to consider criteria that act as a proxy for wealth and race and thus concentrate privilege on 10 UC campuses. 3. 11 Consequently, the UC admissions process—as deliberately operated by the 12 Regents—creates formidable barriers to access to public higher education for deserving students 13 from low-income families, students from historically underrepresented racial and ethnic groups, 14 and students with disabilities. The requirement that all applicants submit SAT or ACT scores 15 systematically and unlawfully denies talented and qualified students with less accumulated 16 advantage a fair opportunity to pursue higher education at the UC. Every UC admissions cycle 17 that evaluates applicants based on their SAT and ACT scores irreparably damages the futures of 18 tens of thousands of students who are capable of excelling at the UC campuses of their choice and 19 benefitting from the opportunities and supports a UC education provides, causing unjustifiable 20 squandering of time and resources and intense stress for them, their families, and their schools. 4. 21 The UC is well aware that, rather than providing meaningful information about a 22 student’s ability to succeed in college, SAT and ACT scores are largely a proxy for a student’s 23 socioeconomic background and race. Instead of measuring students’ academic ability or mastery 24 of curriculum, SAT and ACT scores reflect demographic and socioeconomic characteristics like 25 family income, parental education, and race. With its mission of providing public higher education 26 to all Californians, the UC would never explicitly decline to admit a student because the student is 27 28 5 John A. Douglass, The Conditions for Admission: Access, Equity, and the Social Contract of Public Universities 90 (2007) (quoting BOARS Chairman Charles Jones). 4 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 low-income, a first-generation college applicant, or a member of an underrepresented minority 2 group. But that is precisely the result of UC’s decision to consider SAT and ACT scores: an 3 admissions process that rations access to public higher education on the bases of race, privilege, 4 and wealth. 5. 5 Over the past six decades, the UC has commissioned study after study on what, if at 6 all, SAT and ACT scores contribute to the prediction of first-year grades, and repeatedly arrived at 7 the same answer: almost nothing. Taken alone, SAT and ACT scores predict less than two percent 8 of the variation in first-year grade point average (“GPA”) among UC students. 6 First-year grades 9 are a dubious metric around which to build a student body: nowhere in any of the voluminous 10 publications and presentations about the UC system is it ever stated that maximizing first-year 11 grades is either a goal of the system or meaningful in any relevant way to the objectives of a 12 college education. Putting aside the questionable choice to structure admissions decisions around 13 the metric of first-year grades, UC’s findings have remained strikingly consistent over time: SAT 14 and ACT scores add very little incremental validity to predictions of first-year grades, and that 15 already marginal contribution dwindles even further once socioeconomic characteristics are taken 16 into account. UC researchers have demonstrated that high school grades are consistently the best 17 predictor of college success—a finding that has also been established by hundreds of studies at 18 other colleges and universities. Faced with such studies, the UC has not only admitted the 19 “considerable redundancy” of using both high school GPA and SAT or ACT scores as admissions 20 criteria, but has also recognized that high school GPA “has less adverse impact on disadvantaged 21 groups.”7 Moreover, as the UC has acknowledged, but validity studies ignore, “students can be 22 coached, to advantage,” on the SAT and ACT. 8 The fact that affluent students can effectively 23 6 Univ. of Cal. Bd. of Admissions & Rel. with Schs. (BOARS), Revised Proposal to Reform UC’s Freshman Eligibility Policy app. IV 95 (Feb. 20, 2008), available at 25 https://senate.universityofcalifornia.edu/_files/ underreview/sw.rev.eligibility.02.08.pdf. 26 7 BOARS, Admissions Tests and UC Principles for Admissions Testing 15, 24 (Dec. 2009), 27 available at https:// senate.universityofcalifornia.edu/_files/reports/hp2mgy_boars-testing_010609.pdf. 28 8 Id. at 21. 24 5 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 purchase higher scores through expensive private tutoring services diminishes the already limited 2 predictive value of the tests still further. 6. 3 The design of the SAT and ACT places underrepresented minority students at a 4 significant disadvantage. UC Provost Michael Brown recently objected to UC’s continued use of 5 the test on precisely these grounds, noting that rather than measuring individual merit, SAT and 6 ACT results artificially “compare students against one another in a way designed to produce high 7 and low scores.”9 To repeatedly produce this score distribution, the test development process tends 8 iteratively to discard items on which underrepresented minority students perform well and to 9 retain questions on which they do not do well. Over multiple administrations of the test, this 10 adverse effect reinforces and perpetuates itself, such that subsequent tests are systematically 11 biased against underrepresented minority students. UC psychometricians have found that up to 12 12 percent of items are biased against Black students, and up to 10 percent of items are biased against 13 Latinx students.10 7. 14 The highly discriminatory nature of the SAT and ACT has resulted in starkly 15 disparate student outcomes. According to the College Board’s 2019 data, among students taking 16 the SAT in California, 45 percent of White students scored 1200 or above, compared to only nine 17 percent of Black students and 12 percent of Latinx students.11 Only one percent of Black students 18 and two percent of Latinx students scored in the top score bracket, compared to 12 percent of 19 White students.12 And although Asian students have the highest scores when treated 20 monolithically by the College Board, such a grouping hides the fact that certain subgroups score 21 much lower than average. 22 23 24 9 Watanabe, supra note 4. 25 10 27 11 Maria Veronica Santelices & Mark Wilson, On the Relationship Between Differential Item Functioning and Item Difficulty: An Issue of Methods? Item Response Theory Approach to 26 Differential Item Functioning, 72 Educ. & Psychol. Measurement 5, 24 (2012). 28 College Board, SAT Suite of Assessments Annual Report: California (2019), available at https:// reports.collegeboard.org/pdf/2019-california-sat-suite-assessments-annual-report.pdf. 12 Id. 6 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 8. 1 Unsurprisingly, given their enormous stakes, the SAT and ACT have spawned a 2 lucrative test preparation industry which capitalizes on the desires of students and their parents to 3 maximize their odds of admission by securing even incremental score increases. Well aware of the 4 decisive effect SAT and ACT scores can have on college admission opportunities, affluent 5 students and their parents spend thousands of dollars to learn test-taking tricks and strategies from 6 private tutors. As UC’s Board of Admissions and Relations with Schools (“BOARS”) and 7 individual Regents have recognized, unequal access to high-quality test preparation gives affluent 8 students a competitive advantage over their peers with less wealth. According to UC Regents 9 Chairman John Pérez, “[t]he highest predictive value of an SAT isn’t in how well a student will do 10 in school, but how well they were able to avail themselves of prep material. And access to that 11 prep material is still disproportionately tied to family income.” 13 Moreover, by incentivizing 12 mastery of test-taking tricks over mastery of curriculum, test preparation “distracts participants 13 from college preparatory coursework.”14 The need to prepare students for high-stakes, one-off 14 tests has forced already underresourced schools and school districts to divert their limited funds 15 away from substantive education and toward test preparation. As then-UC President Richard 16 Atkinson stated when he recommended eliminating the use of the SAT I in UC admissions: 17 “[O]veremphasis on the SAT is compromising our education system.” 15 9. 18 The Regents have been aware of these inequities for years. Nevertheless, the 19 Regents have determined that the minimal added value of SAT and ACT scores in predicting first20 year GPA outweighs their harms to underrepresented minority students, students with disabilities, 21 and students with less wealth. 22 23 13 Teresa Watanabe, Q&A: Raise UC Tuition? Eliminate SAT Tests? Board of Regents Chairman John A. Pérez Has Something to Say, L.A. Times (Nov. 1, 2019), 25 https://www.latimes.com/california/story/2019-11-01/q-a-uc-board-of-regents-chairman-john-aperez. 26 14 BOARS, supra note 7, at 21. 27 15 Richard C. Atkinson, Robert H. Atwell Distinguished Lecture at the Eighty-third Annual 24 28 Meeting of the American Council on Education: Standardized Tests and Access to American Universities 139 (Feb. 18, 2001), available at https://escholarship.org/uc/item/6182126z. 7 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 10. 1 Plaintiffs are students and organizations that are committed to college access for 2 underrepresented minority students and students with disabilities. Student plaintiffs are four 3 talented and accomplished young people whose ability to access public higher education has been 4 hindered by UC’s SAT or ACT score requirement. Organizational plaintiffs represent diverse 5 communities and serve thousands of well-qualified students who have been adversely impacted by 6 UC’s unwillingness to cease its unlawful reliance on a discriminatory metric. 11. 7 UC’s Academic Senate is yet again studying the issue of whether to continue 8 requiring SAT or ACT scores, and speculates that it might have a recommendation by the summer 9 of 2020. As UC’s Provost and UC Berkeley’s Chancellor recently recognized, the research is 10 clear: “performance on the SAT and ACT [is] so strongly influenced by family income, parents’ 11 education and race that using them for high-stakes admissions decisions [is] simply wrong.” 16 But 12 UC’s continued use of a flatly discriminatory metric is not only bad educational policy; it is 13 unlawful discrimination. BOARS’s evaluation therefore ignores the most critical reason for ending 14 UC’s use of SAT and ACT scores in its admission process: its illegality. UC’s requirement that all 15 applicants submit SAT or ACT scores in order to be considered for admission violates the Equal 16 Protection Clause of the California Constitution and multiple State antidiscrimination statutes, and 17 is actively preventing Plaintiffs from accessing public higher education and its attendant 18 opportunities. 19 PARTIES 20 Plaintiffs 21 Individual Student Plaintiffs 22 12. Plaintiff Kawika Smith is a 17-year-old Black student who is a senior at Verbum 23 Dei High School. Kawika is an active student advocate and organizer in the South Los Angeles 24 community. He serves as the youth representative on his neighborhood council, engages in 25 community organizing with Plaintiff Community Coalition, and advocates for workers through the 26 long-term homecare union SEIU Local 2015. He also works on youth poverty issues with 27 28 16 Watanabe, supra note 4. 8 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 Children’s Defense Fund and acts as a youth ambassador for Imagine LA, a nonprofit organization 2 that works to end family homelessness and chronic poverty. Kawika has fought discriminatory 3 policies at his high school and helped to bring about state legislation banning discrimination based 4 on natural hair, Senate Bill 188 (effective January 1, 2020). Kawika has taken multiple honors 5 courses, and has a cumulative GPA of 3.56 and an unweighted GPA of 3.32. His PSAT score is 6 1140. 13. 7 Kawika aspires to a career in government, through which he can continue to engage 8 in public advocacy and activism. He plans to work in local or state government and eventually run 9 for office. Ultimately, Kawika hopes to represent his community as either a U.S. Representative or 10 a U.S. Senator. 14. 11 Throughout his years in school, Kawika has experienced significant trauma. He has 12 experienced housing insecurity for much of his life, and was homeless from second grade through 13 eighth grade. Kawika is a survivor of rape and has experienced domestic violence. As a child, he 14 witnessed a murder, and he more recently experienced the death of his brother during his junior 15 year of high school. These challenges have made it extraordinarily difficult for Kawika to perform 16 well on high-stakes tests like the SAT and ACT. His food insecurity impacted his test performance 17 because he was not able to eat breakfast. He also knows that his past trauma directly impacts his 18 capacity to perform on these tests. 15. 19 Kawika’s school partnered with an SAT program through which local college 20 students from California State University, Dominguez Hills helped students prepare for the SAT. 21 Only one of the tutors offered high-quality instruction, and she was unable to tutor all of the 22 students who sought her assistance. Kawika’s more affluent friends were able to afford private 23 SAT tutoring, with costs ranging from $2,500 to $5,000 a month. Although Kawika participated in 24 Khan Academy,17 that program did not help him improve his scores on the PSAT. On simulated 25 SAT tests, Kawika’s score dropped from his 1140 PSAT score to 980. During these tests, Kawika 26 27 17 Under a contract with the College Board, Khan Academy provides free online test preparation services, although these services are limited by its affiliation with the College Board. Infra para. 28 124. 9 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 began questioning whether he was intelligent enough to succeed on the SAT, only to realize that 2 the test was entirely unable to measure his many accomplishments. After paying for private 3 tutoring, Kawika’s friends improved their scores on both the PSAT and SAT. Kawika decided not 4 to take the SAT at the same time as many of his peers because he felt underprepared. 5 16. Kawika’s lifelong dream has been to attend either UC Berkeley or UCLA. He takes 6 pride in the UC system and wishes to study in its leading social science departments. Attending 7 college at a UC campus would afford Kawika the opportunity to continue working with 8 California-based community organizations and to apply his college education and experiences to 9 building a more equitable California. Upon learning that he would have to take the SAT to be 10 considered at any UC campus, Kawika was forced to focus his application efforts on East Coast 11 colleges with test-optional policies. Kawika is scheduled to take the SAT in December 2019, but 12 unless his scores increase dramatically, he is unlikely to gain admission at any UC campus. 13 Kawika’s lifelong ambition to attend UC Berkeley or UCLA has been stymied by UC’s 14 requirement that he submit test scores that fail to measure his true abilities. UC’s test score 15 requirement has subjected Kawika to unlawful discrimination on the basis of race and wealth and 16 has impaired his access to public higher education. 17 17. Plaintiff Gloria D. is a 17-year-old Latinx student who is a senior at a private high 18 school in California. Gloria D.’s first language is Spanish. As a young child, Gloria woke up at 19 four a.m. every day to commute across the border to attend school. 20 18. Gloria D. is a strong student. She has taken a rigorous course load, and has a 21 weighted GPA of 4.0 and an unweighted GPA of 3.78. Gloria D. aspires to be a doctor and has 22 thus taken more science classes than her school requires, including an AP science class. She has 23 received multiple academic awards and is a member of the National Honor Society. Outside of 24 school, Gloria D. volunteers at a hospital and is a dancer. 25 19. Gloria D. has taken the SAT twice: once in the spring of her junior year, and once 26 in the fall of her senior year. As a student for whom English is a second language, Gloria D. has 27 experienced difficulties with the phrasing of SAT questions, including word problems in the math 28 section. She takes a long time to read and understand each word problem, and when she realizes 10 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 that she is running out of time, she panics and guesses the answer. She has subsequently reviewed 2 word problems she answered incorrectly, only to realize that she understood the underlying 3 mathematical concept, but did not understand the question as phrased. As a result, Gloria D. gets 4 nervous when taking college admissions tests. She has experienced headaches when taking mock 5 exams. 6 20. Due to these challenges, Gloria D. scored 1110 on her first SAT. Because Gloria 7 D.’s high school does not offer SAT or ACT test preparation, Gloria D.’s mother has had to 8 devote the family’s limited funds to other test preparation resources. Family friends referred 9 Gloria D.’s mother to a private tutor—who was reputedly very effective—but at $250 per hour, 10 the cost was prohibitive. Instead, Gloria D. and her mother relied on free and relatively low-cost 11 test preparation resources: Gloria D. used Khan Academy’s test preparation programs, and her 12 mother bought her SAT and ACT test preparation books and paid $45 per hour for Gloria D. to 13 meet with an individual who tutored her on the SAT and advised her on the college admissions 14 process. In total, Gloria D.’s mother has spent approximately $800 on college admissions tests and 15 test preparation for Gloria D. This amount is both significant and burdensome for Gloria D.’s 16 mother, who contributes to her children’s private school tuition and also spends large amounts to 17 fund their extracurricular activities. Despite the time, effort, and resources that she and her mother 18 devoted to test preparation, Gloria D. achieved only a 30-point increase at her second SAT sitting, 19 raising her score to 1140. 20 21. Encouraged by her high school to pursue any career to which she aspires, Gloria D. 21 is determined to become a doctor. Gloria D.’s lifelong ambition has been to attend UCLA. The UC 22 system appeals to Gloria D. because its campuses are known as great research schools, which 23 Gloria D. believes will benefit her when she applies to medical school. For Gloria D., obtaining a 24 degree from a UC campus would be like obtaining a “seal of approval” when applying to medical 25 school and pursuing a career as a doctor. Attending a UC campus would also be more affordable 26 for Gloria D. and her family than attending a private college. Gloria D. has applied to three UC 27 campuses, but due to her SAT scores, she is unlikely to gain admission at any of them. When 28 Gloria D. visited one UC campus, a student told her that with her SAT scores, she had no chance 11 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 of acceptance. UC’s requirement that Gloria D. submit test scores that fail to measure her true 2 abilities has devastated Gloria D. and her mother and shattered her dream of attending UCLA. 3 UC’s test score requirement has subjected Gloria D. to unlawful discrimination on the basis of 4 race and wealth and has impaired her access to public higher education. 5 22. Plaintiff Stephen C. is a 16-year-old student who is a sophomore at a public high 6 school in California. He plans to apply to several UC campuses for undergraduate education. 7 Stephen C. has a history of disabilities that have affected his testing performance. Since seventh 8 grade, Stephen C. has had an official school accommodations plan pursuant to Section 504 of the 9 Rehabilitation Act of 1973 (his “504 Plan”), which allows him academic accommodations 10 including extended time on exams, the ability to take exams in a distraction-reduced setting, and 11 preferential seating in class. Stephen C. currently has a 4.0 GPA and takes one of the most 12 rigorous course loads offered at his high school. He is also a varsity athlete. 13 23. Due to his disabilities, Stephen C. requires testing accommodations to take the SAT 14 or ACT. Recently, his high school and school district changed their SAT and ACT 15 accommodation policies. Due to the burdens that the standardized testing accommodations process 16 impose on his high school’s counseling department, Stephen C.’s high school will no longer apply 17 for SAT or ACT accommodations for any of its students or assist students or families with this 18 process. This will make the process more difficult to complete for Stephen C. and other students in 19 his school district, including the 45 percent of students who are socioeconomically disadvantaged. 20 These students will be less likely to receive the testing accommodations to which they are entitled. 21 Moreover, not all test sites permit students with disabilities to take the SAT or ACT with 22 necessary accommodations, such that—unlike examinees without disabilities—Stephen C. may 23 not be able to choose his preferred test location when he registers. Accordingly, he faces 24 significant hurdles to even take the SAT or ACT with the accommodations he needs. Without 25 these necessary accommodations, Stephen C. will be unable to perform to his full ability on the 26 SAT or ACT. 27 24. Stephen C. has also confronted the stigma associated with requesting and using the 28 testing accommodations he requires to demonstrate his abilities. Many of his peers and their 12 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 parents consider the use of accommodations—even by students who are legally entitled to them 2 and whose 504 Plans or Individualized Education Programs (“IEPs”) provide for them—to be 3 “cheating.” He has found that certain individuals, even those who work with children, consider 4 standardized test accommodations to be a “privilege” instead of a legal right. Following the 5 negative press surrounding the recent college admissions scandal, Stephen C. has had to weigh 6 using the accommodations he needs against the embarrassment he will experience and suspicion 7 he will face if his peers or their parents—or even college admissions officers—find out. 8 25. Stephen C. knows that his standardized test scores cannot predict his ability to 9 succeed in college, and that his scores will be an even worse predictor of his college performance 10 if he is unable to obtain the accommodations he needs. He knows that he will have to spend an 11 unreasonable amount of time and money seeking appropriate test accommodations and preparing 12 for the tests, and that this will be extremely burdensome for him and his family. UC’s reliance on 13 SAT and ACT scores to evaluate applicants thus jeopardizes Stephen C.’s ability to be fairly 14 considered in the UC admissions process. UC’s requirement that all applicants submit an SAT or 15 ACT score to be considered for admission has subjected Stephen C. to unlawful discrimination on 16 the basis of disability and has impaired his access to public higher education. 17 26. Plaintiff Alexandra Villegas is an 18-year-old Latinx first-year student in the 18 Honors Transfer Program at Pasadena City College. Alex was raised without a father in a family 19 of six children, and her mother died when Alex was five years old. Together with her siblings, 20 Alex had significant family responsibilities throughout her childhood, including caring for her 21 brother with Down syndrome. Despite these challenges, Alex attended and excelled at Los 22 Angeles public schools from elementary through high school. Alex has a longstanding passion for 23 animals and the environment, which she developed as a middle school student in the 24 Environmental Studies Magnet program at Thomas Starr King Middle School. Participation in the 25 program, which teaches students about environmental issues and community engagement through 26 a project-based curriculum, sparked Alex’s interest in pursuing a career working with animals. 27 27. A strong and motivated student, Alex applied for and was accepted to John 28 Marshall High School’s School for Advanced Studies (SAS). SAS sites are so designated for their 13 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 exemplary gifted/talented education programs, “which offer high-level academic opportunities that 2 meet the unique educational needs of K-12 gifted learners.” 18 In SAS, Alex appreciated being 3 surrounded by other talented and driven students, who pushed each other to perform well in their 4 coursework. Alex’s favorite classes were English and chemistry. Outside of the classroom, Alex 5 played on her high school’s soccer team and participated in Habitat for Humanity. She graduated 6 in June 2019 with a cumulative GPA of approximately 3.4. 28. 7 Alex’s standardized test scores, however, failed to capture her gifts and motivation. 8 Alex took the PSAT as a high school sophomore, and received a score of 950. She took the SAT 9 two times, receiving a score of 1040 on the first administration and 1050 on the second. Alex 10 experienced anxiety while taking the SAT, particularly during the essay section of the exam. 11 During that section, she felt “giddy” under the time pressure and struggled to begin her essay. 12 Alex found it difficult to make all of her points clearly and concisely while worrying about 13 whether she would have sufficient time to finish. 29. 14 Alex could not afford expensive test preparation courses, and so did not participate 15 in any test preparation prior to her first SAT administration. After that test, Alex learned about 16 Khan Academy, which she used to practice test questions before her second SAT administration. 17 Alex also met with a family connection three or four times to work through SAT math problems. 18 Alex’s score on her second SAT administration increased by only 10 points. 30. 19 Alex wanted to attend the UC so that she could study among the best students in 20 the State. Although UCLA was Alex’s ideal college, the high average SAT and ACT scores of its 21 students deterred her from applying. Alex qualified for an application fee waiver, which allows 22 students who would otherwise be unable to afford it the ability to apply to up to four UC campuses 23 for no fee. Instead of applying to UCLA, where she felt she had no chance of admission due to her 24 SAT scores, Alex chose to apply to UC Davis, UC Merced, UC Riverside, and UC Santa Cruz. Of 25 those schools, UC Davis—with its College of Agricultural and Environmental Sciences—appealed 26 most strongly to Alex, although she did not feel confident when applying, again because of her 27 28 18 L.A. Unified Sch. Dist., SAS Home: What Are Schools for Advanced Studies?, http://echoices.lausd.net/sas. 14 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 SAT scores. UC Davis and UC Santa Cruz denied Alex admission. UC Riverside placed Alex on 2 its waitlist, but did not ultimately grant her admission. Although Alex was offered admission at 3 UC Merced, she had little familiarity with its campus and programs, and was concerned about 4 whether those programs would meet her educational needs and goals. Although Alex was capable 5 of succeeding academically at the other UC campuses to which she applied, and which would have 6 been a better fit, her SAT scores prevented her from gaining admission to those campuses. Alex 7 ultimately decided to attend Pasadena City College in the hope that by performing well there, she 8 would have the opportunity to transfer to a UC campus that is closer to her family or more suited 9 to her academic interests. 10 31. Alex aspires to a career in which she can work with animals. She dreams of putting 11 her passion for the environment into practice by running her own farm and caring for many 12 animals. UC’s requirement that all applicants submit SAT or ACT scores has forced Alex to defer 13 this dream. Instead, Alex is enrolled at a community college with fewer supports for low-income 14 students than the UC, and waiting until her SAT scores no longer constitute a barrier to admission 15 at the UC campuses she hopes to attend. UC’s requirement that all applicants submit an SAT or 16 ACT score to be considered for admission has subjected Alex to unlawful discrimination on the 17 basis of race and wealth and has impaired her access to public higher education at the UC. 18 Organizational Plaintiffs 19 32. Plaintiff Chinese for Affirmative Action (“CAA”) is a community-based civil 20 rights organization in San Francisco, California. CAA’s constituents include individuals who 21 reside in California and pay State taxes. The mission of CAA is to defend the civil and political 22 rights of Chinese Americans and to advance multiracial democracy. Programs at CAA currently 23 encompass direct services, leadership development, and civic engagement programs that prioritize 24 the needs and aspirations of its low-income, limited-English proficient, immigrant community 25 members. For five decades, CAA efforts in education equity have featured direct work with 26 English language learners and limited-English proficient parents to ensure that all families have 27 equitable access to high-quality public K-12 education and public higher education. 28 15 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 33. CAA has identified the SAT and ACT as a major discriminatory barrier to college 2 access for its constituents and for all students of diverse backgrounds, including those who are 3 English language learners. In addition, CAA believes that the UC’s SAT and ACT requirement 4 has harmfully skewed the priorities and resources of the families it serves and the schools they 5 attend away from initiatives that would improve high quality and more equitable education 6 opportunities. CAA’s interests, as well as the rights and interests of its constituents and 7 organization stakeholders, are adversely affected by Defendants’ actions and inactions as alleged 8 in this complaint, and CAA has diverted significant organizational resources to counteract 9 Defendants’ discriminatory testing requirements. Neither the claims asserted nor the relief 10 requested in this complaint require the participation of CAA’s individual constituents. 11 34. Plaintiff College Access Plan (CAP) is a nonprofit organization based in 12 Pasadena, California. CAP works to address the gap in college knowledge and matriculation 13 among low-income and college-going students in Pasadena’s public schools. Working closely 14 with counselors and teachers, CAP has developed a comprehensive program that works in 15 partnership with schools to add meaningful and individualized college readiness support that is 16 open to all students. CAP’s opt-in programs prioritize student-driven relationships with trusted 17 adults to embolden students of all ages to envision themselves with a college degree. Because 18 access to education is a right, not a privilege, CAP never selects students based on limiting criteria 19 like grades, test scores, and legal status. 20 35. In addition to its other programs, CAP provides free SAT test preparation courses 21 to students who attend Pasadena’s public high schools. Because the SAT can be stressful for 22 students, CAP’s curriculum addresses not only content and question design, but also students’ test23 related anxiety, using metacognitive tools that consider both what students know and what they 24 feel. CAP knows that these courses are a stopgap measure that cannot eliminate the inequities 25 imposed by these tests on its own students and on students across the State. CAP’s interests, as 26 well as the rights and interests of its members, are adversely affected by Defendants’ actions and 27 inactions as alleged in this complaint, and CAP has diverted significant organizational resources to 28 counteract Defendants’ discriminatory testing requirement, including the time, effort, and 16 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 resources it devotes to its test preparation programs that it could have been spending on 2 substantively preparing students for higher education. Neither the claims asserted nor the relief 3 requested in this complaint require the participation of CAP’s individual members. 4 36. Plaintiff College Seekers is a nonprofit community-based organization whose 5 mission is to improve access to higher education for underserved populations in California. 6 College Seekers serves non-traditional learners, especially those who homeschool, use public 7 independent study, or attend non-traditional schools. College Seekers students include 8 underrepresented minorities, students from lower-income families, and students from remote or 9 underserved communities. College Seekers currently has over 2,500 members, and typically adds 10 between 100 to 150 new members per month. A high percentage of College Seekers students have 11 learning differences or disabilities, or have suffered trauma or bullying in traditional high school 12 environments. 13 37. College Seekers helps create access to postsecondary options through social capital, 14 awareness of college opportunities, and access to accurate information. College Seekers has seen 15 the lack of access to standardized testing constitute a consistent barrier to higher education for its 16 students. The SAT and ACT accommodations request process and its attendant costs have caused 17 College Seekers parents to experience distress, frustration, and concern, which has led some 18 parents to urge their children to attend community college and then transfer rather than take the 19 SAT or ACT without accommodations. Even when College Seekers students are approved for 20 accommodations, they may struggle to find test sites willing to accommodate them if their own 21 schools are not approved test centers. This increased difficulty and expense reduces access to 22 postsecondary opportunities for students with disabilities, particularly those from disadvantaged 23 and underserved backgrounds and rural areas. UC’s requirement that all applicants submit SAT or 24 ACT scores in ordered to be considered for admission harms College Seekers students, who face 25 significant barriers to accessing these tests with necessary accommodations. The interests of 26 College Seekers and the rights and interests of its members are adversely affected by Defendants’ 27 actions and inactions as alleged in this complaint, and College Seekers has diverted significant 28 organizational resources to counteract Defendants’ discriminatory testing requirement. Neither the 17 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 claims asserted nor the relief requested in this complaint require the participation of individual 2 members of College Seekers. 3 38. Plaintiff Community Coalition is a nonprofit social justice organization based in 4 South Los Angeles that helps students reach their full potential and become agents of change. 5 Community Coalition prepares Black and Latinx students for college and success in the twenty6 first century economy and in institutions where they have historically been denied opportunity. 7 Over the last 28 years, Community Coalition’s youth program—South Central Youth Empowered 8 thru Action (“SCYEA”)—has trained high school students to push for reforms that disrupt the 9 “school to prison pipeline,” including by helping young people from South Los Angeles access 10 higher education. Through this work, Community Coalition has repeatedly seen that the UC’s 11 SAT or ACT score requirement impedes that access—without providing any meaningful 12 information to the admissions process—and prevents South Los Angeles youth from achieving 13 their potential and realizing their dreams. Community Coalition has witnessed the UC’s SAT or 14 ACT score requirement impose an obstacle to the advancement of the next generation of young 15 leaders, who are already fighting to transform social and economic conditions in South Los 16 Angeles. Community Coalition’s interests, as well as the rights and interests of its members, are 17 adversely affected by Defendants’ actions and inactions as alleged in this complaint, and 18 Community Coalition has diverted significant organizational resources to counteract Defendants’ 19 discriminatory testing requirement. Neither the claims asserted nor the relief requested in this 20 complaint require the participation of Community Coalition’s individual members. 21 39. Plaintiff Dolores Huerta Foundation is a community benefit organization based 22 in California’s Central Valley. Dolores Huerta Foundation recruits, trains, organizes, and 23 empowers grassroots leaders in low-income communities to attain social justice through systemic 24 and structural transformation. Dolores Huerta Foundation’s education program empowers parents 25 to advocate for the rights of their students by disrupting the school-to-prison pipeline and creating 26 more support for low-income students, multilingual learners, and students with disabilities. 27 Dolores Huerta Foundation’s Vecinos Unidos (neighborhood organizations) have also successfully 28 18 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 advocated for education reforms that will enable students from communities that are primarily 2 Latinx, immigrant, and low-income to achieve their full potential. 3 40. UC’s requirement that all applicants submit an SAT or ACT score creates a 4 discriminatory barrier to college access for the students Dolores Huerta Foundation serves. 5 Dolores Huerta Foundation believes that eliminating this obstacle will enhance its efforts to ensure 6 that low-income students, multilingual learners, and students with disabilities have equal 7 opportunity to attain success in college and beyond. Dolores Huerta Foundation’s interests, as well 8 as the rights and interests of its members and member organizations, are adversely affected by 9 Defendants’ actions and inactions as alleged in this complaint, and Dolores Huerta Foundation has 10 diverted significant organizational resources to counteract Defendants’ discriminatory testing 11 requirement. Neither the claims asserted nor the relief requested in this complaint require the 12 participation of Dolores Huerta Foundation’s individual members. 13 41. Plaintiff Little Manila Rising is a community-based nonprofit organization based 14 in Stockton, California. Little Manila Rising provides education and leadership to revitalize 15 Stockton’s Filipinx American community and advocates for the historic preservation of the Little 16 Manila Historic Site in Stockton. Serving local Stockton high school students, the Little Manila 17 After School Program is an Ethnic Studies program focused on Philippine and Filipinx American 18 history, culture, politics, art, community responsibility, and college access. Through this program, 19 Little Manila Rising has seen the UC’s SAT or ACT score requirement constitute a discriminatory 20 barrier to college access for students from Stockton’s historically disenfranchised Filipinx 21 community. Little Manila Rising’s interests, as well as the rights and interests of its members, are 22 adversely affected by Defendants’ actions and inactions as alleged in this complaint, and Little 23 Manila Rising has diverted significant organizational resources to counteract Defendants’ 24 discriminatory testing requirement. Neither the claims asserted nor the relief requested in this 25 complaint require the participation of Little Manila Rising’s individual members. Defendants 26 27 42. Defendant Regents of the University of California (“Regents”) is a California 28 public corporation responsible for administering the University of California, a public trust, 19 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 pursuant to Article IX, Section 9 of the California Constitution. Pursuant to Standing Order 105.2, 2 although the Regents have delegated to the Academic Senate the determination of “the conditions 3 for admission” to the UC, the Regents retain the authority to approve those conditions. 19 4 43. Defendant Janet Napolitano, sued here in her official capacity, is President of the 5 UC and an ex officio Regent. Pursuant to Standing Order 100.4, as the “executive head of the 6 University,” Defendant Napolitano has, with limited exclusions, “full authority and responsibility 7 over the administration of all affairs and operations of the University.” 20 “The Office of the 8 President is responsible for guiding policy decisions on UC enrollment growth and managing 9 UC’s online application.”21 10 44. Defendants Does 1 through 100 are persons whose true names and capacities are 11 presently unknown to Plaintiffs, who therefore sue these Defendants by fictitious names. Doe 12 Defendants include current and past individual Regents of the University of California who have 13 approved and tolerated the discriminatory policy of requiring SAT and ACT test scores as an 14 admissions criterion, as well as past and current members of BOARS who developed, approved, 15 and tolerated UC’s current testing requirement for admissions. Plaintiffs will amend this 16 Complaint to show Doe Defendants’ true names and capacities when they have been ascertained. 17 Plaintiffs are informed and believe, and herein allege, that such Doe Defendants are residents of 18 California. 19 20 21 22 23 24 19 Univ. of Cal. Bd. of Regents, Standing Order 105.2: Duties, Powers, and Privileges of the 25 Academic Senate, https://regents.universityofcalifornia.edu/governance/standingorders/so1052.html. 26 20 Univ. of Cal. Bd. of Regents, Standing Order 100.4: Duties of the President of the University, 27 https://regents.universityofcalifornia.edu/governance/standing-orders/so1004.html. 28 21 Univ. of Cal. Office of the Pres., Office of the President, https://www.universityofcalifornia.edu/uc-system/office-of-the-president. 20 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 FACTUAL ALLEGATIONS 2 I. Meaningless Tests 3 A. SAT and ACT 4 45. The SAT is owned by the College Board, which had roughly $1.068 billion in 5 revenue in 2017.22 Its creator, eugenicist Carl Brigham, based the test on previous IQ tests that 6 purported to measure “native intelligence” and aptitude, as suggested by its original name, the 7 “Scholastic Aptitude Test.”23 In the years following World War I, IQ tests were used to channel 8 Black students into vocational education tracks, “justify[ing] educational systems that mainly 9 reproduced extant socio-economic” and racial inequalities. 24 The SAT was first administered in 10 1926.25 46. 11 Use of the SAT grew over the next two decades, primarily amongst Ivy League 12 universities and other private institutions on the East Coast. 26 In 1947, the Educational Testing 13 Service (ETS) was founded to consolidate the development and administration of standardized 14 tests, including the SAT.27 As the testing industry grew, so too did ETS’s budget, burgeoning from 15 approximately $2 million in 1947 to nearly $30 million in 1969.28 In 1959, the ACT entered the 16 testing market, touted as an alternative to the SAT that was purportedly more curriculum-based. 29 17 18 22 ProPublica, Nonprofit Explorer: College Board, 19 https://projects.propublica.org/nonprofits/organizations/ 20 131623965. 23 See Nicholas Lemann, The Big Test: The Secret History of American Meritocracy 29–31, 34 21 (2000). 22 23 24 Wayne Au, Hiding Behind High-Stakes Testing: Meritocracy, Objectivity and Inequality in U.S. Education, 12 Int’l Educ. J.: Comparative Perspectives 7, 9–10 (2013). 25 Douglass, supra note 5, at 83. 24 26 Id. at 84. 25 27 26 Clarence J. Karier, Testing for Order and Control in the Corporate Liberal State, 22 Educ. Theory 154, 174 (1972). 28 Id. In 2017, ETS had approximately $1.407 billion in revenue. ProPublica, Nonprofit Explorer: 27 Educational Testing Service, https://projects.propublica.org/nonprofits/organizations/210634479. 28 29 Douglass, supra note 5, at 84. 21 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 47. 1 During the late 1940s and 1950s, ETS aggressively lobbied the UC to adopt the 2 SAT.30 As discussed infra paras. 140–43, although the UC was initially reluctant to adopt the test 3 due to concerns about its lack of validity, it ultimately incorporated the test into its admissions 4 process for the sake of administrative convenience—as a tool to “cull [its] eligibility pool.” 31 UC’s 5 adoption of the SAT was a significant financial boon for ETS: as “the largest single user of the 6 SAT,” UC “helped solidify the SAT as the most widely used national admissions test” until 7 2012.32 48. 8 Despite the test’s ubiquity, socioeconomic and racial disparities in SAT scores led 9 to public criticism of the test as a discriminatory instrument that reinforced social and economic 10 inequality.33 These critiques led the College Board to repeatedly rebrand the test in an (ultimately 11 futile) attempt to define what it was actually measuring. In 1994, the College Board 12 redenominated the SAT as the “Scholastic Assessment Test,” seeking to demonstrate the test’s 13 shift in focus away from “aptitude” and toward curricular knowledge. 34 This rebranding was short14 lived, however: after only three years, the College Board gave up on using the test’s name to 15 describe the material tested, and now, by the College Board’s own account, SAT “doesn’t stand 16 for anything[.]”35 Commentators have suggested that, in light of how SAT scores track student 17 18 30 19 Id. at 84–86. 31 Saul Geiser, Norm-Referenced Tests and Race-Blind Admissions: The Case for Eliminating the 20 SAT and ACT at the University of California 7 (Dec. 2017), available at https://cshe.berkeley.edu/sites/default/files/publications/ 21 2.rops.cshe.15.2017.geiser.testsrace-blind_admissions.12.18.2017.pdf. 22 32 Saul Geiser, The Growing Correlation Between Race and SAT Scores: New Findings from California 1 (Oct. 2015), available at 23 https://cshe.berkeley.edu/sites/default/files/publications/rops.cshe_.10.15.geiser.racesat. 24 10.26.2015.pdf. 25 33 Paul Tough, The Years That Matter Most: How College Makes or Breaks Us 73–74 (2019). 34 Peter Applebome, Insisting It’s Nothing, Creator Says SAT, Not S.A.T., N.Y. Times (Apr. 2, 26 1997), 27 https://www.nytimes.com/1997/04/02/us/insisting-it-s-nothing-creator-says-sat-not-sat.html. 28 35 Id. 22 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 wealth and accumulated advantage, SAT in practice stands for the “Socioeconomic Advantage 2 Test.” 49. 3 The College Board also adapted the format of the test in response to the ACT’s 4 gains in national market share and critiques that the SAT was insufficiently tied to high school 5 curriculum. Following a proposal by then-UC President Richard Atkinson to drop the SAT as a 6 UC admissions criterion, the College Board in 2005 repackaged the SAT into an ostensibly more 7 curriculum-based exam, adding a writing section and making marginal changes to the test’s 8 existing sections.36 In 2014, to increase its competitiveness with the ACT, whose market share had 9 grown larger than that of the SAT, the College Board announced the 2016 redesigned SAT. 37 10 According to the College Board, the 2016 redesigned SAT is “more closely linked with rich, 11 rigorous course work” than its predecessors, 38 although recent studies have refuted the alignment 12 between the SAT and Common Core State Standards.39 The redesigned SAT eliminated the 13 penalty for guessing incorrectly, dropped arcane vocabulary words, focused the math sections on 14 “linear equations; complex equations or functions; and ratios, percentages and proportional 15 reasoning,” and made the essay optional.40 Despite these changes, the SAT remains of limited 16 predictive value, while continuing to adversely affect underrepresented minority students, students 17 with disabilities, and students with less wealth. 18 19 20 21 36 Infra para. 146. 37 Tamar Lewin, A New SAT Aims to Realign With Schoolwork, N.Y. Times (Mar. 5, 2014), 22 https:// www.nytimes.com/2014/03/06/education/major-changes-in-sat-announced-by-college-board.html 23 38 College Board, Test Specifications for the Redesigned SAT 11 (2015), available at https:// 24 collegereadiness.collegeboard.org/pdf/test-specifications-redesigned-sat-1.pdf. 39 Michal Kurlaender & Kramer Cohen, Predicting College Success: How Do Different High School Assessments Measure Up? 21 n.20 (Mar. 2019), available at 26 https://edpolicyinca.org/sites/default/files/R_Kurlaender_Mar-2019.pdf. 40 Lewin, supra note 37. As discussed infra paras. 94–97, the redesigned SAT also featured an 27 increased number of lengthy word problems, placing multilingual learners and students with 28 certain learning disabilities at a significant disadvantage. 25 23 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 50. 1 Notably, the College Board itself has demonstrated that the SAT treats students 2 with less accumulated advantage unfairly. In 2010, College Board researchers studied which 3 groups of students tend to have “discrepant” SAT scores, meaning SAT scores that are either 4 much higher or lower than their high school grades would predict. 41 Its findings were striking: 5 whereas students with inflated scores were “disproportionately male, affluent, white or Asian, and 6 with highly educated parents,” students with scores that were markedly lower than their high 7 school grades would predict were “disproportionately female, black or Latin[x], low-income, and 8 first-generation.”42 Unsurprisingly, the College Board omits this study from the extensive list of 9 research publications on its website.43 51. 10 As discussed infra paras. 60–61, the College Board’s “Landscape” tool—which 11 provides college admissions officers with socioeconomic background information to supplement 12 students’ SAT scores—effectively acknowledges both that SAT scores act as a proxy for 13 socioeconomic status and that the scores taken alone cannot accurately measure the potential of 14 less advantaged students to succeed in college. 15 B. Meaningless Test Results 16 52. The chief purpose of the SAT and ACT is to predict student performance in 17 college.44 According to the tests’ proponents, the SAT and ACT provide a common, standardized 18 metric that enables college admissions officers to sort among large numbers of applicants. 45 But 19 the SAT and ACT fail to deliver on this purpose. Rather than providing meaningful information 20 about a student’s ability to succeed in college, the SAT and ACT are largely a proxy for a 21 student’s socioeconomic background and race. 22 23 41 Tough, supra note 33, at 173–74. 24 42 Id. at 174–75. 25 43 Id. 26 44 Geiser, supra note 31, at 1. 45 Richard C. Atkinson & Saul Geiser, Reflections on a Century of College Admissions Tests 17 (Apr. 2009), available at https://cshe.berkeley.edu/sites/default/files/publications/rops28 atkinsongeiser-tests-04-15-09.pdf. 27 24 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 53. 1 These facts—that the SAT and ACT have extremely limited predictive value and 2 act as a stand-in for students’ wealth and race—are well known to the Regents, and have been for 3 decades.46 On information and belief, in addition to publicly available BOARS reports, the 4 Regents, the Academic Senate, and BOARS possess studies and memoranda—discussing both the 5 SAT and ACT’s weak predictive value and the discrimination that results from their use in 6 admissions decisions—that have not been made public. On information and belief, the Regents 7 have deliberately kept these studies and memoranda concealed so as to mute criticism of their 8 continued use of meaningless and discriminatory test scores. The Regents are also well aware that 9 studies of the tests’ predictive ability fail to take into account the use by affluent students of 10 expensive test preparation services, such as private tutoring. 47 BOARS has “confirm[ed] that 11 students can be coached, to advantage, for both the old and new SAT,” and that “[c]oaching 12 among some students and not others disadvantages low-income groups.” 48 There is no defensible 13 reason for the Regents to ignore this information in continuing to use the test scores as a criterion 14 of admission. 15 SAT and ACT as a Proxy for Socioeconomic Status 16 54. More than any other admissions criterion, SAT and ACT scores act as a proxy for 17 students’ wealth and accumulated advantage. 49 When compared with other metrics like high 18 school GPA and class rank, SAT and ACT scores are more highly correlated with socioeconomic 19 characteristics like family income and parents’ education. 50 In other words, the more money a 20 student’s parents earn, the higher that student’s SAT and ACT scores are likely to be. Similarly, 21 the more education a student’s parents have, the higher his or her SAT and ACT scores are likely 22 to be. 23 24 46 25 Infra paras. 140–49, 153. 47 See Watanabe, supra note 13; BOARS, supra note 7, at 21. 26 48 BOARS, supra note 7, at 21. 27 49 Geiser, supra note 31, at 1. 50 Id. 28 25 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 55. 1 Among applicants to the UC, this trend has increased dramatically since 1994. 51 Figure 1: Variance in SAT/ACT Scores and High School GPA Explained by Family Income, Parents’ Education, and Race/Ethnicity52 2 3 4 5 6 7 8 9 10 11 12 13 14 56. 15 Figure 1 illustrates the growing importance of socioeconomic background 16 characteristics—family income, parents’ education, and race—in predicting UC applicants’ test 17 scores. As depicted in Figure 1, the correlation between socioeconomic characteristics and test 18 scores has risen sharply over time: whereas socioeconomic characteristics accounted for 23 19 percent of the variance in UC applicants’ SAT and ACT scores in 1995, they accounted for 39 20 percent of test score variance in 2016. In other words, students’ socioeconomic characteristics— 21 rather than their individual merit—predict almost 40 percent of the variation in their SAT and 22 ACT scores. Figure 1. 57. 23 By contrast, socioeconomic characteristics are substantially less important in 24 predicting students’ high school grades. Figure 1. In 1995, socioeconomic factors accounted for 25 26 51 Id. at 2. 52 Saul Geiser, Ctr. for Studies in Higher Educ., Univ. of Cal., Berkeley, UC and the SAT/ACT: 27 Research Findings 1994-2019, Presentation at Policy Analysis for California Education (PACE) 28 Conference: Toward a Vision of Equity in College Access: Re-evaluating College Admissions (Nov. 22, 2019). 26 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 only five percent of the variance in students’ high school GPAs, but accounted for 23 percent of 2 the variance in UC applicants’ SAT and ACT scores. The disparity in predictive value increased 3 over time: by 2016, socioeconomic characteristics accounted for only nine percent of the variance 4 in students’ high school GPAs, compared to 39 percent of the variance in applicants’ SAT and 5 ACT scores. Whereas the predictive value of socioeconomic characteristics with respect to test 6 scores rose dramatically between 1995 and 2016, the predictive value of those same characteristics 7 with respect to high school GPA remained low and relatively constant over the same period. 8 Figure 1. 58. 9 These differences are meaningful. The fact that socioeconomic characteristics 10 account for almost 40 percent of the variation in applicants’ SAT and ACT scores—as compared 11 to just nine percent of the variation in applicants’ high school grades—means that the UC is 12 knowingly using a metric that weighs in favor of more affluent students relative to students with 13 less accumulated advantage. 59. 14 In May 2019, the College Board attempted to compensate for the SAT’s unequal 15 treatment of applicants from different groups by introducing a planned “adversity score” that 16 purported to quantify students’ socioeconomic hardship. Designed to contextualize SAT scores for 17 college admissions officers, the “adversity score” was to be calculated based on 15 factors, 18 including “the relative quality of the student’s high school and the crime rate and poverty level of 19 the student’s neighborhood,” with higher scores connoting greater levels of “disadvantage.” 53 20 According to the College Board’s CEO, without such information on socioeconomic background 21 characteristics, “the SAT could be misleading.” 54 The College Board’s concession, through its 22 adoption of the “adversity score,” that SAT scores alone cannot accurately measure the potential 23 of less advantaged students to succeed in college led to widespread condemnation of not only the 24 25 53 Anemona Hartocollis, SAT’s New ‘Adversity Score’ Will Take Students’ Hardships Into 26 Account, N.Y. Times (May 16, 2019), https://www.nytimes.com/2019/05/16/us/sat-score.html. 54 Catherine Gewertz, Counselors Blast College Board’s Plan to Assign Students a ‘Disadvantage’ 27 Score, Educ. Week (May 20, 2019), https://www.edweek.org/ew/articles/2019/05/20/counselors28 blast-college-boards-plan-to-assign.html. 27 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 “adversity score,” but also the test itself. Education advocates questioned why colleges and 2 universities would continue to use a test after its own maker’s acknowledgment that the test 3 requires “a sophisticated contextual framework to make it valid.” 55 The College Board’s 4 announcement of the “adversity score” generated such a powerful backlash among educators and 5 parents that the company withdrew its plan only three months later. 56 60. 6 In response to this backlash, the College Board engaged in another rebranding: 7 although it no longer sums ratings of a student’s school characteristics and neighborhood 8 characteristics into a single numerical score, it continues to provide this data to college admissions 9 officers through the same ratings system—formerly known as the “Environmental Context 10 Dashboard,” which the College Board has renamed “Landscape.” 57 Landscape provides essentially 11 the same information on a student’s school and neighborhood characteristics as the former 12 “adversity score,” albeit in a slightly disaggregated format: Landscape averages six socioeconomic 13 and demographic indicators to provide a neighborhood rating and a high school rating, which are 14 “presented on a 1—100 scale,” with higher values indicating “higher level[s] of challenge related 15 to educational opportunities and outcomes.” 58 Practically speaking, therefore, the College Board 16 has not eliminated the “adversity score” at all: it just presents the score’s component information 17 separately. 61. 18 This superficial repackaging fails to address the hardships or challenges faced by 19 individual students, presenting instead data based solely on school and neighborhood averages. 20 Critically, Landscape also excludes information about the racial composition of a student’s school 21 and neighborhood, which significantly impacts educational outcomes. Ultimately, Landscape is a 22 post-hoc attempt to mitigate the SAT’s biases against less advantaged students rather than redress 23 24 55 25 56 26 27 Hartocollis, supra note 53. Anemona Hartocollis, SAT ‘Adversity Score’ Is Abandoned in Wake of Criticism, N.Y. Times (Aug. 27, 2019), https://www.nytimes.com/2019/08/27/us/sat-adversity-score-college-board.html. 57 Id. 58 College Board, Landscape Data and Methodology Summary (2019), available at https://secure28 media.collegeboard.org/landscape/data-methodology-summary.pdf. 28 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 the socioeconomic and racial disparities that make such mitigation necessary. As one commentator 2 opined, universities “should be trying to level the playing field by providing historically 3 disenfranchised people opportunities to build wealth rather than retrofitting test results around 4 inequality.”59 5 SAT and ACT as a Proxy for Race 6 62. As compared to other socioeconomic background characteristics—parents’ 7 education and family income—race has become the strongest predictor of students’ performance 8 on the SAT and ACT tests.60 After controlling for parents’ education and family income, race “has 9 a large, independent, and growing statistical effect” on applicants’ SAT and ACT scores. 61 10 /// 11 /// 12 13 /// /// 14 15 16 /// /// 17 /// 18 /// 19 /// 20 /// 21 /// 22 /// 23 /// 24 25 59 Andre M. Perry, Students need more than an SAT adversity score, they need a boost in wealth 26 (May 17, 2019), https://www.brookings.edu/blog/the-avenue/2019/05/17/students-need-morethan-an-sat-adversity-score-they-need-a-boost-in-wealth/. 27 60 Geiser, supra note 31, at 4. 28 61 Id. at 3. 29 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 Figure 2: Relative Weight of Family Income, Parents’ Education, and Race/Ethnicity in Predicting SAT/ACT Scores, 1994 to 201162 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 63. 17 Figure 2 illustrates the relative importance of three characteristics in predicting 18 students’ SAT and ACT scores: underrepresented minority status (self-identification as Black or 19 Latinx), parents’ education, and family income. 63 As shown in Figure 2, the strongest predictor of 20 UC applicants’ SAT and ACT test scores has historically been their parents’ level of education. 21 But while the predictive value of parents’ education has remained relatively constant over time, 22 family income and race have become increasingly important in predicting test results. Figure 2. 64. 23 Of the three characteristics, race has seen the most dramatic growth in its 24 importance as a predictor of students’ SAT and ACT scores, increasing from a standardized 25 coefficient of “0.23 in 1994 to 0.29 in 2011 (about a 55-point test-score differential).” 64 For the 26 27 28 62 Id. at 4. 63 Id. 64 Id. 30 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 first time in 2009 and again in 2011, race outweighed parents’ education and family income as the 2 single most important predictor of students’ SAT and ACT scores. Figure 2. Race now does a 3 better job of predicting how well a student will perform on the SAT and ACT than the level of 4 education the student’s parents have or the amount of money the student’s family earns. 5 Limited Predictive Value of SAT and ACT 6 65. Because SAT and ACT scores are so strongly correlated with race and wealth, it is 7 essential to control for those factors when assessing how well those scores predict college 8 performance,65 i.e., their “validity.”66 66. 9 But validity studies conducted by the College Board and ACT fail to do just that. 10 Instead, many such studies take into account only two predictors of college success: SAT or ACT 11 scores and high school GPA.67 Even without controlling for socioeconomic status, these studies 12 demonstrate that SAT and ACT scores add relatively little to the predictive power of high school 13 GPA. That predictive power dwindles even further once socioeconomic factors—which are 14 correlated with both SAT and ACT scores and college outcomes—are taken into account. By 15 failing to control for these factors, simplistic, two-predictor variable models like the College 16 Board’s and ACT’s exaggerate the predictive value of these tests. 68 Nor do such models account 17 for disparate access to expensive, high-quality test preparation services, which provide affluent 18 students a competitive edge over their less advantaged peers. Rather than predicting how well a 19 20 65 Saul Geiser & Maria Veronica Santelices, Validity of High School Grades in Predicting Student Success Beyond the Freshman Year: High-School Record vs. Standardized Tests as Indicators of 21 Four-Year College Outcomes 6–7, available at https://cshe.berkeley.edu/publications/validity22 high-school-grades-predicting-student-success-beyond-freshman-yearhigh-school. 66 Validity is a statistical term that “refers to the degree to which evidence and theory support the interpretations of test scores entailed by proposed uses of tests”—i.e., the degree to which a test 24 succeeds in measuring what it is supposed to measure. Am. Research Ass’n et al., Standards for Educational and Psychological Testing 9 (2014). 25 67 See, e.g., Paul A. Westrick et al., College Board, Validity of the SAT for Predicting First-Year 26 Grades and Retention to the Second Year 8–9 (May 2019), available at https://collegereadiness.collegeboard.org/pdf/ 27 national-sat-validity-study.pdf. 23 28 68 Geiser, supra note 31, at 8–9. 31 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 student is likely to perform in college, “much of the apparent predictive value of test scores 2 reflects the ‘proxy’ effect of [socioeconomic status].”69 67. 3 Relying on this flawed methodology, the College Board and ACT assert that SAT 4 and ACT scores are valid indicators of several measures of college success, including first-year 5 GPA, second-year retention (i.e., a student returning for his or her second year), cumulative GPA, 6 and college completion. But independent research demonstrates that after socioeconomic 7 background characteristics are taken into account, SAT and ACT scores add little meaningful 8 information about a student’s likelihood of college success as measured by any of these outcomes. First-Year GPA 9 68. 10 The College Board defends the use of the SAT through a series of validity studies 11 focused primarily on the SAT’s purported ability to predict first-year grades. 70 First-year grades 12 are a dubious metric: no student attends college to attain first-year grades, and no university 13 should seek to design its student body around that metric. Nor are first-year grades particularly 14 reliable indicators of how students will ultimately perform in college, given the myriad 15 nonacademic factors that can impact first-year academic performance—from living away from 16 home to adjusting to a new social and educational environment. 71 69. 17 In May 2019, the College Board released its first national validity study of the 18 current SAT.72 Like its predecessors, the study considered only two predictor variables: high 19 school GPA and SAT scores. 73 The correlations found between first-year college GPA, high 20 21 69 Id. at 8. 22 70 27 72 Westrick et al., supra note 67, at 4. 73 Id. at 8–9. See Jonathan Beard & Jessica Marini, College Board, Validity of the SAT for Predicting FirstYear Grades: 2013 SAT Validity Sample 2 (2018), available at 23 https://files.eric.ed.gov/fulltext/ED582459.pdf (citing series of College Board validity studies measuring correlations between SAT scores, high school GPA, and first-year grades). 24 71 Geiser & Santelices, supra note 65, at 17 (“[T]he first year or two in college is a difficult 25 transition period for many students who must adjust not only to the more rigorous academic standards of college but often as well to the experience of being away from home for the first 26 time.”). 28 32 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 school GPA, and scores on the current SAT are similar to those shown in previous College Board 2 validity studies.74 70. 3 The study found that, taken individually, high school GPA and SAT scores are 4 similarly correlated to first-year college GPA (“.53 and .51, respectively”). 75 In other words, taken 5 independently, high school GPA is slightly better at predicting a student’s first-year college GPA 6 than SAT scores are. According to College Board researchers, SAT scores are still important 7 because they purportedly add a 15 percent “boost” in the correlation above high school GPA 8 alone.76 Even according to the College Board, any usefulness of SAT and ACT scores in 9 admissions decisions derives entirely from whatever incremental increase in validity (i.e., increase 10 in predictive value) they are able to provide over high school GPA alone. 77 71. 11 ACT Research makes similar claims regarding correlations between first-year 12 college GPA, high school GPA, and ACT scores. An ACT Research study found correlations of 13 .58 between high school GPA and first-year college GPA and .51 between ACT scores and first14 year college GPA.78 ACT Research also claims that the ACT scores add incremental accuracy 15 beyond high school GPA in predicting first-year college GPA.79 16 17 18 19 20 21 22 23 74 For example, the correlation between SAT score and first-year GPA for the 2017 cohort (a sample of students entering college in fall 2017) was .51, and the correlation for cohorts between 2006 and 2010 ranged from .53 to .56. Id. at 10; Krista D. Mattern & Brian F. Patterson, College Board, Synthesis of Recent SAT Validity Findings: Trend Data over Time and Cohorts 76 (2014), available at https://files.eric.ed.gov/fulltext/ED556462.pdf. Similarly, the correlation of combined SAT score and high school GPA with first-year GPA for the 2017 cohort was .61, and between 2006 and 2010 it ranged between .62 and .64. Westrick et al., supra note 67, at 11; Mattern & Patterson at 76. 75 Westrick et al., supra note 67, at 11. 76 Id. at 4, 11. 77 Atkinson & Geiser, supra note 45, at 3 n.1 (“In real-world admissions, the key question is what 24 SAT scores uniquely add to the prediction of college outcomes, beyond what is already provided by a student’s HSGPA.”). 25 78 ACT, The ACT Technical Manual: Fall 2019 Version 3 11.37 (2019), available at 26 http://www.act.org/content/ dam/act/unsecured/documents/ACT_Technical_Manual.pdf. 27 79 Id. at 11.49 (“In most scenarios, using both high school grades and [ACT] scores jointly is 28 better than using either by itself.”). 33 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 72. 1 As independent researchers have long noted, however, the two-predictor variable 2 model most frequently utilized by the College Board and ACT leaves out important information 3 about students’ socioeconomic characteristics, and thus furnishes results that substantially 4 overstate the predictive value of the SAT and ACT.80 According to a UC economist who analyzed 5 UC test score data, “a conservative estimate is that traditional methods and sparse models”—like 6 those typically used by the College Board—“overstate the SAT’s importance to predictive validity 7 by 150 percent.”81 73. 8 This substantial overstatement of the SAT’s predictive value results from 9 traditional models’ failure to account for the proxy effect of socioeconomic status. 82 SAT scores 10 function as a stand-in for socioeconomic status, such that when either the scores or a student’s 11 demographic information is combined with high school GPA, the strength of the resulting 12 prediction about a student’s first-year grades is about the same. 83 Due to this proxy effect, studies 13 that omit socioeconomic characteristics from their prediction models “inflate[] the SAT’s apparent 14 validity.”84 This is because, rather than measuring differences in students’ ability to succeed in 15 college, SAT scores simply reflect the socioeconomic differences that themselves predict first-year 16 GPA.85 74. 17 When College Board studies do take socioeconomic characteristics into account, 18 the proxy effect is clear. In July 2019, the College Board released a follow-up to its national 19 20 21 80 See, e.g., Jesse Rothstein, College Performance Predictions and the SAT, 121 J. of Econometrics 297, 298 (2004); Geiser & Santelices, supra note 65, at 6. 22 81 Rothstein, supra note 80, at 315. 23 82 Id. 24 83 26 84 Id. at 314 (“[T]ogether with HSGPA, school and individual demographic variables explain 45% of the variance in FGPAs, about as much as do SAT and HSGPA together in models excluding 25 background variables.”). Id. at 315. 85 Id. (“[T]he SAT score appears to be a more effective measure of the demographic characteristics that predict UC FGPAs than it is of variations in preparedness conditional on student 28 background.”). 27 34 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 validity study for the current SAT, in which it examined validity results by student subgroups. 86 2 College Board researchers found that all correlations with first-year GPA—including SAT scores 3 alone and SAT scores combined with high school GPA—“tend[ed] to increase as parental 4 education level increases.”87 ACT Research has also found that ACT scores consistently increase 5 with parental education level.88 75. 6 Independent studies that take socioeconomic background characteristics into 7 account have confirmed that SAT and ACT scores add very little incremental validity to 8 predictions of first-year GPA. One such study, prepared in 2008 at the request of BOARS, 9 demonstrated the miniscule contribution of SAT and ACT scores to the prediction of first-year 10 GPA at the UC.89 The study considered all available information from students’ UC applications, 11 including information about students’ socioeconomic characteristics, such as their parents’ 12 education and family income.90 It found that, taken together, this information accounted for 21.7 13 percent of the variance in first-year GPA among students at UC Berkeley. 91 The study then 14 determined the unique contribution of SAT scores by removing them from the prediction model. 92 15 Without SAT scores, the explained variance decreased to 19.8 percent. 93 SAT scores alone thus 16 accounted for only 1.9 percent of the variance in first-year GPA among students at UC Berkeley. 94 17 18 86 Jessica P. Marini et al., Differential Validity and Prediction of the SAT: Examining First-Year 19 Grades and Retention to the Second Year 5 (July 2019), available at https://research.collegeboard.org/pdf/differential-validity-and-prediction-sat.pdf. 20 87 Id. at 16. 21 88 Dina Bassiri, ACT, ACT Composite Score by Parental Education Level, 2012–2016 (Oct. 22 2016), https:// www.act.org/content/dam/act/unsecured/documents/R1608_ACT_Composite_Score_by_Parental 23 _Education_Level_Web.pdf. 24 89 BOARS, supra note 6, at 90–132. 90 25 Id. at 90–92. 91 Id. at 107. 26 92 Id. 27 93 Id. 94 See id. 28 35 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 Across the UC system’s undergraduate campuses, SAT scores increased predictive value by only 2 1.6 percent.95 This result accords with earlier findings that, when controlled for socioeconomic 3 factors, SAT scores “add nothing to the prediction of freshman grades beyond that which [high 4 school GPA] and the SAT II already provide.” 96 76. 5 A study of more recent UC data demonstrates that California’s far less 6 discriminatory Smarter Balanced Assessment has the ability to predict first-year GPA as well as 7 the SAT. Although the correlation between first-year GPA and combined SAT scores and high 8 school GPA was stronger than that between first-year GPA and high school GPA alone, the 9 difference diminished after controlling for socioeconomic disadvantage, UC campus differences, 10 and high school characteristics.97 Once such factors were taken into account, the unique 11 contributions of the SAT and California’s statewide Smarter Balanced Assessment over high 12 school GPA alone were quite similar.98 Longer-Term College Outcomes 13 14 77. 15 The College Board and ACT also assert that SAT and ACT scores are valid 16 indicators of longer-term measures of college success, including second-year retention, cumulative 17 GPA, and college completion. Like College Board and ACT validity studies predicting first-year 18 GPA, however, many of these studies fail to control for socioeconomic background factors 19 entirely. A College Board study of four-year college completion, for example, stated that it 20 “intentionally excluded” socioeconomic status as a “predictor[] in our model, knowing that 21 omitted variable bias may be present to a limited extent.”99 Similarly, an ACT study of six-year 22 23 95 Id. at 95. 96 Saul Geiser & Roger Studley, Univ. of Cal. Office of the Pres., UC and the SAT: Predictive 24 Validity and Differential Impact of the SAT I and SAT II at the University of California 9 (Oct. 29, 25 2001). 26 27 97 Kurlaender & Cohen, supra note 39, at 12–13. 98 Id. 99 Krista D. Mattern et al., College Board, How Useful Are Traditional Admission Measures in 28 Predicting Graduation Within Four Years? 6 (2013), available at 36 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 college completion expressly noted its exclusion of “sociodemographic factors,” stating that “[t]he 2 use in this paper of one or two predictors is a mathematical simplification.” 100 Unlike these 3 industry studies, independent studies conducted by UC researchers do take socioeconomic 4 characteristics into account, and their results are striking: across these measures, SAT and ACT 5 scores add minimal or no incremental predictive value over high school GPA, which is 6 consistently a superior predictor of long-term college outcomes. These findings have been 7 corroborated by respected researchers studying universities across the nation. 101 78. 8 With respect to second-year retention, UC researchers have found that, after 9 controlling for socioeconomic characteristics, combining SAT scores with high school GPA 10 produced only a miniscule increase in predictive value. 102 In fact, even before controlling for 11 socioeconomic characteristics, SAT scores added very little additional predictive power when 12 combined with high school GPA.103 79. 13 With respect to long-term college outcomes—cumulative GPA and college 14 completion—UC researchers have found that high school GPA “is the best single predictor.” 104 In 15 predicting cumulative four-year GPA, SAT I scores have “considerably less predictive weight 16 [than high school GPA] after controlling for student background characteristics.” 105 The 17 18 https://files.eric.ed.gov/fulltext/ED562684.pdf. 19 100 25 103 27 104 Geiser & Santelices, supra note 65, at 9, 21. 105 Id. at 10. Justine Radunzel & Julie Noble, ACT, Predicting Long-Term College Success Through Degree Completion Using ACT Composite Score, ACT Benchmarks, and High School Grade Point 20 Average 11 (Aug. 2012), available at 21 http://www.act.org/content/dam/act/unsecured/documents/ACT_RR2012-5.pdf. 101 See, e.g., William G. Bowen at al., Crossing the Finish Line: Completing College at America’s 22 Public Universities 113–16 (2009). 23 102 Kurlaender & Cohen, supra note 39, at 13, Panel B, Column 4 (showing that, after controlling for socioeconomic characteristics and other factors, the difference between the adjusted correlation 24 coefficients described infra n.103 drops to .02). Id. at 13, Panel B, Column 1 (showing that the adjusted correlation coefficient of high school GPA and second-year retention is .18, whereas the adjusted correlation coefficient of high school 26 GPA and SAT scores combined and second-year retention is .22). 28 37 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 superiority of high school GPA over SAT I scores for predicting cumulative four-year GPA held 2 across all freshman cohorts, UC campuses, and academic fields. 106 As to four-year graduation, 3 SAT I scores “produce[d] no incremental improvement in prediction” over high school GPA and 4 SAT II scores combined.107 80. 5 Figure 3 illustrates the superiority of high school GPA over SAT scores as a 6 predictor of five-year graduation from UC. 7 8 Figure 3: Relative Weight of High School GPA and SAT Scores, Before and After Controlling for 9 SES, in Predicting 5-Year Graduation: All UC Freshmen vs. Underrepresented Minorities, 1994 to 2005108 10 11 12 13 14 15 16 17 18 19 20 21 22 23 81. 24 As shown in Figure 3, high school GPA is consistently a better predictor of five- 25 year graduation than SAT scores. The predictive value of SAT scores plummets even further when 26 27 28 106 Id. at 10–12. 107 Id. at 22. 108 Geiser, supra note 32, at 14. 38 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 controlling for race and socioeconomic status. Figure 3. When family income and parental 2 education are taken into account (“All Students + SES”), SAT scores lose about one-third of their 3 predictive value. Figure 3. By contrast, the predictive value of high school GPA actually increases 4 when controlling for the same socioeconomic characteristics. Figure 3. 82. 5 This predictive disparity is magnified among underrepresented minority students. 6 The predictive value of SAT scores declines precipitously when underrepresented minorities are 7 considered alone (“URMs, No SES”), particularly after controlling for socioeconomic 8 characteristics (“URMs + SES”). Figure 3. Thus, “[w]hile SAT scores remain a ‘statistically 9 significant’ predictor of UC graduation for students of color, the effect size is very small 10 indeed.”109 83. 11 These findings about UC data accord with findings that considered data from 12 across the country’s public higher education systems. After examining graduation rates from 21 13 public flagship universities and four state higher education systems, researchers found that “[h]igh 14 school grades are a far better predictor of both four-year and six-year graduation rates than are 15 SAT/ACT test scores,” which “routinely fail to pass standard tests of statistical significance when 16 included with high school GPA in regressions predicting graduation rates.” 110 17 Limited Value, Discriminatory Effects 18 84. Empirical evidence thus demonstrates the limited value of SAT and ACT scores in 19 predicting college performance. The independent research studies discussed supra show that, with 20 respect to the outcomes most commonly touted by the College Board—first-year GPA and 21 second-year retention—SAT and ACT scores add very little additional predictive power over high 22 school GPA once socioeconomic characteristics are taken into account. 85. 23 More importantly, the studies make plain that with respect to the college outcomes 24 that matter most—cumulative GPA and college completion—high school GPA is simply a better 25 predictor than SAT and ACT scores. As discussed infra paras. 123–24, the already limited 26 27 28 109 Id. 110 Bowen at al., supra note 101, at 113, 115. 39 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 predictive value of the SAT and ACT is further undermined by the fact that many affluent students 2 utilize expensive test preparation services to maximize their scores. Given SAT and ACT scores’ 3 adverse impact on students with less accumulated advantage, there can be no justification for UC’s 4 continued use of a metric that is both more discriminatory and worse at predicting long-term 5 college outcomes than high school GPA. 6 II. Bias in the Tests and Testing Conditions 7 A. Bias Against Underrepresented Minority Students 8 86. The limited predictive value of the SAT and ACT tests—particularly for 9 underrepresented minority students—arises in part from biases built into the development of the 10 tests themselves. 11 Norm-Referenced Tests and Reliability 12 87. The SAT and ACT are norm-referenced tests: they measure a test-taker’s 13 performance against the performance of other test-takers, instead of against a fixed standard. 111 14 Norm-referenced tests are designed to produce a bell curve, wherein most test-takers score in the 15 middle of the curve, and fewer test-takers score at the high and low tails. 112 By forcing student 16 scores into a bell curve, norm-referenced tests reflect the unscientific assumption that intelligence 17 in the population is distributed in that pattern. 113 Biases against underrepresented minority students 18 are inherent in the design of norm-referenced tests.114 88. 19 Designers of norm-referenced tests create items that must meet two requirements: 20 first, each item must be “reliable,” i.e., “internally consistent with the other items on the same 21 22 111 24 112 Geiser, supra note 31, at 11; see College Board, SAT Suite of Assessments Annual Report: Total Group (2019), available at https://reports.collegeboard.org/pdf/2019-total-group-sat-suite23 assessments-annual-report.pdf. Geiser, supra note 31, at 11. 113 See James J. Heckman, Lessons from the Bell Curve, 103 J. Pol. Econ. 1091, 1102 (1995) (contesting the “empirically incorrect claim that a single factor—g or IQ—that explains linear 26 correlations among test scores is primarily responsible for differences in individual performance in society at large”). 27 114 William C. Kidder & Jay Rosner, How the SAT Creates Built-in-Headwinds: An Educational 28 and Legal Analysis of Disparate Impact, 43 Santa Clara L. Rev. 131, 156–59 (2002). 25 40 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 test.”115 Second, each item must conform to requirements regarding its level of difficulty, so that 2 the test as a whole can sort students into a bell curve pattern. 116 Through a process known as “pre3 testing,” items are reviewed for reliability before being added to the SAT or ACT. 117 An item’s 4 reliability is gauged “by the internal correlation between performance on that item” and 5 performance on the test overall. 118 Put simply, an item is “reliable” if higher-scoring test-takers 6 tend to answer it correctly, and lower-scoring test-takers tend to answer it incorrectly. 119 Its 7 reliability is not tied to any fixed standard outside of the test itself. 120 89. 8 If White students have higher overall scores than underrepresented minority 9 students on a given set of pre-test items, a facially neutral reliability analysis will tend to find 10 strong positive correlations for items on which White students perform better, as opposed to 11 “weakly positive or even negative correlations” for items on which underrepresented minority 12 students perform better.121 Items that favor White students will thus be deemed reliable—resulting 13 in their inclusion on the test—whereas those that favor underrepresented minority students will be 14 deemed unreliable—resulting in their exclusion. 122 90. 15 Over multiple administrations of the test, this adverse effect on underrepresented 16 minority test-takers compounds itself: items that are strongly correlated to items from previous 17 administrations are more likely to remain on the test, whereas more weakly correlated items may 18 be dropped or revised.123 Because underrepresented minority test-takers have historically scored 19 20 115 Id. at 156. 21 116 Id. 117 Id. at 156–57. 118 Geiser, supra note 31, at 10. 119 Kidder & Rosner, supra note 114, at 157. 24 120 See Geiser, supra note 31, at 10–11. 25 121 Kidder & Rosner, supra note 114, at 158. 26 122 Id. 22 23 123 Id. at 158–59; Martin Shapiro, A Psychometric Model for Preserving Discrimination, 12 Berkeley La Raza L.J. 387, 390–91 (2001) (“Pre-test items which the higher-scoring group 28 answers correctly and the lower-scoring group answers incorrectly are considered to be good items 27 41 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 lower on the SAT and ACT tests for reasons including structural biases in test design and 2 stereotype threat, discussed infra, items on which those test-takers perform well are iteratively 3 discarded in order to reproduce score distributions on subsequent tests. 124 By contrast, facially 4 neutral questions that in fact favor White test-takers tend to be retained due to their higher 5 correlations with previously administered items. 125 Reliability analyses during test development 6 thus create an “ongoing feedback loop” that systematically disadvantages underrepresented 7 minority test-takers.126 8 Differential Item Functioning 9 91. Test developers attempt to detect bias in the SAT and ACT by searching for pre- 10 test items that exhibit differential item functioning (DIF). 127 DIF is a measure of bias that “occurs 11 ‘when equally able test takers [as measured by their overall scores] differ in their probabilities of 12 answering a test item correctly as a function of group membership.’” 128 92. 13 Critically, however, DIF controls for overall test scores before it assesses item 14 bias.129 That is, because it compares test-takers with the same overall scores, DIF necessarily 15 begins from the assumption that the test overall is unbiased. 130 Since DIF assumes that the test as a 16 whole is unbiased, it cannot be used to detect and eliminate systemic biases that affect the test 17 overall.131 Instead, DIF identifies only major outliers—items that are “grossly unfair to one or 18 19 and are likely to be included . . . on a subsequent test form. The same process is repeated on each subsequent test form. . . . [T]he selection process for pre-tested items continues to discard items 20 which would reduce the test-score difference between the two groups.”). 21 22 124 Brief for Nat’l Center for Fair & Open Testing (FairTest) as Amicus Curiae Supporting Respondents at 18, 539 U.S. 306 (2003) (No. 02-241) [hereinafter FairTest]. 125 See id. 23 126 Kidder & Rosner, supra note 114, at 158–59. 24 127 Shapiro, supra note 123, at 391. 25 128 Geiser, supra note 31, at 11 (citing Am. Research Ass’n et al., Standards for Educational and Psychological Testing 51 (2014)). 26 129 Kidder & Rosner, supra note 114, at 163. 27 130 Id. 131 Id. 28 42 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 another group.”132 93. 2 Test developers attempt to remove items that exhibit DIF from the tests that are 3 ultimately administered.133 But as UC psychometricians have demonstrated, these efforts have 4 failed to eliminate statistically significant item bias from the tests. Analyzing multiple SAT forms 5 from different years, the researchers found that up to 12 percent of items exhibited large DIF for 6 Black as compared to White test-takers, and up to 10 percent of items exhibited large DIF for 7 Latinx as compared to White test-takers. 134 Thus, even beyond the systemic biases that DIF fails to 8 detect, a substantial number of questions on the tests remains “grossly unfair” to underrepresented 9 minority test-takers. 10 Multilingual Learners 11 94. Multilingual learners face additional barriers to success on the SAT and ACT. The 12 UC requires all applicants to submit scores from the otherwise optional writing sections of the 13 SAT and ACT.135 The SAT’s extensive use of word-heavy math problems also places multilingual 14 learners at a disadvantage relative to their native English-speaking peers. For test-takers whose 15 first language is not English, lengthy word problems can slow their pace (which is particularly 16 problematic on “speeded” exams like the SAT and ACT) or even limit their comprehension of the 17 underlying math questions.136 95. 18 These disparities have only worsened with the latest iteration of the SAT, which 19 focuses on assessing test-takers’ ability to apply their math skills to “real-world contexts”— 20 “problems drawn from science, social studies, and careers.” 137 In practice, this emphasis on 21 22 23 24 132 FairTest, supra note 124, at 19. 133 Geiser, supra note 31, at 11. 134 Santelices & Wilson, supra note 10, at 23. 135 Univ. of Cal. Admissions, Exam Requirement, 25 https://admission.universityofcalifornia.edu/admission-requirements/freshmanrequirements/exam-requirement/. 26 136 James S. Murphy, New SAT, New Problems, The Atlantic (Jan. 20, 2015), available at https:// 27 www.theatlantic.com/education/archive/2015/01/new-sat-new-problems/384596/. 28 137 College Board, The Redesigned SAT: Problems Grounded in Real-World Contexts 1 (2014), 43 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 contextualized math items results in test-takers having to read and understand more text—placing 2 multilingual learners at a significant disadvantage. 138 96. 3 Even before the redesigned SAT became operational, the College Board was well 4 aware of these problems. In 2014, it administered a prototype of the new SAT and found that 5 approximately half of the test-takers could not complete the math sections of the test. 139 Moreover, 6 revisions to the math section disproportionately affected test-takers at lower-scoring schools (i.e., 7 schools where students posted lower overall test scores on the old SAT): on the calculator portion 8 of the test, seventy-eight percent of students from higher-scoring schools were able to finish, as 9 compared to only 41 percent of students from lower-scoring schools. 140 97. 10 Faced with these disparities, test developers decided to reduce the proportion of 11 “heavy” word problems (defined as problems exceeding 60 words) to only 10 percent of items on 12 the redesigned SAT.141 But this reduction never happened: in 2015, a College Board executive 13 reported that “heavy” word problems comprised 45 percent of the contextualized math items on 14 redesigned SAT practice tests.142 After the redesigned SAT became operational, independent 15 analysis of six practice tests—including two previously administered SAT tests—demonstrated 16 that “heavy” word problems continued to constitute at least 45 percent of the contextualized math 17 items on each test.143 98. 18 These disparities are exacerbated by the difficulties multilingual learners encounter 19 20 available at https://www.collegeboard.org/sites/default/files/problems_grounded_in_real_world_contexts.pdf. 21 138 Murphy, supra note 136. 22 139 Renee Dudley, Despite warnings, College Board redesigned SAT in way that may hurt neediest students, Reuters (Sept. 21, 2016), available at https://www.reuters.com/investigates/special23 report/college-sat-redesign/ (“Only 47 percent of all students finished one part of the math section 24 – the 20-question portion during which calculators aren’t allowed. And just 50 percent completed the longer 38-question math section, which allows the use of a calculator.”). 25 140 Id. 26 141 Id. 27 28 142 Id. 143 Id. 44 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 in trying to access English learner (“EL”) accommodations for the SAT and ACT. EL supports are 2 not available at all to students who take regular weekend SAT administrations, SAT Subject Tests, 3 or the PSAT/NMSQT.144 Moreover, because EL supports on the SAT and ACT are approved by 4 testing agencies rather than school districts or high schools, students who have accommodations at 5 school may not be able to use those accommodations on the SAT or ACT. Like accommodations 6 for students with disabilities, EL supports must be requested by a high school counselor, and many 7 multilingual learners at underserved, highly segregated schools have limited access to counselors 8 who can request EL supports on their behalf. During the 2016 to 2017 academic year, the ratio of 9 students to counselors in California was 663 to one.145 Monolingual parents seeking EL supports 10 for their children also lack adequate information in their native languages about the 11 accommodations process. Critically, because English language proficiency may improve over 12 time, EL supports are temporary (i.e., limited to the test date for which they were requested and 13 granted).146 The temporary nature of EL supports undermines the assumption that test scores 14 obtained by students receiving these supports will accurately predict future academic performance. 15 Stereotype Threat 16 99. Beyond biases in the content of the SAT and ACT, underrepresented minority 17 students face the additional barrier of stereotype threat: “the pressure that [students] feel when 18 they fear that their performance could confirm a negative stereotype about their group,” which 19 “manifests itself in anxiety and distraction that interferes with intellectual functioning.” 147 By 20 “artificially depress[ing]” the scores of underrepresented minority students on tests like the SAT 21 22 144 See College Board, Testing Supports for English Learners, 23 https://collegereadiness.collegeboard.org/educators/k-12/english-learner-supports (EL supports are available only “during SAT School Day, the PSAT 10, and the PSAT 8/9”). 24 145 Am. Sch. Counselor Ass’n, Student-to-School-Counselor Ratio 2016-2017, 25 https://www.schoolcounselor.org/ asca/media/asca/home/Ratios16-17.pdf. 26 146 College Board, supra note 144. 27 147 Jerry Kang et al., Brief of Experimental Psychologists as Amici Curiae in Support of 28 Respondents at 4, Fisher v. Univ. of Texas at Austin, 136 S. Ct. 2198 (2016) (No. 14-981). 45 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 and ACT, stereotype threat results in those tests underestimating how well underrepresented 2 minority students are likely to perform in college.148 100. 3 Stereotype threat is a well-documented psychological phenomenon. 149 Multiple 4 studies have confirmed its existence in real world educational and testing environments. 150 One 5 meta-analysis combining data from more than 3,000 participants found that standard measures of 6 academic performance “underestimate the capacity of students targeted by negative stereotypes by 7 an average of 0.18 standard deviations”—or 63 points on the SAT. 151 101. 8 Although all students can potentially experience stereotype threat, 152 the 9 phenomenon systematically disadvantages underrepresented minority students, who belong to 10 groups that historically have been stereotyped as intellectually inferior. 153 Perversely, the more a 11 student cares about his or her performance on a test, the more likely that student is to experience 12 stereotype threat-induced anxiety.154 Consequently, “stereotype threat hits the most dedicated 13 students the hardest.”155 14 Trauma and Other Adverse Life Experiences 15 102. Research shows that students’ exposure to trauma and other adverse life 16 experiences—including poverty and discrimination—also depresses their performance on 17 18 148 19 Claude M. Steele, Expert Report of Claude M. Steele, 5 Mich. J. Race & L. 439, 440 (1999). 149 Toni Schmader et al., An Integrated Process Model of Stereotype Threat Effects on 20 Performance, 115 Psychol. Rev. 336, 336 (2008) (“[A] large body of work now testifies to the reliability and generalizability of stereotype threat effects on performance.”). 21 150 Kang et al., supra note 147, at 15–18. 22 151 Id. at 18 (citing Greg Walton & Steven Spencer, Latent Ability: Grades and Test Scores Systematically Underestimate the Intellectual Ability of Negatively Stereotyped Students, 20 23 Psychol. Sci. 1132, 1135 (2009)). 24 152 See Tough, supra note 33, at 27 (describing how an SAT tutor for wealthy high school students 25 seeks primarily to “reduce his students’ anxiety” by dispelling their beliefs that the SAT and ACT are “objective, reliable measures” of their academic ability). 26 153 Kang et al., supra note 147, at 8. 27 28 154 Id. at 15. 155 Id. 46 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 standardized tests.156 Compared to their White and more affluent counterparts, underrepresented 2 minority students and students with less wealth are disproportionately exposed to (and forced to 3 cope with) trauma and adverse life experiences. Researchers’ analysis of data on 84,837 children’s 4 exposure to adversity—including “financial hardship,” “parental divorce/separation,” “parental 5 death,” “parental imprisonment,” “witness to domestic violence,” “victim or witness of 6 neighborhood violence,” “lived with mentally ill/suicidal person,” “lived with someone with 7 alcohol/drug problem,” and “treated unfairly because of race/ethnicity”—revealed striking racial 8 and socioeconomic disparities.157 The researchers found that Black, Latinx, and poorer children 9 had greater rates of exposure to adverse life experiences than White children and wealthier 10 children, noting that racial disparities were “consistent with extensive evidence of racial/ethnic 11 disadvantages across all major sectors of society,” including education and employment. 158 12 Studies have also shown that students with lower socioeconomic status experience significantly 13 higher rates of traumatic stress than their more affluent counterparts, suggesting higher rates of 14 exposure to trauma or fewer resources to mitigate its effects. 159 103. 15 Exposure to trauma and adverse life experiences has been shown to hinder 16 working memory—the temporary storage and manipulation of information—in children, 160 which 17 limits the ability of students exposed to trauma to prepare for and succeed on standardized tests. 18 Analyzing data from the Early Childhood Longitudinal Study, researchers found that students with 19 traumatic stress had significantly lower standardized test scores than students without such 20 21 22 156 Rachael D. Goodman et al., Traumatic Stress, Socioeconomic Status, and Academic Achievement Among Primary School Students, 4 Psychol. Trauma: Theory, Res., Prac., & Pol’y 23 252, 256 (2012). 24 157 Natalie Slopen et al., Racial Disparities in Child Adversity in the U.S.: Interactions with 25 Family Immigration History and Income, 50 Am. J. Preventive Med. 47, 48–49 (2016). 158 Id. at 52. 26 159 Goodman et al., supra note 156, at 256. 27 160 Anne P. DePrince et al., Executive Function Performance and Trauma Exposure in a 28 Community Sample of Children, 33 Child Abuse & Neglect 353, 359–60 (2009). 47 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 stress.161 The researchers concluded that “traumatic stress impairs student performance on 2 standardized tests,” placing students with exposure to trauma at risk of being incorrectly tracked 3 into “low-ability programs.”162 The researchers’ caution to educators—that “the 4 neuropsychological impact of trauma must be differentiated from actual cognitive ability” 163— 5 arises from the recognition that exposure to trauma distorts student performance on standardized 6 tests, reducing their predictive value still further. 7 B. Discrimination Against Students with Disabilities 8 Inaccessible Test Sites 9 104. Neither the College Board nor the ACT mandates that test sites be accessible to all 10 students with disabilities who require accommodations to access the SAT or ACT. Disability 11 accommodations vary depending on student need, and may include format changes such as Braille 12 or large-type test books, MP3 audio, readers, extended time, and additional or extended breaks 13 (such as for a student requiring insulin). College Board and ACT test sites—typically high 14 schools—may not provide all necessary accommodations for test-takers for all test 15 administrations. Students with disabilities are thus unable to choose their preferred test location, 16 and may not be guaranteed the ability to test on their desired dates, if at all. 17 ACT Accessibility 18 105. A student without disabilities who does not require accommodations to take the 19 ACT is able to choose his or her preferred test site and date, and may thus select the site that is 20 most convenient for him or her. Generally, a student testing without accommodations will choose 21 to take the ACT at his or her local high school. However, a student requesting disability 22 accommodations on the ACT cannot choose a test location during registration, because not all test 23 sites offer accommodations or have sufficient space and proctors for accommodated test-takers. 24 Students with disabilities who cannot test at their chosen location may have to travel out of the 25 26 161 Goodman et al., supra note 156, at 253, 255. 27 162 Id. at 256–57. 163 Id. at 257. 28 48 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 area to test, which can result in hardship—including additional stress and expenses—for test2 takers and their families. 106. 3 ACT divides its disability accommodations into “National” and “Special” 4 categories.164 “National” accommodations can be provided at standard or “national” test centers 5 with students who do not have accommodations. National accommodations are limited and 6 generally include 50 percent extended time, preferential seating, and EL supports. ACT’s policies 7 state that the “test location will be provided on the examinee’s admission ticket” for students who 8 are approved for national accommodations. These examinees may not learn their test location 9 under shortly before their scheduled test administration. 107. 10 ACT “special” accommodations are for documented disabilities requiring 11 accommodations that “cannot be provided in a National test center.” 165 Test-takers requiring 12 special accommodations must have the ACT administered by their schools. Despite mandating 13 that students with special accommodations take the ACT at their schools, ACT does not require 14 those schools to administer the test with special accommodations. 166 Students with disabilities may 15 thus be left without a location at which to test on their desired date. Students with sufficient 16 resources may drive long distances to test locations willing to accept their necessary 17 accommodations.167 For example, a student whose parent is a member of College Seekers received 18 notice from ACT informing the student that “you have been approved for Special Testing with 19 Accommodations, but . . . your home school is unable to administer the test during your scheduled 20 21 164 ACT, Accommodations and English Learner Supports for the ACT test, 22 https://www.act.org/content/act/en/ products-and-services/the-act/registration/accommodations.html (stating that “[b]ased on your 23 approved accommodations ACT will place you in either National Testing” or “Special Testing”). 24 165 Id. 166 ACT, Accommodations and English Learners Supports for Educators, 25 https://www.act.org/content/ 26 act/en/products-and-services/the-act-educator/accommodations.html (“If the school is not able to test the examinee, . . . [t]he school should notify the examinee that Special testing cannot be 27 administered and the examinee will need to work with ACT for an alternate location.”). 28 167 ACT, supra note 164 (“If the examinee’s school is unable to administer Special testing examinees may find a willing school and work with ACT to request testing at this location.”). 49 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 testing window.” He was told that “[o]nce we receive your information, we will attempt to arrange 2 testing at a school within 75 miles of your home. We will notify you via email as soon as we’ve 3 found a location.” However, ACT made clear that there was no guarantee that the student would 4 be able to test at all, providing that “[w]e cannot guarantee we will be able to arrange a location or 5 that the location we arrange will be able to test at the time of your choice.” These practices deter 6 students with disabilities from exercising their rights to request necessary accommodations, from 7 testing with their approved accommodations, and from retesting (which is already cost-prohibitive 8 for many test-takers). 9 108. ACT offers students the opportunity to test online to “better align with 10 how today’s students learn and the comfort many of them feel taking tests.” 168 According to ACT, 11 the benefits for students of online testing include students’ ability to choose their “most 12 comfortable” test format, to test “without distraction,” and to receive their scores more quickly (as 13 soon as two business days after taking the test). 169 However, ACT’s online testing is not 14 accessible to all test-takers with disabilities, because it does not offer the same accommodations as 15 are available for paper testing. Students whose accommodations are unavailable via online testing 16 must take the ACT in paper format. ACT states that eligible students whose accommodations are 17 not available via online testing may take the ACT via paper testing. 170 Certain members of College 18 Seekers cannot access ACT’s online testing due to their disabilities and required accommodations. 19 SAT Accessibility 20 109. Whereas all test-takers without disabilities can register to take the SAT at National 21 test centers, students who require certain testing accommodations due to their disabilities (such as 22 23 168 ACT, NEW FIRSTS FOR STUDENTS TAKING THE ACT TEST: FAQs, 24 https://www.act.org/content/act/en/ products-and-services/more-choices-for-the-act-sept-2020/faqs.html. 25 169 Id. 26 170 Id. (“Eligible students who require any amount of extended time, including 50% extended time 27 in a single session or other accommodations not available via online testing, may take the ACT via paper testing.”). 28 50 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 more than 50 percent extended time, breaks as needed, the use of a computer for the SAT essay, or 2 certain nonstandard materials such as Braille or assistive technology) cannot. These students must 3 instead participate in “school-based” testing to access the SAT. 4 110. Only College Board-approved test centers can administer school-based SAT testing 5 to students with disabilities who require accommodations. However, many California high school 6 students attend schools—such as public charter schools, independent study schools, and small 7 private schools—that are not College Board-approved test centers. Many students with disabilities 8 thus cannot take the SAT with accommodations at their own schools. These students may register 9 and pay for the SAT and obtain approval for necessary accommodations, only to subsequently 10 receive notice that they do not have a test date or location. 11 111. If a student cannot take the SAT with accommodations at his or her home school, 12 the College Board provides the student with a list of schools that may be willing to test outside 13 students. But the College Board does not require these schools to accept outside students, and if no 14 school is willing to administer the test with accommodations, the student will miss that test 15 administration. One College Seeker member with disabilities was approved for school-based SAT 16 testing. However, because his high school was not a College Board-approved test center, the 17 student could not take the SAT with accommodations there. The student was unable to locate a 18 test center willing to provide school-based testing for two separate exam administrations. 19 Realizing that he would miss early college application deadlines, the student decided not to take 20 the SAT with school-based accommodations at all. 21 Failure to Approve Necessary Accommodations 22 112. The challenge of finding an accessible test site is only the final hurdle in the testing 23 accommodations process, and it is a step that many students with disabilities never even reach. 24 Students and their families face both administrative and structural (socioeconomic, racial, and 25 language) barriers to obtaining the accommodations they need to access the SAT and ACT exams. 26 As a result of these barriers, even though approximately 14 percent of California students qualify 27 28 51 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 for special education supports,171 only approximately four to five percent of students take the SAT 2 or ACT with accommodations.172 113. 3 Unlike the California State assessment tests, for which schools provide disability 4 accommodations and EL supports without requiring outside agency approval, the SAT and ACT 5 may be taken with accommodations only with approval from the College Board or ACT. For a 6 student’s scores to be reported to colleges, his or her accommodations must be approved by one of 7 the testing agencies, even if the student takes the test during the school day. ACT and the College 8 Board do not guarantee students the same accommodations and supports they receive in their 9 school or district plans. This leaves some students without needed accommodations that they are 10 accustomed to receiving in school. 114. 11 The SAT and ACT accommodations processes place high demands on schools’ 12 resources. Because the testing agencies advise students to work with their schools in requesting 13 testing accommodations, high school counselors must prepare and submit time-consuming 14 accommodations applications. As discussed supra para. 96, students at underresourced and highly 15 segregated schools have more limited access to their high school counselors, and are thus less 16 likely than their more privileged peers to receive the accommodations they need to access the SAT 17 and ACT. A California public high school attended by a College Seekers member recently 18 informed its students that due to the extreme burdens imposed by the SAT and ACT 19 20 21 22 23 24 25 26 171 Cal. Dep’t of Educ. DataQuest, 2018-19 Enrollment by Ethnicity, https://dq.cde.ca.gov/dataquest/dqcensus/ EnrEthLevels.aspx?cds=00&agglevel=State&year=2018-19 (reporting 6,186,278 total students and 725,412 students who receive services under the Individuals with Disabilities Education Act (IDEA) (found by filtering by “Students with Disabilities”) enrolled in California K-12 public schools). Based on these numbers, approximately 12 percent of California students receive services through IEPs under IDEA. Because this percentage excludes students who receive accommodations pursuant to 504 Plans, disability experts estimate that the percentage of California students who qualify for special education supports is closer to 14 percent. 172 Nick Anderson, Abuse of ‘Extended Time’ on SAT and ACT Outrages Learning Disability Community, Wash. Post (Mar. 29, 2019), https://www.washingtonpost.com/local/education/abuse27 of-extended-time-on-sat-and-act-outrages-learning-disability-community/2019/03/29/d58de3c628 4c1f-11e9-9663-00ac73f49662_story.html. 52 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 accommodations processes, “school counselors will no longer submit accommodations requests on 2 behalf of students. It is cumbersome and we can no longer handle the volume.” Parents of students 3 at underresourced public high schools may receive no information at all on the SAT and ACT 4 accommodations process: one mother of a student with A.D.H.D. and a reading disability only 5 learned of the opportunity to request ACT accommodations after transferring her son to a private 6 school, where staff informed her about the process.173 Her son was granted an accommodation to 7 take the ACT over multiple days, earning a score of 33.174 115. 8 Even those students who submit accommodations requests with the support of their 9 schools must meet strict deadlines and documentation requirements. The College Board requires 10 requests to be submitted seven weeks prior to the test date, and it requires an additional seven 11 weeks for an appeal. ACT also has strict application deadlines. Accommodations and EL supports 12 are not available for ACT standby testing. 116. 13 ACT and SAT accommodations require frequent and often expensive evaluations, 14 as well as early diagnosis. Whereas schools with fewer resources may not be able to complete 15 evaluations as often as the testing agencies require, students with more resources can use private 16 testing to obtain early diagnosis and to update their evaluations to meet the strict review standards. 17 For many disabilities, ACT requires testing to have been updated within three years, and the 18 College Board requires testing to have been updated within five years. For psychiatric conditions, 19 both ACT and the College Board require documentation updated within the past year. 175 High 20 21 173 Dana Goldstein & Jugal K. Patel, Need Extra Time on Tests? It Helps to Have Cash, N.Y. Times (July 30, 2019) 22 https://www.nytimes.com/2019/07/30/us/extra-time-504-sat-act.html. 23 174 Id. Among the graduating class of 2018, an ACT composite score of 33 would place the student in the 98th percentile. ACT, ACT Profile Report – National: Graduating Class 2018 13 (2018), 24 available at https://www.act.org/content/dam/act/unsecured/documents/cccr2018/P_99_999999_N_S_N00_A 25 CT-GCPR_National.pdf. 26 175 ACT, Requesting Accommodations, https://www.act.org/content/act/en/products-andservices/the-act/registration/accommodations/policy-for-accommodations27 documentation.html#types (“[D]ocumentation of a psychiatric disorder must be within the past 28 year.”); College Board, Psychiatric Disorders, https:// 53 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 school mental health resources are already strained, and mental health professionals cannot always 2 reassess students as frequently as the College Board and ACT require. Unlike their less 3 advantaged peers, affluent students can pay for private testing to meet College Board and ACT 4 requirements for accommodations applications. A private psychologist in New York, for example, 5 charges $6,000 per evaluation and reports that “70 percent of the patients he sees leave with a 6 diagnosis, . . . and testing agencies usually approve accommodations for them.” 176 Relatedly, 7 although the College Board and ACT consider a student’s history of school accommodations and 8 school-provided information in determining whether to grant SAT or ACT accommodations, 9 underserved students may not be diagnosed soon enough or evaluated frequently enough by their 10 public schools to show a lengthy history of school accommodations. One student whose parent 11 could not afford a new evaluation had her request for accommodations on the SAT denied after 12 she submitted a seven-year-old A.D.H.D. evaluation. 177 Students in underresourced and highly 13 segregated schools may be undiagnosed, diagnosed late, or misdiagnosed. 117. 14 Students with temporary conditions may also be ineligible for accommodations. 15 The College Board does not consider temporary medical or physical conditions to be disabilities 16 for purposes of the SAT.178 In most cases, students with temporary conditions must “reregister to 17 take the test at a date when they have healed.”179 118. 18 The “speeded” nature of the ACT and SAT further discriminates against disabled 19 students, because there is no evidence that the time limits are essential to the test, and because 20 21 22 accommodations.collegeboard.org/documentation-guidelines/psychiatric-disorders 23 (“documentation must . . . include . . . [a] current psychiatric update, completed within the past year” (emphasis omitted)). 24 176 Goldstein & Patel, supra note 173. 25 177 Id. 178 College Board, Temporary Medical Conditions, https://accommodations.collegeboard.org/temporary-conditions (“[T]emporary medical or 27 physical conditions . . . are not disabilities.”). 26 28 179 Id. 54 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 students with disabilities struggle to complete the tests. 180 Studies do not consistently show that 2 the time accommodations granted are sufficient for all test-takers, and, as discussed supra, the 3 accommodations process for obtaining extended time is onerous and not equally accessible. 119. 4 Even when the College Board and ACT do approve accommodations, many 5 students with disabilities are not properly served by those accommodations. Studies have 6 demonstrated that students with certain disabilities, such as hearing loss and dyslexia, are not 7 effectively accommodated in testing,181 and that their test scores do not predict their likelihood of 8 success in the same way as for their peers without disabilities. For example, researchers have 9 noted the lack of information about the “effects of accommodations for deaf and hard of hearing 10 students who have additional disabilities” and “the range of quality in the implementation of 11 [American Sign Language] accommodations.”182 Moreover, “despite the importance of matching 12 accommodations to individual student characteristics and needs, few studies disaggregated 13 findings beyond broad categories” such as deaf, hard of hearing, or hearing. 183 120. 14 Finally, students with disabilities who do receive and use the accommodations they 15 need are experiencing unprecedented levels of suspicion and discrimination. Problematically, test 16 scores are sometimes reported in ways that identify students as having disabilities or as having 17 tested with accommodations. Although students with disabilities have historically feared 18 discrimination based upon self-identification, students who use testing accommodations have been 19 increasingly stigmatized following the recent college admissions scandal, in which wealthy 20 21 180 See generally Ruth Colker, Test Validity: Faster Is Not Necessarily Better, 49 Seton Hall L. 22 Rev. 679 (2019). 181 See, e.g., Juliana M. Taymans et al., Learning to Achieve: A Review of the Research Literature on Serving Adults With Learning Disabilities 134 (2009), available at 24 https://lincs.ed.gov/publications/pdf/L2ALiteratureReview09.pdf (noting “evidence that, despite receiving extended time on standardized test administrations, the population with LD [learning 25 disabilities] continues to perform significantly below their peers academically”). 23 182 Stephanie Cawthorn & Rachel Leppo, Assessment Accommodations on Tests of Academic Achievement for Students Who Are Deaf or Hard of Hearing: A Qualitative Meta-analysis of the 27 Research Literature, 158 Am. Annals of the Deaf 363, 373 (2013). 26 28 183 Id. at 367. 55 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 parents of students without disabilities abused the testing accommodations process to give their 2 children a competitive advantage.184 3 Non-College Reportable Scores 4 121. Rather than honoring all special education supports provided by schools to students 5 with disabilities, the College Board and ACT have devised a system of “non-college reportable” 6 scores for students who take their tests during the school day with their ordinary school 7 accommodations, but who have not been approved for SAT or ACT accommodations. These 8 students—who have IEPs, 504 Plans, or access to language supports in school, but who do not 9 receive permission to use these accommodations on the ACT or SAT—receive “non-college 10 reportable scores,” which cannot be sent to colleges, scholarship agencies, or any other outside 11 agencies.185 This system of non-college reportable scores exists solely because testing providers 12 are unwilling to honor the necessary and legal accommodations and supports provided by schools 13 to their students during school-day testing. 14 Exclusion from Validity Studies 15 122. Students with disabilities who test with necessary accommodations are excluded 16 from the College Board’s current validity studies. A concordance study comparing the 2016 17 redesigned SAT to its predecessor omitted “SSD answer sheets”—those from test-takers with 18 accommodations approved by the College Board’s Services for Students with Disabilities (“SSD”) 19 office—from its analysis.186 Thus, although the College Board claims that its Technical Manual 20 21 22 23 24 25 26 27 28 184 Anderson, supra note 172. 185 ACT, The ACT Test Administration Manual: State Testing District Testing Non-College Reportable Accommodations and English Learner Supports 6 (2019), available at https://www.act.org/content/dam/ act/secured/documents/pdfs/ACT-Admin-Manual-SD-NCR-Accoms-Paper-Fall-Secured.pdf (“ACT will not report scores earned with non-college reportable accommodations and/or supports to colleges, scholarship agencies, or any other entities.”); see College Board, SAT School Day Accommodated Testing Manual 14 (2019), available at https://collegereadiness.collegeboard.org/pdf/sat-school-day-accommodated-testing-manualfall.pdf (“College Board requires preapproval for all accommodations” in order for accommodated students to “receive reportable scores”). 186 College Board, SAT Suite of Assessments Technical Manual 115–16 (Dec. 2017), available at 56 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 demonstrates that “validity considerations permeate every aspect of the SAT,” 187 this is plainly not 2 the case for students with disabilities who take the SAT with necessary accommodations. ACT 3 similarly excludes special-tested students from its National and State Annual Reports of College 4 and Career Readiness, which consequently furnish no support for the validity of the ACT with 5 respect to these students. 6 III. Unequal Access to Test Preparation 7 123. The enormous stakes of the SAT and ACT tests have spawned a lucrative test 8 preparation industry which capitalizes on the desires of students and their parents to secure even 9 incremental score increases.188 Unequal access to test preparation services exacerbates the 10 discriminatory nature of the SAT and ACT. 124. 11 Affluent students have superior access to high-quality test preparation programs, 12 which can cost “upwards of $10,000” over the course of multiple sessions. 189 One $400-per-hour 13 SAT and ACT tutor teaches his students “clever test-taking tricks,” like “plugging a simple 14 number” into algebra equations rather than spending time working through the math. 190 Learning 15 these strategies not only boosts his tutees’ scores, but also reduces their anxiety about the test, 16 which becomes less daunting once they realize that simple “tricks actually work.” 191 125. 17 The College Board has historically insisted that the only way for students to 18 increase their SAT scores is by learning content. 192 However, the College Board publishes and 19 sells its own test preparation book, the “Official SAT Study Guide,” which it touts as “[t]he best 20 21 https://collegereadiness.collegeboard.org/pdf/sat-suite-assessments-technical-manual.pdf. 187 Id. at vi. 22 188 27 191 Id. at 28–29. 192 Id. at 74. See James Wellemeyer, Wealthy parents spend up to $10,000 on SAT prep for their kids, MarketWatch (July 7, 2019), https://www.marketwatch.com/story/some-wealthy-parents-are23 dropping-up-to-10000-on-sat-test-prep-for-their-kids-2019-06-21 (“A March 2019 report from 24 IBISWorld valued the tutoring and test preparation industry at $1.1 billion, with exam prep services making up 25% of the industry.”). 25 189 Id. 26 190 Tough, supra note 33, at 28. 28 57 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 source of information about the SAT.”193 The Official SAT Study Guide promises students that its 2 “information, advice, and sample questions will help you prepare to take the test with 3 confidence.”194 According to the College Board, “[l]earning about the SAT through this guide 4 . . . will help you be well prepared when your test date arrives.” 195 126. 5 In 2015, in the wake of criticisms regarding the SAT’s lack of predictive value, the 6 College Board dramatically shifted its stance on test preparation by contracting with Khan 7 Academy, a provider of online test preparation services. 196 Although Khan Academy test 8 preparation is free, its content—particularly the test-taking “strategies” it teaches students—is 9 limited by its affiliation with the College Board. Because the College Board continues to assert 10 that learning content is the best way to succeed on the SAT, Khan Academy does not teach 11 students strategies for “gaming” the test. For example, whereas private tutors instruct students on 12 how to quickly and correctly solve math items without working through every step of the problem, 13 supra para. 122,197 Khan Academy’s math strategies section advises students that their best 14 strategy is to “stop rushing and stop cutting corners.”198 Similarly, Khan Academy cautions 15 against “mental math,” advising students to “[g]et out of your head” and “do your work on paper” 16 by, for example, “[w]rit[ing] down every step of algebra problems.” 199 Even with these 17 18 193 College Board, The Official SAT Study Guide (2018). 19 194 Id. at 3. 20 195 Id. 196 Jacqueline Thomsen, New Website for a New Test, Inside Higher Ed (June 2, 2015), https:// www.insidehighered.com/news/2015/06/02/college-board-and-khan-academy-team-ease-access22 new-sat (“The announcements are designed to help the College Board try to rebrand a test that some have criticized for not providing an accurate prediction of success in higher education.”). 23 197 Tough, supra note 33, at 28 (describing how a $400-per-hour private tutor instructs students to 24 “see if you can figure out the answer” to geometry questions “by just eyeballing the diagram, rather than doing the math”). 25 198 Khan Academy, SAT Math Test Strategies: Controlling Careless Errors on the SAT Math Test, 26 https:// www.khanacademy.org/test-prep/sat/new-sat-tips-planning/sat-math-test-strategies/a/controlling27 careless-errors-on-the-sat-math-test?modal=1. 21 28 199 Id. 58 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 limitations, Khan Academy is yet another tool for test preparation that is leveraged only with 2 access and assistance. It is thus used more by affluent students and students whose parents have 3 advanced degrees, furnishing those students with yet one more tool for preparation. 200 127. 4 Given these disparities in student access to test preparation, underresourced schools 5 and school districts divert valuable instruction time and large amounts from their limited funds 6 toward test preparation, at the expense of providing students with substantive, college preparatory 7 education. 8 IV. Use of SAT and ACT in UC Admissions 9 A. Mission and Mandate of the UC 128. The “distinctive mission” of the UC “is to serve society as a center of higher 10 11 learning, providing long-term societal benefits” through the transmission and discovery of 12 advanced knowledge.201 The UC thus has an obligation to students and to the people of the State 13 of California to educate students and to prepare them to contribute to the civic and economic life 14 of the State. By continuing to consider SAT and ACT scores in admissions, however, the UC 15 deprives the people of California of the contributions of talented young people who could bring 16 their intelligence, diverse backgrounds, and industry to serve the needs of the State. UC’s 17 continued use of the SAT and ACT thus harms both well-qualified students excluded from public 18 higher education and the people of the State by stymieing these students’ immense potential to 19 contribute to the State’s political and economic vitality. 129. 20 UC’s mission is constitutionally mandated: the California Constitution provides 21 that the UC shall be a “public trust” administered by the Regents, and envisions the UC as the 22 culmination of the State’s responsibility to provide public education to its citizens. 202 Within the 23 200 Tough, supra note 33, at 91 (describing Khan Academy and College Board findings that “[s]tudents in the top income quartile used the site for more time than students in the bottom 25 income quartile” and students whose parents attended graduate school used the site for more time than students whose parents had only a high school degree). 26 201 Univ. of Cal. Office of the Pres., UC’s Mission, https://www.ucop.edu/uc-mission/ (citing 27 Univ. of Cal. Academic Plan, 1974-1978) [hereinafter UC Mission]. 24 28 202 Cal. Const. art. IX, § 9(a); see id. § 1. 59 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 framework of the California Constitution, the UC exists to further the “general diffusion of 2 knowledge and intelligence,” which is “essential to the preservation of the rights and liberties of 3 the people,” and to promote California citizens’ “intellectual, scientific, [and] moral . . . 4 improvement.”203 This mission—providing higher education to California citizens—is so critical 5 to the population’s wellbeing that the California Constitution explicitly articulates that no person 6 shall be excluded from higher education at the UC on the basis of race or ethnicity. 204 Nor is the 7 constitutional vision of UC’s inclusivity and diversity limited to UC’s student population: the 8 California Constitution expressly provides that UC’s leadership shall be “broadly reflective of the 9 economic, cultural, and social diversity of the State, including ethnic minorities.” 205 130. 10 The UC frames this constitutional mandate as three “fundamental missions”: 11 “teaching, research[,] and public service.” 206 Undergraduate education is central to UC’s teaching 12 mission. According to the UC, “[u]ndergraduate programs are available to all eligible California 13 high-school graduates.”207 These programs furnish students with the knowledge and skills they 14 need not only to have successful careers, but also to continue growing intellectually and “to 15 contribute to the needs of a changing society.” 208 Taken in the aggregate, UC’s education of 16 talented California students “create[s] an educated workforce that keeps the California economy 17 competitive.”209 131. 18 UC’s Policy on Undergraduate Admissions unequivocally confirms UC’s 19 commitment to provide higher education to all talented California students, regardless of 20 background: 21 22 23 24 203 Id. § 1. 204 Id. § 9(f). 205 25 Id. § 9(d). 206 UC Mission, supra note 201. 26 207 Id. 27 208 Id. 209 Id. 28 60 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 Mindful of its mission as a public institution, the University of California has an 2 historic commitment to provide places within the University for all eligible 3 applicants who are residents of California. The University seeks to enroll, on each 4 of its campuses, a student body that, beyond meeting the University’s eligibility 5 requirements, demonstrates high academic achievement or exceptional personal 6 talent, and that encompasses the broad diversity of cultural, racial, geographic, and 7 socioeconomic backgrounds characteristic of California. 210 8 9 Thus, although the UC permits individual campuses to establish their own admissions procedures, 10 it mandates that those procedures align with UC’s commitment to building a student body that 11 reflects the “cultural, racial, geographic, and socioeconomic” diversity of the State. 211 132. 12 The UC has repeatedly reaffirmed this commitment to diversity, recognizing that 13 building a diverse and representative student population is integral to fulfilling its “core mission . . 14 . to serve the interests of the State of California.” 212 Noting that “[t]he diversity of the people of 15 California has been the source of innovative ideas and creative accomplishments throughout the 16 state’s history into the present,” the UC has recognized the myriad ways in which a diverse 17 educational community fosters its “achievement of excellence”: by “broaden[ing] and deepen[ing] 18 both the educational experience and the scholarly environment”; by “promot[ing] mutual respect” 19 and “preparing [students] to participate in an increasingly complex and pluralistic society”; by 20 giving rise to, nurturing, and enriching ideas and practices; and, most critically, by enabling “the 21 22 23 210 Univ. of Cal. Bd. of Regents, Regents Policy 2102: Policy on Undergraduate Admissions, 24 https:// regents.universityofcalifornia.edu/governance/policies/2102.html. 25 211 Id. (“[E]ach campus shall establish procedures for the selection of applicants to be admitted 26 from its pool of eligible candidates. Such procedures shall be consistent with the principles stated above and with other applicable University policies.”). 27 212 Univ. of Cal. Bd. of Regents, Regents Policy 4400: Policy on University of California Diversity 28 Statement, https://regents.universityofcalifornia.edu/governance/policies/4400.html. 61 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 full, effective use of the talents and abilities of all to foster innovation and train future 2 leadership.”213 133. 3 Beyond contributing to UC’s academic mission, UC’s assembling of a diverse 4 student population fulfills an even more fundamental purpose: it “sustain[s] the social fabric of the 5 State” with the “knowledge that the University of California is open to qualified students from all 6 groups, and thus serves all parts of the community equitably.” 214 Because the UC exists to serve 7 the people of the California, the “State of California has a compelling interest in making sure that 8 people from all backgrounds perceive that access to the University is possible for talented students 9 . . . from all groups.”215 These sentiments are simply irreconcilable with UC’s continued use of the 10 SAT and ACT in admissions. By basing admissions decisions on tests that discriminate against 11 underrepresented minority students, students with less wealth, and students with disabilities, the 12 UC has failed to provide a level playing field in which all applicants are assessed based on 13 individual merit, drive, and character. Admission to the UC, therefore, is plainly not “open to 14 qualified students from all groups,” rendering the UC utterly unable to “serve[] all parts of the 15 community equitably.”216 Unsurprisingly, UC’s persistent use of these tests has also undermined 16 public confidence in the fairness of UC admissions, as Californians correctly “perceive that access 17 to the University” is premised on access to wealth and other advantages. 217 18 B. UC’s Mandate and Fundamental Rights 19 134. Education is a fundamental right under the California Constitution, and UC’s 20 constitutional mandate—to prepare all qualified students to contribute meaningfully to the State’s 21 economy and society—is premised on that right.218 California courts have long recognized that K22 23 213 Id. 24 214 Id. 215 Id. 216 Id. 217 Id. 25 26 27 218 Butt v. State, 4 Cal. 4th 668, 692 (1992) (“[E]ducation is a fundamental interest in 28 California.”). 62 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 12 education constitutes a fundamental right precisely because it enables students to access higher 2 education and its attendant opportunities. Permitting equal access to higher education—including 3 “entrance into . . . the University of California”—is the ultimate purpose of public elementary and 4 secondary schooling, which opens “doorways . . . into chambers of science, art and the learned 5 professions, as well as into fields of industrial and commercial activities.” 219 135. 6 As the California Supreme Court has recognized, “the ‘fundamental’ nature of the 7 right to an equal education derives in large part from the crucial role that education plays in 8 ‘preserving an individual’s opportunity to compete successfully in the economic marketplace, 9 despite a disadvantaged background.’”220 The “public schools of this state”—and none more so 10 than the UC—facilitate the “entry” of students from underrepresented groups into the economic 11 and political spheres from which those groups have historically been excluded. 221 Thus, not only 12 does education “prepare[] individuals to participate in the institutional structures . . . that distribute 13 economic opportunities,”222 it also empowers students to engage meaningfully in civic life: 14 education “provide[s] the understanding of, and the interest in, public issues which are the spur to 15 involvement in other civic and political activities.” 223 The right to education is therefore 16 fundamental because it enables students of all backgrounds to access social mobility and become 17 successful participants in California’s economy and democracy. 136. 18 Because the principal value of education is as a vehicle of economic and social 19 mobility, the fundamental right to education mandates “more than access to a classroom.” 224 In 20 Butt v. State, the California Supreme Court made clear that the right requires public institutions 21 not just to provide instruction, but to do so in a way that facilitates students’ “high school 22 23 219 24 220 25 Piper v. Big Pine Sch. Dist. of Inyo Cty., 193 Cal. 664, 673 (1924). Crawford v. Bd. of Educ., 17 Cal. 3d 280, 295 (1976) (quoting Serrano v. Priest, 5 Cal. 3d 584, 609 (1971)). 221 Id. 26 222 Hartzell v. Connell, 35 Cal. 3d 899, 908 (1984). 27 223 Serrano, 5 Cal. 3d at 607. 224 Id. at 607. 28 63 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 graduation[] and college entrance.”225 According to the Court, the constitutional harm arising from 2 the early closure of a district’s schools was not the loss of instruction itself, but rather the 3 “extensive educational disruption” that could “seriously compromise[]” “the ability of departing 4 seniors to qualify for college admission.” 226 By jeopardizing students’ ability to access higher 5 education, the Court concluded, “the proposed closure would have a real and appreciable impact 6 on the affected students’ fundamental California right to basic educational equality.” 227 137. 7 The fundamental right to K-12 education thus means nothing if it does not also 8 encompass the right to access public higher education. A college degree has become a virtual 9 prerequisite for securing steady, gainful employment in today’s economy. Disparities in 10 employment rates and wages between individuals with bachelor’s degrees and those without are 11 persistent and well-documented. According to 2018 data from the Bureau of Labor Statistics, 12 those with bachelor’s degrees had median weekly earnings of $1,198, as compared to $862 for 13 those with associate degrees and $730 for those with only high school diplomas. 228 Those with 14 bachelor’s degrees also had lower rates of unemployment than those without. 229 A college degree 15 has become a common prerequisite even for jobs that have traditionally been available to high 16 school graduates: a 2017 Harvard Business School study reviewed more than 26 million job 17 postings and found that postings for many jobs that previously did not require a college degree 18 19 20 21 22 23 225 4 Cal. 4th at 687. 24 226 Id. & n.16. 25 227 Id. at 688. 228 U.S. Dep’t of Labor, Bureau of Labor Statistics, Education pays (Feb. 2019), https://www.bls.gov/careeroutlook/ 27 2019/data-on-display/education_pays.htm. 26 28 229 Id. 64 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 “now stipulate a college degree as a minimum education requirement.” 230 Employer demand for 2 workers with college degrees is on the rise “in the vast majority of occupations.” 231 138. 3 These trends are particularly pronounced in California, where a dynamic State 4 economy fuels a “[s]trong and growing demand for highly educated workers.” 232 According to 5 researchers at the Public Policy Institute of California, “[i]f current trends continue, about 40 6 percent of jobs in California will require at least a bachelor’s degree by 2030.” 233 In California, 7 individuals with college degrees have manifold advantages over individuals with only high school 8 diplomas: whereas a typical full-time worker with a college diploma earns $81,000, a typical full9 time worker with a high school diploma earns only $36,000. 234 A college degree produces striking 10 long-term dividends: Georgetown University researchers have found, for example, that economic 11 value of a UC Berkeley degree 40 years after enrollment is $1,383,000, and that of a UCLA 12 degree is $1,300,000.235 The “wage premium” associated with a college degree has grown steadily 13 over the last two decades.236 Beyond higher wages, California workers with college degrees have 14 more job security and non-wage benefits, lower rates of poverty and unemployment, and higher 15 rates of homeownership.237 College graduates “are more likely to have long-lasting marriages and 16 to be civically engaged,” and even “have lower mortality rates.” 238 These disparities make plain 17 230 Joseph B. Fuller & Manjari Raman, Dismissed by Degrees: How Degree Inflation Is Undermining U.S. Competitiveness and Hurting America’s Middle Class 2 (Oct. 2017), available 19 at https://www.hbs.edu/managing-the-future-of-work/Documents/dismissed-by-degrees.pdf. 18 231 Pub. Policy Inst. of Cal., Higher Education in California: Meeting California’s Workforce Needs (Oct. 2019), available at https://www.ppic.org/wp-content/uploads/higher-education-in21 california-october-2019.pdf. 20 22 23 232 Id. 233 Id. 234 Id. 24 235 27 237 Id. 238 Id. Anthony P. Carnevale et al., A First Try at ROI: Ranking 4,500 Colleges 43 (2019), available at https://1gyhoq479ufd3yna29x7ubjn-wpengine.netdna-ssl.com/wp25 content/uploads/College_ROI.pdf. 26 236 Pub. Policy Inst. of Cal., supra note 231. 28 65 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 that in the State of California, the fundamental right to education cannot end at grade 12. Rather, 2 the ability to access public higher education has become the “major determinant of an individual’s 3 chances for economic and social success in our competitive society.” 239 In today’s labor market, 4 the constitutional right to “equal access to a public education system that will teach [students] the 5 skills they need to succeed as productive members of modern society” necessarily means equal 6 access to public higher education.240 139. 7 Closely related to the fundamental right of access to public higher education is the 8 fundamental right to pursue an occupation, which California courts have repeatedly recognized. 241 9 It has long been established in California that “[a]ny unreasonable limitation that deprives 10 qualified persons of the equal opportunity to qualify for work is unconstitutional.” 242 UC’s 11 requirement that applicants submit SAT and ACT scores in order to be considered for admission is 12 one such “unreasonable limitation”: by hindering the ability of well-qualified, less advantaged 13 students to access high-quality public higher education at the UC, the requirement ultimately 14 deprives those students of the opportunity to pursue a wide range of occupations that is granted to 15 their more privileged peers. 16 C. UC’s Historical Use of the SAT and ACT 17 140. UC’s concerns about the validity of the SAT and ACT are longstanding and well- 18 founded. Concerns about the tests’ weak ability to predict student success prevented UC from 19 adopting standardized testing requirements “until later than most other selective US 20 universities.”243 As early as 1956, BOARS—the Academic Senate committee that formulates and 21 regulates UC’s undergraduate admissions policies, subject to ultimate approval by the Regents— 22 rejected a proposal that would have required SAT scores from entering students, based on 23 24 25 239 Serrano, 5 Cal. 3d at 605. 240 O’Connell v. Superior Court, 141 Cal. App. 4th 1452, 1482 (2006). 241 See, e.g., Hughes v. Bd. of Architectural Examiners, 17 Cal. 4th 763, 788 (1998); Terry v. Civil 26 Serv. Comm’n, 108 Cal. App. 2d 861, 870 (1952). 27 28 242 Terry, 108 Cal. App. at 870. 243 Geiser, supra note 31, at 6. 66 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 members’ “grave concerns regarding the validity of standardized tests in predicting the academic 2 performance of university students.” 244 141. 3 Following this initial rejection, the UC embarked on an experimental study to 4 assess the SAT’s effectiveness.245 For a period of two years, beginning with the class of 1960, the 5 UC required applicants to take the SAT or similar tests. 246 During this experimental period, SAT 6 scores were not used to evaluate applicants for admission to the UC, but were instead collected to 7 analyze the capacity of the SAT, as a potential admissions criterion, to predict students’ academic 8 performance.247 The results were unequivocal: BOARS’s chair reported that after “[e]xtensive 9 analysis of the data,” BOARS was “wholly convinced that [SAT] scores add little or nothing to the 10 precision with which existing admissions requirements are predictive of success in the 11 University.”248 BOARS thus rejected an SAT requirement by unanimous vote. 249 The 1960–62 12 study was the first of several commissioned by BOARS over the past six decades, all of which 13 have demonstrated the limited predictive value of standardized tests like the SAT and ACT in 14 predicting college performance. Despite these results, UC has persisted in using SAT and ACT 15 scores as an admissions criterion. 142. 16 In 1963, “[l]obbied by the Educational Testing Service,” 250 BOARS agreed to 17 undertake another study of standardized testing, which would examine the ability of achievement 18 tests (like SAT II subject tests) to predict student performance at the UC. 251 The study “echoed the 19 findings of [UC’s] previous analysis of the validity of the SAT,” determining that achievement 20 21 22 244 Douglass, supra note 5, at 85 & n.16. 23 245 Id. at 86. 246 Id. 24 247 25 Id. 248 Id. at 90 (quoting BOARS Chairman Charles Jones). 26 249 Id. 27 250 Geiser, supra note 31, at 7. 251 Douglass, supra note 5, at 91. 28 67 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 tests were only “of marginal value in predicting academic success.” 252 Notably, even this 2 “marginal value” outweighed the predictive value of the SAT. 253 The study found that high school 3 GPA remained the best predictor of college performance, explaining 22 percent of the variance in 4 first-year grades (as compared to the achievement tests, which explained only eight percent). 254 5 Although the study also determined that combining achievement test scores with high school GPA 6 “appeared to offer a marginal improvement” in predicting first-year grades, this “admittedly 7 slight” gain was not enough to persuade BOARS to adopt an achievement test requirement. 255 8 Without a stronger demonstration of the predictive value of the SAT and achievement tests, 9 BOARS refused to impose a testing requirement on applicants to UC. 256 143. 10 Despite these dismal results, the UC gradually expanded its use of the SAT and 11 ACT tests. In 1968, the UC required SAT and ACT scores from all applicants, but deployed the 12 scores only for the narrow purposes of assessing out-of-state applicants and “in-state students with 13 very low GPAs.”257 By 1979, in an attempt to “cull [its] eligibility pool,” the UC incorporated 14 SAT and ACT scores into an “eligibility index” that combined grades and test scores, effectively 15 “extend[ing] a minimum test-score requirement to most UC applicants.” 258 UC faculty expressed 16 multiple concerns about the index, questioning the SAT’s predictive value and its efficacy as a 17 “device[] for encouraging increased preparation” and warning of “disproportionate adverse 18 effects” on low-income and underrepresented minority students. 259 Despite these concerns, the 19 Regents approved the eligibility index by a close 15-to-12 vote. 260 20 21 252 Id. 22 253 Id. 23 254 Id. 255 Id. at 91–92. 24 256 25 Id. at 92. 257 Geiser, supra note 31, at 7. 26 258 Id. 27 259 Douglass, supra note 5, at 115 (quoting then-Santa Barbara division Chair Robert Michaelsen). 260 Id. at 117. 28 68 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 144. 1 In 1988, following UC’s adoption of a centralized application system and 2 subsequent increases in application volume, the UC adopted a new admissions policy at “heavily 3 impacted campuses.”261 The policy provided for the admission of “[t]he top 40% to 60% of the 4 freshman class . . . based solely on high-school grades and SAT/ACT scores.” 262 145. 5 The increasing weight placed on SAT and ACT scores in UC admissions during 6 this period did not reflect corresponding increases in the tests’ validity, as BOARS’s own studies 7 demonstrated.263 In 1997, BOARS commissioned a study that found that the combination of SAT 8 II math and writing scores outperformed SAT I scores in predicting first-year grades at the UC, 9 likely due to the fact that, unlike the SAT I, SAT II tests were “curriculum driven.” 264 Based on 10 this study, the UC revised its eligibility index, including SAT II scores and giving them twice as 11 much weight as SAT I or ACT scores. 265 146. 12 In 2001, BOARS commissioned a study that again found that SAT I scores 13 contributed “very little, if any, incremental power in predicting UC freshman grades” after taking 14 into account high school GPA and SAT II scores. 266 Moreover, the study found that much of the 15 correlation between SAT I scores and first-year grades was “conditioned by socioeconomic 16 factors,” such that after controlling for those factors, “SAT I scores add[ed] nothing to the 17 prediction of freshman grades beyond that which HSGPA and the SAT II already provide[d].” 267 18 Relying on this analysis, then-UC President Richard Atkinson called for the elimination of the 19 20 261 Geiser, supra note 31, at 7. 262 Id. 22 263 Id. 23 264 27 266 Geiser & Studley, supra note 96, at 4. 267 Id. at 9. 21 BOARS, The Use of Admissions Tests by the University of California 7–8 (Jan. 2002), available at https://senate.universityofcalifornia.edu/_files/committees/boars/admissionstests.pdf 24 (quoting Judy Kowarsky et al., Predicting University Grade-Point Average in a Class of 25 University of California Freshmen: An Assessment of the Validity of A-F GPA and Test Scores as Indicators of Future Academic Performance 6 (1998)). 26 265 Id. at 6–8. 28 69 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 SAT I requirement in UC admissions, and the use instead of curriculum-based achievement tests 2 that measure “what [students] have accomplished during four years of high school, taking into 3 account their opportunities.”268 Ceasing to consider scores on tests like the SAT I and ACT, 4 Atkinson recognized, would redress adverse impacts on students with less wealth and 5 underrepresented minority students and foster “greater public confidence in the fairness of [UC]’s 6 admissions process.”269 147. 7 In 2002, BOARS cited its 2001 study in a paper proposing a “new testing array” 8 that expressly excluded the SAT I.270 Instead, BOARS recommended a battery of tests consisting 9 of a yet to be developed “core achievement examination,” testing students’ mastery of language 10 arts and math, and two other subject examinations. 271 148. 11 Recognizing the threat posed by Atkinson’s and BOARS’s proposals, the College 12 Board soon announced that it would modify the SAT, starting in 2005, to include a writing 13 section, evaluate “reading for understanding,” and emphasize “mathematical problem solving” 14 through questions modeled on “real-life situations.”272 The College Board implied that the revised 15 SAT would be less coachable, stating that rigorous coursework was the best preparation for the 16 test.273 Through this “repackaging,” the College Board “appeared to respond to criticism, modified 17 the existing test at the margin,” and ultimately “preserved the market for the SAT I.” 274 149. 18 These changes—including the much-touted writing section—did not substantially 19 enhance the predictive power of the test. BOARS has continued to study the utility of the SAT and 20 ACT in UC admissions, and has continued to find SAT I scores largely redundant in evaluating 21 applicants. For example, a 2008 study commissioned by BOARS determined that SAT scores 22 23 24 268 Atkinson, supra note 15, at 144. 269 Id. at 139. 270 25 BOARS, supra note 264, at 16, 18. 271 Id. at 16–17. 26 272 Douglass, supra note 5, at 228. 27 273 Id. 274 Id. at 230. 28 70 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 accounted for only 1.9 percent of the variance in first-year GPA among UC Berkeley students, and 2 only 1.6 percent of the variance across the UC system’s undergraduate campuses. 275 3 D. UC’s Justification for Continued Use of SAT and ACT 4 150. Despite BOARS’s own studies demonstrating the large redundancy of SAT and 5 ACT scores in predicting college performance, the UC has continued to use those scores in 6 admissions decisions. In 2002, BOARS established a set of principles governing UC’s use of 7 “admissions tests.”276 The principles provide that UC will use admissions tests for four purposes: 8 “to assess academic preparation and achievement of UC applicants”; “to predict success at UC 9 beyond that predicted by high school GPA”; “to aid in establishing UC eligibility”; and “to aid in 10 selecting students for admission at individual UC campuses.” 277 The principles further enumerate 11 four “desired properties” of admissions tests.278 According to BOARS, an admissions test should: 12 “be a reliable measurement that provides uniform assessment and should be fair across 13 demographic groups”; “measure levels of mastery of content in UC-approved high school 14 preparatory coursework and . . . provide information to students, parents, and educators enabling 15 them to identify academic strengths and weaknesses”; “be demonstrably useful in predicting 16 student success at UC and provide information beyond that which is contained in other parts of the 17 application”; and “be useful in a way that justifies its social and monetary costs.” 279 BOARS 18 specifically “recognized that predictors of success [were] currently limited” to first-year GPA and 19 graduation rate, and called for the identification and use of “better predictors” to validate 20 admissions tests.280 Having established these principles, BOARS recommended that “the faculty 21 22 23 24 275 BOARS, supra note 6, at 95, 107. 276 25 BOARS, supra note 264. 277 Id. at 15. 26 278 Id. 27 279 Id. at 15–16. 280 Id. at 16. 28 71 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 regularly review UC’s admissions testing policy and practices to ensure that tests are being used in 2 a way that is consistent with these principles and desired properties of admissions tests.” 281 151. 3 BOARS’s testing principles have continued to govern UC’s use of the SAT and 4 ACT in admissions decisions. In 2002, BOARS applied these principles to UC testing 5 requirements, and ultimately recommended a testing array that did not include the SAT I. 282 6 Following the College Board’s announcement of revisions to the test, however, BOARS and UC 7 decided to wait to study the predictive validity of the 2005 SAT—a wait of more than two 8 years.283 152. 9 BOARS provided its most comprehensive justification for UC’s continued use of 10 the SAT and ACT in 2009. BOARS found that, compared to their predecessors, the revised SAT 11 and ACT-with-Writing were better aligned with college preparatory curricula and, when combined 12 with high school GPA, “add[ed] a small but notable increment of predictive power” to the 13 prediction of first-year GPA.284 According to BOARS, without SAT and ACT scores’ admittedly 14 “small” contribution to predicting first-year GPA, “UC would lose much of its predictive 15 modeling capability.”285 BOARS emphasized that the scores could be used to address “anomalies” 16 in the application process, by facilitating the evaluation of “non-traditional applicants without high 17 school GPAs” and identifying applicants “whose high school GPAs are at odds with their test 18 scores.”286 Without test scores to provide a “uniform reference point,” BOARS asserted, high 19 schools might “engage in more grade inflation” and public confidence in the fairness of UC’s 20 admissions process could dwindle.287 Finally, BOARS expressed concern about UC losing its 21 22 23 24 281 Id. (emphasis omitted). 282 Id. at 18. 283 25 Douglass, supra note 5, at 231. 284 BOARS, supra note 7, at 25. 26 285 Id. at 4. 27 286 Id. 287 Id. 28 72 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 “capacity to compare itself to other national institutions”—that is, its ability to maintain its 2 reputation through national rankings.288 153. 3 BOARS weighed these justifications against the SAT and ACT’s numerous 4 deficiencies. First, BOARS noted the “considerable redundancy” of using both high school GPA 5 and SAT or ACT scores to predict first-year GPA. 289 BOARS also recognized that, because the 6 tests could be coached, disparate access to test preparation “disadvantages low-income groups and 7 distracts participants from college preparatory coursework.” 290 Most importantly, BOARS 8 acknowledged the tests’ adverse effects on underrepresented minority students and students with 9 less wealth: BOARS stated that despite revisions to both the ACT and SAT, neither test had 10 “escaped the persistent differences between racial/ethnic groups” that had characterized their 11 predecessors.291 By contrast, BOARS observed that “[h]igh school GPA has less adverse impact 12 on disadvantaged groups and is the better predictor of student retention over time.” 292 13 Nevertheless, BOARS ultimately decided that the justifications discussed supra outweighed not 14 only the limited predictive value of the tests, but also the inequalities they reinforced, and 15 recommended continued use of both tests.293 154. 16 BOARS most recently justified its continued use of the SAT in a statement 17 announcing its adoption of the 2016 redesigned SAT as an acceptable admissions exam. 294 18 Relying entirely on information provided by the College Board, BOARS stated that it “expects the 19 revised SAT to be a fairer test that is more closely aligned with BOARS’ testing principles than 20 the existing SAT, in that the new test will be more aligned with college-level expectations and 21 288 Id. 289 Id. at 15. 290 Id. at 21. 24 291 Id. at 24. 25 292 Id. 26 293 Id. at 4–5. 22 23 294 BOARS, Statement on the Redesigned SAT (July 8, 2014), available at https://senate.universityofcalifornia.edu/ 28 _files/committees/boars/SAT_Redesign.pdf. 27 73 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 high school curriculum.”295 Without having studied the redesigned SAT’s predictive validity, 2 BOARS nevertheless asserted that “the new SAT meets BOARS general objectives and principles 3 for admissions tests and should be an improvement over the current exam.” 296 155. 4 In fact, as explained infra paras. 184–92, none of these reasons is necessary to UC 5 admissions processes and none outweighs the significant discriminatory impact of the use of such 6 exams. 7 E. Current Use of SAT and ACT in UC Admissions 8 156. The UC requires all prospective first-year students applying to a UC campus to 9 “submit scores from either the ACT with Writing or SAT with Essay.” 297 In addition to submitting 10 SAT or ACT scores, prospective first-year students must complete a minimum of 15 academic or 11 college preparatory courses in certain subjects (“A-G courses”) and submit scores from two 12 supplementary subject tests.298 Students who meet these minimum criteria are “entitled to review 13 (ETR)” at any UC campus to which they apply, but are not guaranteed admission to any UC 14 campus.299 15 Statewide and Local Admissions Guarantees 16 157. There are two paths to guaranteed admission to a UC campus: a statewide path and 17 a local path known as Eligibility in the Local Context.300 The statewide path identifies the top nine 18 percent of California graduates using an admissions index of high school GPA and test scores. 301 19 The local path identifies the top nine percent of students at participating California high schools 20 21 295 Id. 22 296 Id. 297 Univ. of Cal. Admissions, supra note 135. 23 298 Univ. of Cal. Bd. of Regents, Regents Policy 2103: Policy on Undergraduate Admissions 24 Requirements, https://regents.universityofcalifornia.edu/governance/policies/2103.html. 25 299 Id. 300 Univ. of Cal. Admissions, California Residents, https://admission.universityofcalifornia.edu/admission-requirements/freshman27 requirements/california-residents/. 26 28 301 Univ. of Cal. Bd. of Regents, supra note 298. 74 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 based solely on grades in A-G coursework. 302 Prospective first-year students who qualify for 2 either the statewide path or the local path and are not admitted to any campus to which they apply 3 “will be offered admission at a UC campus with available space.” 303 Taken together, the statewide 4 and local guarantees are known as the “9-by-9 policy.” 304 158. 5 The statewide admissions index places considerable weight on SAT and ACT 6 scores, essentially granting a wealth and race bonus to affluent and White students. A student’s 7 score on the index is calculated by: (1) converting the student’s high school GPA to a standardized 8 “UC GPA”; (2) converting the student’s SAT or ACT score to a “UC Score” using UC conversion 9 tables; and (3) using a UC index table to determine whether the student’s UC Score meets the 10 minimum requirement for the student’s UC GPA range. 305 A student falls within the top nine 11 percent of students statewide—and is thus guaranteed admission to a UC campus—if his or her 12 UC Score “meets or exceeds” the minimum score for his or her UC GPA range. 306 Thus, a 13 student’s SAT or ACT score is a deciding factor in determining whether the student will be 14 guaranteed admission through the statewide path. 159. 15 Although the Eligibility in the Local Context formula relies solely on high school 16 grades to determine which students are guaranteed admission to a UC campus, students are 17 required to submit either SAT or ACT scores with their applications, and those scores are 18 considered by the individual campuses to which they apply. To determine Eligibility in the Local 19 20 21 302 Id. 303 Id. 304 BOARS, Impact of the New Freshman Eligibility Policy at the University of California 3 (Nov. 22 2013), available at 23 https://senate.universityofcalifornia.edu/_files/committees/boars/Nov52013BOARSReporttoRege nts-Final.pdf. 24 305 Univ. of Cal. Admissions, Statewide Guarantee, https://admission.universityofcalifornia.edu/admission-requirements/freshman25 requirements/california-residents/statewide-guarantee/; Univ. of Cal. Admissions, Admissions 26 Index Instructions, https://admission.universityofcalifornia.edu/admission-requirements/freshmanrequirements/california-residents/statewide-guarantee/admissions-index-instructions.html 27 [hereinafter Admissions Index]. 28 306 Admissions Index, supra note 305. 75 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 Context, UC identifies the top nine percent of students at participating California high schools 2 based on “GPA in UC-approved coursework completed in the 10 th and 11th grades.”307 To be 3 considered, a student must have at least a 3.0 GPA and complete required A-G coursework prior to 4 his or her senior year.308 UC then compares the student’s GPA “to the historic top GPA” from the 5 student’s school to determine whether the student falls within the top nine percent of his or her 6 class, and is thus guaranteed admission to a UC campus. 309 160. 7 Taking these two policies together, the wealth and race bonus provided by SAT and 8 ACT scores is clear: a student whose GPA does not place him or her in the top nine percent of his 9 or her high school class will not be Eligible in the Local Context, but that student may 10 nevertheless be guaranteed admission through the statewide path if, once his or her GPA is 11 supplemented by his or her SAT or ACT score, he or she falls in the top nine percent of students 12 statewide. By contrast, a student who falls just outside of the top nine percent of his or her high 13 school class, but whose SAT or ACT score is not high enough to place the student in the top nine 14 percent of students statewide, may not be able to attend a UC at all. 15 “Holistic” Admissions 16 161. Although UC professes to employ a “holistic” review process to evaluate 17 applicants, that process is irretrievably tainted by its requirement that applicants submit SAT and 18 ACT scores. UC ostensibly follows a “comprehensive review” policy in undergraduate 19 admissions, which it defines as evaluating students “using multiple measures of achievement and 20 promise, while considering the context in which each student has demonstrated academic 21 accomplishment.”310 UC’s guidelines for implementing its comprehensive review policy 22 23 307 25 308 Id. 26 309 Id. Univ. of Cal. Admissions, Local Guarantee (ELC), https://admission.universityofcalifornia.edu/admission-requirements/freshman24 requirements/california-residents/local-guarantee-elc.html. 310 BOARS, Annual Report on Undergraduate Admissions Requirements and Comprehensive Review 8 (Apr. 2019), available at 28 https://senate.universityofcalifornia.edu/_files/committees/boars/boars-2019-cr-report.pdf (quoting 27 76 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 enumerate 14 criteria that individual campuses use to select applicants for admission, including 2 SAT or ACT scores.311 In 2011, the Regents endorsed a resolution stating that a single-score 3 holistic evaluation is “the expected implementation” of comprehensive review. 312 Single-score 4 holistic review assigns one score to an applicant based on the totality of the information in his or 5 her application.313 All but three UC undergraduate campuses use a single-score review system. 314 6 7 Univ. of Cal. Bd. of Regents, Regents Policy 2104: Policy on Comprehensive Review in 8 Undergraduate Admissions (Nov. 15, 2001)). 311 Id. The 14 criteria are (1) academic GPA; (2) SAT or ACT scores; (3) “number, content of, and 9 performance in courses completed in academic subjects beyond the minimum specified by the 10 University’s eligibility requirements”; (4) “number of and performance in University approved honors courses, College Board Advanced Placement courses, International Baccalaureate courses, 11 and transferable college courses completed”; (5) Eligibility in the Local Context; (6) “quality of the senior year program”; (7) “quality of academic performance relative to” available educational 12 opportunities; (8) “[o]utstanding performance in one or more specific academic subject areas”; (9) “[o]utstanding work in one or more special [academic] projects”; (10) “[r]ecent, marked 13 improvement in academic performance”; (11) “special talents, achievements, and awards in a 14 particular field . . . ; special skills . . . ; special interests . . . ; . . . experiences that demonstrate unusual promise for leadership . . . ; or other significant experiences or achievements that 15 demonstrate the applicant’s promise for contributing to” a campus’s “intellectual vitality”; (12) “[c]ompletion of special projects”; (13) “[a]cademic accomplishments in light of the applicant’s 16 life experiences and special circumstances”; and (14) “[l]ocation of the applicant’s secondary 17 school and residence.” Univ. of Cal., Guidelines for Implementation of University Policy on Undergraduate Admissions, available at http://www.ucop.edu/student-affairs/_files/guidelines-for18 implementation-of-undergraduate-admissions--rev-7-2019.pdf [hereinafter Comprehensive Review Guidelines]. 19 312 Univ. of Cal. Bd. of Regents, Regents Policy 2108: Resolution Regarding Individualized 20 Review and Holistic Evaluation in Undergraduate Admissions, available at https://regents.universityofcalifornia.edu/governance/ 21 policies/2108.html. The resolution nevertheless permitted campuses to use other approaches to “achiev[e] campus and University goals.” Id. 22 313 See BOARS, supra note 310, at 8. 23 314 Id. at 32. At the three campuses that do not use a single-score review system—UC Merced, UC Riverside, and UC Santa Barbara—the adverse effect of consideration of SAT and ACT scores is 24 particularly severe. At UC Merced, although all applicants receive a “point-driven comprehensive 25 review on academic factors,” only a “subset of the applicant pool receiv[es] a human read score.” Id. at 37. Although UC Riverside intends to transition to single-score holistic review, it currently 26 uses a “fixed-weight calculation” to admit applicants that weights and “sums a subset of the fourteen BOARS criteria” into an “Academic Index Score.” Id. at 38–39. At UC Santa Barbara, 27 only 50 percent of an applicant’s score is determined by individualized review; the other 50 28 percent is based solely on high school GPA and test scores. Id. at 41. 77 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 162. 1 The UC assures applicants that its admissions officers “look beyond . . . test scores” 2 and that their “academic achievements” will be evaluated “in light of the opportunities available to 3 [them].”315 UC’s first-year admissions information website downplays the significance of SAT 4 and ACT scores by listing them among 13 other admissions criteria, and explains that different 5 campuses “often apply these factors differently.” 316 This lack of transparency as to how each UC 6 campus weighs SAT and ACT scores against other admissions criteria harms students even before 7 they submit applications. Because prospective applicants have no way of knowing to what degree 8 their SAT or ACT scores will be utilized at most UC campuses, they cannot make informed 9 decisions about where to apply and, more importantly, whether to expend precious time and 10 resources in preparing for the SAT or ACT. 163. 11 In its Comprehensive Review Guidelines, UC instructs admissions officers to 12 consider academic indicators “in the context of other factors that impact performance, including 13 personal and academic circumstances (e.g. low-income status, access to honors courses, and 14 college-going culture of the school).”317 Like the College Board’s failed “adversity score,” 15 however, UC’s guidance to admissions officers cannot cure the unlawful discrimination injected 16 into the admissions process by UC’s requirement that all applicants submit SAT and ACT scores, 17 which act as a proxy for socioeconomic status and race. UC’s attempt to mitigate the adverse 18 effects of its consideration of SAT and ACT scores by assessing those scores “in the context of 19 other factors” does not and cannot compensate for its refusal to eliminate the discriminatory 20 requirement itself. 164. 21 The adverse effects of UC’s consideration of SAT and ACT scores are particularly 22 pronounced with respect to underrepresented minority students. Whereas the UC at least attempts 23 to redress SAT and ACT scores’ wealth-based discrimination by considering applicants’ 24 25 315 27 316 Id. 317 Comprehensive Review Guidelines, supra note 311. Univ. of Cal. Admissions, How Applications Are Reviewed, https://admission.universityofcalifornia.edu/how-to-apply/applying-as-a-freshman/how26 applications-are-reviewed.html. 28 78 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 socioeconomic status, UC ceased considering applicants’ race in admissions decisions following 2 the approval of Proposition 209.318 Thus, despite UC’s own research demonstrating that race “has 3 a large, independent, and growing statistical effect” on applicants’ SAT and ACT scores, UC 4 admissions officers are effectively prohibited from taking into account how those scores are 5 “conditioned by race.”319 The use of SAT and ACT scores therefore incorporates systematic racial 6 discrimination into the structure of UC’s decision-making. 165. 7 An admissions process that knowingly requires all applicants to submit a criterion 8 that is biased against underrepresented minority students, students with disabilities, and students 9 with less wealth is, by definition, not “merits-based.” Indeed, UC’s intentional discrimination 10 against the less advantaged and in favor of the wealthy in the admissions process is not limited to 11 the consideration of SAT and ACT scores: despite its representations to the contrary, the UC gives 12 preferential access to “legacy” applicants who are related to alumni and to children of wealthy and 13 prospective donors. In the wake of the 2019 college admissions scandal, the UC published on its 14 website a list of frequently asked questions regarding its admissions policies. 320 The UC denied 15 “grant[ing] preferential admission to the children of alumni or donors,” citing “long-established 16 UC admissions policy.”321 But as UC’s own internal audit noted, the referenced policy fails to 17 expressly prohibit legacy admissions, providing only that “admissions motivated by concern for 18 financial, political, or other such benefit to the University do not have a place in the admissions 19 process.”322 This failure led UC’s internal audit service to conclude that the policy’s language 20 21 318 Geiser, supra note 31, at 6. 22 319 Id. 23 24 320 Univ. of Cal., Current Issues: Frequently Asked Questions on UC Admissions Policies (June 20, 2019), https://www.universityofcalifornia.edu/current-issues/admissions/faqs. 321 Id. 322 Univ. of Cal. Ethics, Compliance & Audit Servs., Systemwide Audit of Undergraduate Admissions 18 (June 2019), available at 26 https://www.universityofcalifornia.edu/sites/default/files/audit-report.pdf (quoting Univ. of Cal. Bd. of Regents, Regents Policy 2202: Policy Barring Development Considerations from 27 Influencing Admissions Decisions, available at 28 https://regents.universityofcalifornia.edu/governance/policies/2202.html). 25 79 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 could be “strengthen[ed] . . . to more explicitly prohibit development and legacy considerations 2 from influencing admissions decisions.”323 166. 3 UC’s internal audit also determined that additional protocols were needed to ensure 4 “that admissions decisions are not motivated by concern for financial benefit to the University.” 324 5 Specifically, the audit report instructed each UC campus to “limit[] communication between 6 development personnel and the admissions office regarding admissions matters,” and to prohibit 7 “communication regarding the admission status of specific applicants.” 325 The audit report also 8 called for “retrospective review[s] of donations . . . to identify admissions decisions that could 9 have been influenced by these donations.” 326 10 V. Disparate Outcomes 11 A. Disparate Outcomes for Underrepresented Minority Students 12 167. For the reasons discussed supra, the SAT and ACT tests are highly discriminatory. 13 They have also resulted in starkly disparate student outcomes. 168. 14 According to College Board’s 2019 data for students taking the SAT in California, 15 45 percent of White students scored 1200 or above, compared to only nine percent of Black 16 students and 12 percent of Latinx students.327 Only one percent of Black students and two percent 17 of Latinx students scored in the top score bracket, compared to 12 percent of White students. 328 By 18 contrast, 63 percent of Black students and 57 percent of Latinx students scored in the three lowest 19 score brackets, compared to 17 percent of White students.329 169. 20 These disparities are replicated on the national scale, where only one percent of 21 Black students and two percent of Latinx students were in the top-scoring bracket, as compared to 22 23 24 323 Id. 324 Id. 325 25 Id. 326 Id. 26 327 College Board, supra note 11. 27 328 Id. 329 Id. 28 80 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 eight percent of White students.330 Sixty-seven percent of Black students and 56 percent of Latinx 2 students were in the three lowest score brackets, compared to 27 percent of White students. 331 170. 3 These disparities are longstanding. A Brookings Institute report analyzing College 4 Board data found that between 1996 and 2015, the wide racial gap in SAT math scores “remained 5 virtually unchanged.”332 171. 6 The ACT exhibits similar disparities. Among the national graduating class of 2018, 7 the average scores were 16.9 for Black students and 18.8 for Latinx students, as compared to 22.2 8 for White students.333 In California, racial disparities in 2018 graduates’ ACT scores were even 9 greater: Black and Latinx students had average scores of 19.0 and 19.7, respectively, as compared 10 to 25.5 for White students.334 172. 11 Although Asian students have the highest scores when grouped together by the 12 College Board and ACT,335 such groupings mask the demographic diversity of Asian American 13 and Pacific Islander (AAPI) populations. 336 The College Board and ACT’s monolithic treatment of 14 15 16 330 College Board, supra note 111. 17 331 Id. 18 332 Richard V. Reeves & Dimitrios Halikias, Race gaps in SAT scores highlight inequality and hinder 19 upward mobility (Feb. 1, 2017), https://www.brookings.edu/research/race-gaps-in-sat-scoreshighlight20 inequality-and-hinder-upward-mobility/. 21 333 ACT, supra note 174, at 15. 22 334 ACT, ACT Profile Report – California: Graduating Class 2018 15 (2018), available at 23 https://www.act.org/ content/dam/act/unsecured/documents/cccr2018/P_05_059999_S_S_N00_ACT24 GCPR_California.pdf. 25 335 College Board, supra note 11; College Board, supra note 111; ACT, supra note 334, at 15. 336 See, e.g., Robert Teranishi et al., iCount: A Data Quality Movement for Asian Americans and 26 Pacific Islanders in Higher Education 9–10 (2013), available at https://files.eric.ed.gov/fulltext/ED573772.pdf (describing differences in educational attainment, 27 median household income, and patterns of immigration among Asian American and Pacific 28 Islander subgroups). 81 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 AAPI students obscures the fact that certain subgroups score much lower than average. 337 For 2 example, one study found that 48.3 percent of California test-takers received a total SAT score 3 greater than 1500, whereas only 7.6 percent of test-takers at a predominantly Hmong high school 4 and 12.8 percent at a predominantly Filipinx high school received a score in the same range. 338 5 B. Adverse Effect on Underrepresented Minority Students 6 173. The disparate outcomes described above systematically harm underrepresented 7 minority students in the UC admissions process. Because the disparities between underrepresented 8 minority students and others are much greater for SAT and ACT scores than for high school 9 grades, consideration of SAT and ACT scores places underrepresented minority students at a 10 pronounced disadvantage relative to consideration of high school GPA alone. 339 11 12 13 14 15 16 17 18 19 20 21 22 23 24 337 See, e.g., id. at 18; Univ. of Cal. San Diego, SAT Reasoning Test Scores by Gender, Ethnicity, 25 Major Field of Study, Home Location, 1st Generation College Status, Income & Athletic Status (2016), available at https://ir.ucsd.edu/_files/stats-data/admissions/freshmen/hssat.pdf (showing 26 that from 2011 to 2016, Filipinx students scored lower than both White and Asian Students on the SAT). 27 338 Teranishi et al., supra note 336, at 18. 28 339 Geiser, supra note 31, at 10. 82 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 Figure 4: Distribution of Underrepresented Minority Applicants by SAT/ACT vs. HSGPA 2 Deciles, 1994 to 2011340 3 4 5 6 7 8 9 10 11 12 13 14 15 16 174. Figure 4 illustrates how consideration of SAT and ACT scores displaces high- 17 performing (as measured by high school GPA) underrepresented minority applicants from the top 18 tiers of the UC applicant pool—and thus reduces their likelihood of admission. When ranked by 19 high school grades, underrepresented minority students comprised 12 percent of applicants in the 20 top decile. When ranked by SAT and ACT scores, however, the representation of 21 underrepresented minority students fell to only five percent. Conversely, ranking applicants by 22 SAT and ACT scores resulted in underrepresented minority students comprising 60 percent of the 23 bottom decile, as opposed to only 39 percent when ranked by high school grades. Consideration of 24 SAT and ACT scores thus creates severe “racial stratification” among UC applicants, making it 25 more difficult for high-performing underrepresented minority students to reach the highest levels 26 27 28 340 Id. 83 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 of the applicant pool.341 When ranked by SAT and ACT scores, rather than by high school GPA, 2 “[u]nderrepresented minority applicants are less than half as likely to rise to the top of the pool”— 3 a major roadblock to admission.342 175. 4 Relative to all students and to White students alone, Black and Latino students are 5 admitted at substantially lower rates across the UC system. Figure 5. 6 7 Figure 5: Acceptance Rates for Freshman Applicants to UC Campuses by Race/Ethnicity, Fall 2018343 8 9 10 11 12 13 176. 14 Among Asian American and Pacific Islander populations, racial stratification in 15 SAT and ACT scores is also reflected in disparate rates of acceptance to the UC. Duplicated data 16 from UC Berkeley, for example, show that the acceptance rate for Fall 2018 first-years was 16.06 17 percent for White applicants and 15.06 percent for aggregated Asian applicants. 344 Disaggregating 18 19 341 Id. 20 342 Id. 343 See Univ. of Cal. Infocenter, Undergraduate admissions summary, https://www.universityofcalifornia.edu/ 22 infocenter/admissions-residency-and-ethnicity. Acceptance rates were calculated by dividing the raw number of admitted students by the raw number of applicants from the applicable racial 23 category and year. 21 24 25 26 27 28 344 See Univ. of Cal. Infocenter, Disaggregated data, https://www.universityofcalifornia.edu/infocenter/ disaggregated-data (“[C]ounts are duplicated if a student selects multiple categories.”). Acceptance rates were calculated by dividing the raw number of admitted students by the raw number of applicants from the applicable racial category, campus, and year. These percentages differ from those represented in Figure 5 because they are based on duplicated data, which is the format in which the UC provides disaggregated data. By contrast, the percentages in Figure 5 are based on apparently unduplicated data. Univ. of Cal. Infocenter, supra note 343. 84 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 the data on Asian applicants, however, demonstrates large disparities in admissions rates: Korean 2 and Taiwanese applicants had acceptance rates of 19.18 percent and 18.40 percent, respectively, as 3 compared to acceptance rates of 12.35 percent for Filipinx applicants and 9.55 percent for Hmong 4 applicants.345 5 C. Disparate Outcomes for Students with Disabilities 6 177. For the reasons discussed supra, test scores of students with disabilities (even with 7 accommodations) underpredict their chances of college success and are lower than the scores of 8 students without disabilities. Although the College Board and the ACT currently obscure these 9 disparities by excluding accommodated test-takers from their annual reports and most of their 10 validity studies, historical data confirms the disparate outcomes experienced by students with 11 disabilities. According to ACT’s 2019 Technical Manual, among test-takers in 2013 and 2014, 12 “[o]n average, students with disabilities testing with accommodations earn[ed] lower scores than 13 those of students from the overall ACT-tested population. The few exceptions are students with 14 motor impairments and psychiatric disorders (e.g., mood or anxiety).” 346 15 // 16 // 17 18 19 20 21 22 23 24 25 26 27 28 345 Id. 346 ACT, supra note 78, at 11.59. 85 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL Figure 6: Average ACT Scores for Students Tested with Accommodations in 2013-2014 347 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Although students with disabilities have lower overall scores, ACT nevertheless maintains that 17 those students must achieve higher scores on the ACT to meet college readiness benchmarks. 348 18 The College Board’s 2014 College-Bound Seniors: Total Group Profile Report demonstrated 19 similar disparities in SAT performance: students testing with accommodations had a mean critical 20 reading score of 477 and a mean math score of 479, as compared to a mean critical reading score 21 of 497 and a mean math score of 513 for the overall group. 349 22 Higher Education and Employment 23 178. For many people with disabilities, finding and sustaining employment is a 24 challenge. In the United States, only approximately one in three individuals with disabilities is 25 26 347 Id. at 11.60. 27 348 See id. at 11.70–11.71. 28 349 College Board, 2014 College-Bound Seniors: Total Group Profile Report 1–2 (2014), available at https://secure-media.collegeboard.org/digitalServices/pdf/sat/TotalGroup-2014.pdf. 86 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 employed, as compared to three quarters of their counterparts without disabilities. This gap does 2 not appear to be closing. Even when employed, workers with disabilities are more likely than their 3 counterparts without disabilities to report underemployment, involuntary part-time or contingent 4 employment, and lower than average salaries. 5 179. The limited academic opportunities for students with disabilities based on their 6 standardized test scores has produced a persistent gap between students with and without 7 disabilities who enroll in college. Individuals who graduate from college are two to three times 8 more likely to maintain employment than those who do not obtain a postsecondary degree. 9 Limited access to public higher education thus contributes to the ongoing unemployment and 10 underemployment of persons with disabilities in California and the United States. 11 180. Until testing accommodations can be proven effective and equally accessible to all 12 students who need them, and until students with legitimate disabilities are no longer treated with 13 suspicion and skepticism, the only unbiased way for the UC to evaluate students with disabilities 14 in its admissions process is to eliminate its test score requirement. 15 VI. Alternatives to UC’s Discriminatory Use of SAT and ACT Scores 16 181. The UC expresses a commitment to a holistic review process, but its current 17 requirement that all applicants submit test scores that act as a proxy for race and wealth renders 18 that process unlawful and invalid. As a threshold matter, UC’s statewide admissions index— 19 which identifies the top nine percent of California graduates by a calculation that takes into 20 account only their high school GPAs and SAT or ACT scores—incorporates no holistic review at 21 all into the determination of which students are guaranteed admission through the statewide path. 22 As discussed supra paras. 158–60, the statewide admissions index thus provides a wealth and race 23 bonus to White and affluent students, who—even if they fall outside the top nine percent of their 24 high school classes as ranked by GPA—may nevertheless secure guaranteed admission to the UC 25 if their SAT or ACT scores, coupled with their high school GPA, place them within the top nine 26 percent of students statewide. 27 182. By contrast, eliminating SAT and ACT scores (and their corresponding race and 28 wealth bonus) and basing guaranteed admission on high school grades alone results in substantial 87 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 equity gains. For example, the University of Texas at Austin—a prestigious university that, like 2 many UC campuses, receives a high volume of applications—automatically admits Texas students 3 whose grades place them within a specified percentile of their high school graduating class. 350 4 Researchers have found that this process immensely benefits the students it “pulls in” to UT 5 Austin (i.e., high-performing students at schools that have traditionally sent few or no students to 6 UT Austin), without resulting in significant harm to the students it crowds out. 351 183. 7 Although UC’s discriminatory use of SAT and ACT scores is most egregious as 8 applied in its statewide admissions index, the inclusion of SAT and ACT scores in admissions 9 officers’ decision-making also compromises its “holistic” review process. As currently used by the 10 UC, SAT and ACT scores create severe socioeconomic and racial stratification in the UC 11 applicant pool, preventing talented underrepresented minority students, students with less wealth, 12 and students with disabilities from rising to the top. Requiring SAT and ACT scores from all 13 applicants concentrates privilege on UC campuses not only by systematically placing White and 14 affluent students at a significant competitive advantage relative to their equally talented peers, but 15 also by preventing admissions officers from according appropriate weight to more meaningful, 16 less biased metrics. Rather than redressing these harms, the Regents have chosen to continue 17 requiring applicants to submit scores with known racial and socioeconomic disadvantages for 18 students from historically underrepresented groups and benefits for students with race and wealth 19 privilege. 184. 20 There are many alternative metrics to SAT and ACT scores that provide 21 comparable or superior information about a student’s ability to succeed in college without creating 22 23 350 UT Austin is required to automatically admit enough students to fill 75 percent of available resident spaces, and annually adjusts the percentile rank required for automatic admission 24 accordingly. Univ. of Tex. at Austin Office of Admissions, Admission Decisions, https://admissions.utexas.edu/apply/decisions#fndtn-freshman-admission. In the Fall 2017 and 25 Fall 2018 admissions cycles, for example, applicants who ranked in the top seven percent of their 26 high school graduating class were automatically admitted to UT Austin. Id. 351 Those students—who fall outside of the top 10 percent of graduating classes at “feeder” high 27 schools that have traditionally sent many students to UT Austin—do not experience declines in 28 overall college enrollment, graduation, or earnings. 88 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 such extreme socioeconomic and racial disparities. Most of these metrics are already in applicant 2 files, which contain a wealth of data on the student’s academic achievements, personal qualities, 3 extracurricular activities, community involvement, and special talents. Placing more weight on 4 quantitative metrics such as high school GPA and qualitative evaluations by teachers and 5 counselors is not only more equitable, it also results in a more meaningful assessment of each 6 student’s capacity to thrive in college and contribute to the UC community. 352 185. 7 A student’s high school GPA is both less discriminatory and more predictive than 8 his or her SAT and ACT scores. As UC researchers have demonstrated, high school GPA “is 9 consistently the strongest predictor of four-year college outcomes for all academic disciplines, 10 campuses and freshman cohorts in the UC sample,” and it has “less adverse impact than [the SAT 11 and ACT] on disadvantaged and underrepresented minority students.” 353 Recent analyses have 12 confirmed that using “[h]igh school GPA as a predictor of college success results in a much higher 13 representation of low income and underrepresented minority students in the top of the UC 14 applicant pool.”354 186. 15 UC officials are well aware of both the superiority of high school GPA as a 16 predictor of college performance and the adverse impact of consideration of SAT and ACT scores 17 on underrepresented minority students and students with less wealth. BOARS itself has recognized 18 not only that “[m]ost studies point to high school GPA as the best single predictor of college 19 success,” but also that standardized tests like the SAT and ACT have a “negative impact on 20 underrepresented groups” that is “less readily mitigated” than those of other criteria. 355 21 Furthermore, BOARS has conceded that using both high school GPA and SAT and ACT scores to 22 predict college outcomes is “highly redundant in California.” 356 The miniscule contribution of 23 24 25 352 Tough, supra note 33, at 176 (describing how SAT and ACT scores act as a “false signal” for less advantaged students that distracts admissions officers from more meaningful criteria). 353 Geiser & Santelices, supra note 65, at 1. 26 354 Kurlaender & Cohen, supra note 39. 27 355 BOARS, supra note 7, at 4. 356 Id. 28 89 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 SAT and ACT scores to predicting first-year GPA is thus unnecessary for making admissions 2 decisions. 187. 3 According to BOARS, SAT and ACT scores are needed in addition to high school 4 GPA to “provide students and schools with a uniform reference point that is independent of a 5 particular school’s grading system or associated biases.” 357 Yet despite variations in grading 6 systems and “associated biases,” high school GPA has consistently outperformed SAT and ACT 7 scores as a predictor of long-term student success at the UC. 358 That is, even across high schools 8 with different grading standards, a student’s GPA is a better predictor of his or her long-term 9 performance at the UC than his or her SAT or ACT scores. As UC researchers have demonstrated, 10 “[i]n highly competitive applicant pools like UC’s, . . . [u]sing SAT/ACT scores to rank applicants 11 introduces a substantial amount of error in admissions decision-making” due to the scores’ small 12 effect size and the wide error bands around predicted outcomes. 359 BOARS’s related worry—that 13 placing more emphasis on high school GPA could “encourage high schools to engage in more 14 grade inflation”360—can also be addressed by considering information already in applicants’ files: 15 class rank. Class rank counterbalances potential grade inflation by contextualizing high school 16 GPAs for admissions officers. 188. 17 Critically, BOARS’s emphasis on biases in high school grading systems ignores 18 well-documented biases in the SAT and ACT themselves. 361 As a tool for comparing students of 19 different racial backgrounds, students with different abilities, and students with different levels of 20 wealth, SAT and ACT scores are a “uniform reference point” only insofar as they uniformly 21 correlate with students’ access to advantage. 189. 22 With respect to California applicants and students in the six other states whose high 23 24 357 25 Id. 358 Geiser & Santelices, supra note 65, at 1. 26 359 Geiser, supra note 31, at 9–10. 27 360 BOARS, supra note 7, at 4. 361 Supra paras. 54–64. 28 90 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 schools use the same exam,362 the statewide Smarter Balanced Assessment also provides a 2 “uniform reference point” whose unique contribution to predicting college outcomes is 3 comparable to that of the SAT, but with less adverse impact on underrepresented minority students 4 and students with less wealth.363 Moreover, unlike the SAT and ACT, which are norm-referenced 5 tests, the Smarter Balanced Assessment is an achievement test that measures students against fixed 6 curricular standards, rather than forcing their scores into an artificial bell curve pattern. Because it 7 is tied to curricular standards, the Smarter Balanced Assessment has the advantage of reinforcing 8 the importance of a rigorous college preparatory curriculum in high schools. Use of the Smarter 9 Balanced Assessment may also limit harms arising from unequal access to test preparation by 10 rewarding mastery of curriculum instead of test-taking strategy. 190. 11 The UC maintains that—unlike comparable, less discriminatory metrics—SAT and 12 ACT scores provide the UC with the “capacity to compare itself to other national institutions.” 364 13 To the extent that this statement reflects BOARS’s concern with maintaining UC’s national 14 rankings, such “capacity” is a dubious metric for assessing the utility of SAT and ACT scores in 15 individual admissions decisions. Requiring SAT and ACT scores in order to bolster UC’s national 16 reputation is fundamentally at odds with UC’s pedagogical obligation to provide undergraduate 17 education to all eligible California high school graduates on a nondiscriminatory basis. 365 UC’s 18 capacity to be ranked against other universities has become less salient, moreover, as colleges and 19 universities across the country increasingly stop requiring students to submit SAT and ACT 20 21 362 Catherine Gewertz, Which States Are Using PARCC or Smarter Balanced, Educ. Week (updated Apr. 9, 2019), https://www.edweek.org/ew/section/multimedia/states-using-parcc-or23 smarter-balanced.html. 22 24 363 27 364 BOARS, supra note 7, at 4. 365 Supra paras. 128–33. Kurlaender & Cohen, supra note 39, at 21 (describing the “difference in magnitude” between the predictive power of the SAT and SBAC as “trivial”); id. at 18 (describing how consideration 25 of high school GPA with the Smarter Balanced Assessment results in more low-income students and more underrepresented minority students in the top decile of the UC applicant pool than 26 consideration of high school GPA with the SAT). 28 91 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 scores.366 191. 2 In recent years, a significant wave of colleges and universities—including large 3 research universities like the University of Chicago and George Washington University—have 4 either ceased requiring applicants to submit SAT and ACT scores (adopting a “test-optional” 5 policy) or otherwise deemphasized those scores in their admissions processes. 367 As of December 6 2019, more than 1,050 colleges and universities—about 40 percent of all accredited higher 7 education institutions—no longer use SAT or ACT scores to evaluate many or all of their 8 applicants.368 The experience of these institutions illustrates that SAT and ACT requirements may 9 be eliminated without disrupting admissions processes. 192. 10 At the University of Chicago, for example, admissions officers continue to consider 11 traditionally required information such as a student’s secondary school report, high school 12 transcript, and teacher recommendations.369 Rather than requiring students to submit SAT and 13 ACT scores, however, U. Chicago allows all applicants to submit a range of supplemental 14 materials, including “creative writing projects, highlights from music/dance/visual art/theater 15 performance, school capstone projects such as AP Capstone or the equivalent, research projects, 16 business plans, or other work of note.” 370 Following its adoption of a test-optional policy, U. 17 Chicago realized a 20 percent increase in “[t]he number of first-generation and low-income 18 19 366 See FairTest, More Than 1050 Accredited Colleges and Universities That Do Not Use ACT/SAT Scores to Admit Substantial Numbers of Students Into Bachelor-Degree Programs 21 (2019), https://www.fairtest.org/university/ optional. 22 367 Id. 23 368 Id.; Valerie Strauss, A Record Number of Colleges Drop SAT/ACT Admissions Requirement 24 Amid Growing Disenchantment with Standardized Tests, Wash. Post (Oct. 18, 2019), https://www.washingtonpost.com/education/ 25 2019/10/18/record-number-colleges-drop-satact-admissions-requirement-amid-growingdisenchantment-with-standardized-tests/. 26 369 Univ. of Chicago, First-Year Applicants, https://collegeadmissions.uchicago.edu/apply/first27 year-applicants. 20 28 370 Id. 92 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 students who committed to attend.”371 The experience of U. Chicago and other institutions 2 demonstrates not only that—notwithstanding BOARS’s claims to the contrary—it is feasible to 3 evaluate applicants without considering meaningless and discriminatory SAT and ACT scores, but 4 also that ceasing to require those scores enables universities to expand access to higher education 5 for traditionally underrepresented students while maintaining the strength and quality of their 6 academic programs.372 193. 7 However, simply ceasing to require SAT and ACT scores—without taking into 8 appropriate consideration their failure to contribute meaningfully to the admissions process, their 9 disproportionate impact, and the stigmatizing consequences for students who choose not to 10 submit—does not render an admissions process lawful under California law. Test-optional policies 11 can potentially preserve the structural advantage conferred by SAT and ACT scores on applicants 12 who choose to submit (i.e., higher-scoring applicants who are, as research demonstrates, 13 disproportionately White and affluent), and may also result in the stigmatization of 14 underrepresented students, whose peers may assume that they were only able to secure admission 15 by not submitting scores. Moreover, adoption of a test-optional policy does not transform SAT or 16 ACT scores into meaningful predictors of students’ ability to succeed in college. Thus, for a test17 optional admissions policy to be lawful, a university must demonstrate that the policy: (1) uses 18 SAT and ACT scores in a way that contributes meaningfully to the assessment of applicants who 19 20 371 22 372 Univ. of Chicago, UChicago Empower Meeting Goal of Removing Barriers to College Access, https:// 21 news.uchicago.edu/story/uchicago-empower-meeting-goal-removing-barriers-college-access. 23 24 25 26 27 28 At Wake Forest University, for example, adoption of a test-optional policy increased diversity among undergraduates by 90 percent, without affecting the academic quality of the student body: Wake Forest has found “no difference in academic achievement” between applicants who submitted scores and those who did not. Wake Forest Univ., Test Optional, https://admissions.wfu.edu/apply/test-optional/. This experience is not unique to Wake Forest: one recent study found both that ceasing to require SAT and ACT scores “can lead to . . . an increase in the representation of URM students (both numeric and proportionate) in [a university’s] applicant pool and . . . freshman class,” and also that students who did not submit test scores “graduated at rates equivalent to or slightly above” students who did. Steven T. Syverson et al., Defining Access: How Test-Optional Works 3, 44 (2018), available at https://www.nacacnet.org/globalassets/documents/publications/ research/defining-access-report-2018.pdf (emphasis omitted). 93 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 submit those scores; (2) is transparent in application; (3) does not produce disparate results based 2 on socioeconomic status or race; and (4) does not stigmatize those students who do not submit test 3 scores. Unless and until the Regents demonstrate that a proposed test-optional policy will satisfy 4 these criteria, the UC must cease its consideration of SAT and ACT scores in admissions decisions 5 entirely. 6 CAUSES OF ACTION 7 FIRST CAUSE OF ACTION 8 Violation of State Equal Protection Guarantees 9 (California Constitution, Article I, Section 7(a) & Article IV, Section 16(a); 10 Cal. Civ. Proc. Code §§ 526, 526 (a)) 11 Against All Defendants 12 194. Plaintiffs reallege and incorporate herein by this reference each of the allegations in 13 Paragraphs 1 through 193, inclusive. 14 195. Under Article I, Section 7(a) and Article IV, Section 16(a) of the California 15 Constitution, Defendants are required to ensure that Plaintiffs receive the equal protection of the 16 laws. 17 196. Defendants have knowingly violated, and continue to violate, Plaintiffs’ equal 18 protection rights by maintaining a discriminatory policy of requiring either SAT or ACT scores for 19 students applying for admission to the UC and further considering those scores as part of the 20 criteria for admission. For the reasons explained above, this policy illegally discriminates against 21 applicants on the basis of race and wealth, and thereby denies them equal protection under the 22 California Constitution. 23 197. The equal protection clause of the California Constitution also requires that 24 Defendants do not diminish Plaintiffs’ fundamental right to education, which includes the right of 25 access to public higher education, or Plaintiffs’ fundamental right to pursue an occupation. 26 198. Defendants have knowingly violated, and continue to violate, Plaintiffs’ 27 fundamental rights by maintaining a discriminatory policy of requiring either SAT or ACT scores 28 for students applying for admission to the UC and further considering those scores as part of the 94 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 criteria for admission, and thereby denying them their right of access to public higher education as 2 well as hindering their right to pursue an occupation. 3 199. Defendants’ discriminatory policy denies Plaintiffs equal access to public higher 4 education within California—and all its attendant benefits and opportunities—both by deterring 5 students from underrepresented minority groups, students from disadvantaged socioeconomic 6 backgrounds, and students with disabilities from applying to the UC, and by adversely impacting 7 the evaluation of those students’ applications and their likelihood of being granted admission. 8 200. Plaintiffs are harmed by Defendants’ failure to comply with all applicable 9 provisions of law and their legal duties, as set forth herein. 10 201. Plaintiffs have no plain, speedy, and adequate remedy for Defendants’ 11 constitutional violations in the ordinary course of law. 12 202. Unless enjoined, Defendants will continue to violate the right to receive equal 13 protection of the laws under the California Constitution, and Plaintiffs will suffer irreparable harm. 14 203. Wherefore, Plaintiffs pray for relief, as set forth below. 15 SECOND CAUSE OF ACTION 16 Discrimination on the Basis of a Protected Classification 17 (Government Code § 11135; Cal. Civ. Proc. Code §§ 526, 526(a)) 18 Against All Defendants 19 204. Plaintiffs reallege and incorporate herein by this reference each of the allegations in 20 Paragraphs 1 through 193, inclusive. 21 205. California Government Code section 11135 provides: (a) No person in the State of California shall, on the basis of . . . race, color, . . . ancestry, national origin, ethnic group identification, . . . mental disability, physical disability, [or] medical condition, . . . be unlawfully denied full and equal access to the benefits of, or be unlawfully subjected to discrimination under, any program or activity that is conducted, operated, or administered by the state or by any state agency, is funded directly by the state, or receives any financial assistance from the state. 22 23 24 25 26 27 206. Section 11135 further provides that, with respect to discrimination on the basis of 28 disability, covered programs and activities “shall meet the protections and prohibitions contained 95 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 in Section 202 of the federal Americans with Disabilities Act of 1990 (42 U.S.C. Sec. 12132)” and 2 its implementing regulations, except that if the laws of California prescribe stronger protections 3 and prohibitions, covered programs and activities shall be subject to the stronger protections and 4 prohibitions. Cal. Gov’t Code § 11135(b). 5 207. Through the UC, Defendants operate programs or activities that are conducted, 6 operated, or administered by the State or by any State agency, are funded directly by the State, or 7 receive any financial assistance from the State. 8 208. Plaintiffs include qualified individuals with a disability within the meaning of 9 California law, Cal. Gov’t Code § 12926. 10 209. Defendants’ policy of requiring SAT and ACT scores in the UC admissions 11 process, despite knowing for years of the discriminatory impacts of those tests, amounts to 12 intentional discrimination by Defendants against Plaintiffs on the basis of race, color, ancestry, 13 national origin, ethnic group identification, mental disability, physical disability, and/or medical 14 condition. 15 210. Defendants’ policy of requiring SAT and ACT scores in the UC admissions process 16 constitutes an ongoing violation of the right of Plaintiffs to be free from discrimination, through its 17 well documented disparate impact in disadvantaging underrepresented minority students and 18 students with disabilities. 19 211. Defendants knew or should have known that their conduct was directed toward 20 underrepresented minority students and students with disabilities. 21 212. Defendants cannot justify their policy of requiring SAT and ACT scores in the UC 22 admissions process, in light of the existence of more effective and less discriminatory methods to 23 evaluate applicants, as explained above. 24 213. Defendants’ acts and omissions as alleged herein also violate Section 11135 by 25 violating the Americans with Disabilities Act. 26 214. Plaintiffs are harmed by Defendants’ failure to comply with all applicable 27 provisions of law and their legal duties, as set forth herein. 28 96 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 215. Plaintiffs have no plain, speedy, and adequate remedy for Defendants’ legal 2 violations in the ordinary course of law. 3 216. Unless enjoined, Defendants will continue to violate Plaintiffs’ right to receive 4 nondiscriminatory treatment under California law, and Plaintiffs will suffer irreparable harm. 5 217. Wherefore, Plaintiffs pray for relief, as set forth below. 6 THIRD CAUSE OF ACTION 7 Prohibited Discrimination in Education 8 (Education Code § 66270; Cal. Civ. Proc. Code §§ 526, 526(a)) 9 Against All Defendants 10 218. Plaintiffs reallege and incorporate herein by this reference each of the allegations in 11 Paragraphs 1 through 193, inclusive. 12 219. No person shall be subjected to discrimination on the basis of disability, . . . nationality, [or] race or ethnicity, . . . or any characteristic listed or defined in Section 11135 of the Government Code or any other characteristic that is contained in the prohibition of hate crimes set forth in subdivision (a) of Section 422.6 of the Penal Code, including immigration status, in any program or activity conducted by any postsecondary educational institution that receives, or benefits from, state financial assistance or enrolls students who receive state student financial aid. 13 14 15 16 17 18 California Education Code section 66270 provides: 220. The UC is a state-run postsecondary educational institution that receives, or 19 benefits from, State financial assistance and which enrolls students who receive State student 20 financial aid. 21 221. Overwhelming data, including Defendants’ own extensive research, demonstrate 22 that the SAT and ACT tests discriminate against underrepresented minority students and students 23 with disabilities. 24 222. Despite this knowledge, Defendants continue to require SAT and ACT scores as 25 part of the UC’s admissions criteria, despite knowing that consideration of SAT and ACT scores 26 skews admissions decisions and deters applications from underrepresented, but equally talented, 27 minority students and students with disabilities. 28 97 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 223. As a result of Defendants’ policy, Plaintiffs were and continue to be subjected to 2 discrimination on the basis of disability, nationality, race, and/or ethnicity, depriving them of the 3 right of equal access to the educational benefits and opportunities afforded by the UC. 4 224. Plaintiffs are harmed by Defendants’ failure to comply with all applicable 5 provisions of law and their legal duties, as set forth herein. 6 225. Plaintiffs have no plain, speedy, and adequate remedy for Defendants’ legal 7 violations in the ordinary course of law. 8 226. Unless enjoined, Defendants will continue to violate Plaintiffs’ rights to equal 9 access to the UC, and Plaintiffs will suffer irreparable harm. 10 227. Wherefore, Plaintiffs pray for relief, as set forth below. 11 FOURTH CAUSE OF ACTION 12 Violation of the Unruh Act 13 (Civil Code §§ 51, 52; Cal. Civ. Proc. Code §§ 526, 526(a)) 14 Against All Defendants 15 228. Plaintiffs reallege and incorporate herein by this reference each of the allegations in 16 Paragraphs 1 through 193, inclusive. 17 229. The Unruh Act provides that: [a]ll persons within the jurisdiction of this state are free and equal, and no matter what their . . . race, color, . . . ancestry, national origin, disability, medical condition, . . . [or] primary language . . . are entitled to the full and equal accommodations, advantages, facilities, privileges, or services in all business establishments of every kind whatsoever. 18 19 20 21 22 Cal. Civ. Code § 51(b). The Unruh Act makes liable any person who “denies, aids or incites a 23 denial, or makes any discrimination or distinction contrary to Section 51.” Id. § 52(a). 24 230. Plaintiffs include persons with disabilities within the meaning of California Civil 25 Code section 54(b)(1). 26 231. Defendants’ requirement that all applicants submit SAT or ACT scores to be 27 considered for admission to the UC denies Plaintiffs full and equal accommodations, advantages, 28 facilities, privileges, and/or services on the basis of race, color, ancestry, national origin, disability, 98 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 medical condition, and/or primary language. This policy does so either directly—through the SAT 2 and/or ACT’s discriminatory bias against students from historically underrepresented racial and 3 ethnic groups and students with disabilities—or indirectly, as the mere presence of the testing 4 requirement deters otherwise-qualified applicants from these groups from applying to the UC at 5 all. 6 232. Nevertheless, Defendants intentionally continue to rely on SAT and ACT scores as 7 admissions criteria, despite Defendants’ own copious data revealing that consideration of SAT and 8 ACT scores skews admissions decisions in favor of White students to the detriment of their 9 underrepresented, but equally talented, peers. 10 233. As a result, Plaintiffs have been denied the myriad benefits of public higher 11 education, including, without limitation, higher wages and non-wage benefits, greater job security, 12 greater opportunities for homeownership, greater overall personal and family stability, and even a 13 decreased mortality rate. 14 234. Defendants’ own data, as well as the documented experience of other colleges and 15 universities who have dropped the requirement, shows the continued reliance on the SAT and 16 ACT tests to be a substantial factor in excluding these deserving Plaintiffs from their rightful 17 college opportunities and all the benefits that flow therefrom. 18 235. By enacting unlawful admissions barriers for underrepresented minority students 19 and students with disabilities and by failing to operate their programs on a nondiscriminatory 20 basis, Defendants denied, aided, or incited a denial, or made a discrimination or distinction 21 contrary to Section 51. 22 236. Plaintiffs are harmed by Defendants’ failure to comply with all applicable 23 provisions of law and their legal duties, as set forth herein. 24 237. Plaintiffs have no plain, speedy, and adequate remedy for Defendants’ legal 25 violations in the ordinary course of law. 26 238. Unless enjoined, Defendants will continue to deny Plaintiffs full and equal 27 accommodations, advantages, facilities, privileges, and/or services, and Plaintiffs will suffer 28 irreparable harm. 99 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 239. Wherefore, Plaintiffs pray for relief, as set forth below. 2 FIFTH CAUSE OF ACTION 3 Violation of the California Disabled Persons Act 4 (Civil Code §§ 54 et seq.; Cal. Civ. Proc. Code §§ 526, 526(a)) 5 240. Plaintiffs reallege and incorporate herein by this reference each of the allegations in 6 Paragraphs 1 through 193, inclusive. 7 241. The California Disabled Persons Act, Cal. Civ. Code §§ 54 et seq. (“CDPA”), 8 guarantees individuals with disabilities equal access to all public places. Any violation of the right 9 of an individual under the Americans with Disabilities Act also constitutes a violation of the 10 CDPA. 11 242. Plaintiffs include persons with disabilities within the meaning of California Civil 12 Code section 54(b)(1). 13 243. The University of California constitutes a public place within the meaning of the 14 CDPA, Cal. Civ. Code § 54.1. 15 244. Defendants have violated the CDPA by failing to operate their programs on a 16 nondiscriminatory basis and by failing to ensure that their programs are in compliance with civil 17 rights and access laws, including the Americans with Disabilities Act. 18 245. By enacting unlawful admissions barriers for persons with disabilities and by 19 failing to operate their programs on a nondiscriminatory basis, Defendants were and are in 20 violation of the CDPA. 21 246. Defendants knew or should have known that their conduct was directed toward 22 persons with disabilities. 23 24 25 26 27 247. The CDPA, Cal. Civ. Code § 54.3, provides that: (a) [a]ny person or persons, firm or corporation who denies or interferes with admittance to or enjoyment of the public facilities . . . or otherwise interferes with the rights of an individual with a disability under [the CDPA] is liable for each offense for the actual damages and any amount as may be determined by a jury, or the court sitting without a jury, up to a maximum of three times the amount of actual damages but in no case less than one thousand dollars ($1,000). 28 100 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 248. Defendants’ discriminatory conduct is ongoing, and pursuant to the remedies, 2 procedures, and rights set forth in Sections 54, 54.1, and 54.3, Plaintiffs pray for relief, as set forth 3 below. 4 SIXTH CAUSE OF ACTION 5 Declaratory Relief 6 Against All Defendants 7 249. Plaintiffs reallege and incorporate herein by this reference each of the allegations in 8 Paragraphs 1 through 193, inclusive. 9 250. An actual controversy has arisen and presently exists between Plaintiffs and 10 Defendants, because Plaintiffs contend, and Defendants dispute, that Defendants’ actions and 11 inactions as described above have violated article I, section 7(a) and article IV, section 16(a) of the 12 California Constitution; California Government Code section 11135; California Education Code 13 section 66270; the Unruh Act, California Civil Code section 51; and the California Disabled 14 Persons Act, California Civil Code sections 54 et seq. 15 16 251. Plaintiffs therefore seek a judicial declaration of each of the following: a. That Defendants’ discriminatory policy of requiring SAT and ACT test 17 scores for admission to the UC violates Plaintiffs’ rights to equal protection 18 under the California Constitution, art. I, § 7(a) and art. IV, § 16(a), by 19 discriminating on the bases of race and wealth; 20 b. That Defendants’ discriminatory policy of requiring SAT and ACT test 21 scores for admission to the UC violates Plaintiffs’ fundamental right to 22 education under the California Constitution, art. I, § 7(a) and art. IV, § 23 16(a), by denying Plaintiffs access to public higher education, and also 24 violates Plaintiffs’ fundamental right to pursue an occupation under the 25 California Constitution, art. I, § 7(a) and art. IV, § 16(a); 26 c. That Defendants’ discriminatory policy of requiring SAT and ACT test 27 scores for admission to the UC constitutes intentional discrimination against 28 members of protected classes and therefore constitutes illegal 101 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL discrimination under Section 11135 of the California Government Code; 1 d. 2 That Defendants’ discriminatory policy of requiring SAT and ACT test 3 scores for admission to the UC results in a disproportionate adverse impact 4 on protected classes and therefore constitutes illegal discrimination under 5 Section 11135 of the California Government Code; e. 6 That Defendants knowingly deprived Plaintiffs of the right of equal access 7 to educational benefits and opportunities, in violation of Section 66270 of 8 the California Education Code; f. 9 That Defendants’ discriminatory policy of requiring SAT and ACT test 10 scores for admission to the UC constitutes intentional discrimination against 11 students from underrepresented ethnic and racial backgrounds and students 12 with disabilities in violation of the Unruh Act, California Civil Code section 13 51; g. 14 That Defendant’s discriminatory policy of requiring SAT and ACT test 15 scores for admission to the UC violates the California Disabled Persons 16 Act, California Civil Code sections 54 et seq., by creating unlawful 17 admissions barriers for persons with disabilities, failing to evaluate 18 applicants on a nondiscriminatory basis, and failing to comply with civil 19 rights and access laws; and h. 20 That Defendants are required to prepare a plan, with Court approval and 21 consideration of any objections by Plaintiffs, designed to cure Defendants’ 22 violations of law, and bring them into compliance with the law. 23 REQUEST FOR RELIEF 24 Wherefore, Plaintiffs request the following relief: 25 a. Injunctive relief requiring Defendants and their officers, agents, and employees 26 to: 27 28 i. Cease requiring applicants to submit SAT or ACT scores to be considered for admission at the UC; 102 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL ii. 1 Cease using SAT or ACT scores in UC’s statewide admissions index and any other pathway to guaranteed admission; and 2 iii. 3 Cease considering SAT or ACT scores in UC admissions decisions, unless 4 Defendants are able to demonstrate UC’s ability to assess those scores in a 5 rigorous and meaningful, transparent, nondiscriminatory, and non- 6 stigmatizing manner; 7 b. A declaration that Defendants, through their actions and omissions and the 8 policies and procedures complained of, violate: i. 9 The right to equal protection under the California Constitution, art. I, § 7(a) and art. IV, § 16(a); 10 11 ii. Section 11135 of the California Government Code; 12 iii. Section 66270 of the California Education Code; 13 iv. The Unruh Act, California Civil Code §§ 51, 52; and 14 v. The California Disabled Persons Act, California Civil Code §§ 54 et seq.; 15 c. The issuance of an order requiring the preparation of a plan, with Court approval 16 and consideration of any objections by Plaintiffs, designed to cure Defendants’ violations of law, 17 and bring them into compliance with the law; 18 d. An award of costs, disbursements, and reasonable attorneys’ fees and expenses; and 19 e. Such other relief as this Court deems just and proper. 20 21 22 23 24 25 26 27 28 103 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 DATED: December 10, 2019 3 PUBLIC COUNSEL MARK ROSENBAUM ALISA HARTZ AMANDA SAVAGE By: 4 Mark Rosenbaum 2 5 6 By: 7 Alisa Hartz 8 9 10 By: 11 Amanda Savage 12 13 14 15 16 17 DATED: SCHEPER KIM & HARRIS LLP GREGORY A. ELLIS KATHERINE B. FARKAS MICHAEL LAVETTER 18 19 20 21 22 23 By: Gregory A. Ellis Attorneys for Plaintiffs 24 25 26 27 28 104 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL 1 DATED: EQUAL JUSTICE SOCIETY 2 MONA TAWATAO 3 LISA HOLDER 4 5 6 7 By: 8 Mona Tawatao 9 Attorneys for Plaintiffs 10 11 DATED: MILLER ADVOCACY GROUP 12 MARCI LERNER MILLER 13 14 15 16 17 18 By: Marci Lerner Miller Attorneys for Plaintiffs 19 20 21 22 23 24 25 26 27 28 105 PLAINTIFFS COMPLAINT AND DEMAND FOR JURY TRIAL