A.M. v. N.Y.C. Dep't of Educ.

• Court: Court of Appeals for the Second Circuit
• Decided: January 10, 2017
• Docket: 15-4076
• Status: Published

Opinion

15‐4076 A.M. v. N.Y.C. Dep’t of Educ.

UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

______________

August Term, 2016

(Argued: November 16, 2016 Decided: January 10, 2017)

Docket No. 15‐4076

_____________

A.M., individually and on behalf of E.H., a child with a disability,

Plaintiff‐Appellant,

v.

NEW YORK CITY DEPARTMENT OF EDUCATION,

Defendant‐Appellee.*

* The Clerk of Court is directed to amend the official caption to conform with the above. Before: KEARSE, WESLEY, and DRONEY, Circuit Judges. ______________ Appeal from an order of the United States District Court for the Southern District of New York (Oetken, J.), entered on December 7, 2015, granting judgment for Defendant‐Appellee New York City Department of Education (“DOE”) and denying Plaintiff‐Appellant A.M., on behalf of herself and her disabled, autistic son, E.H., relief under the Individuals with Disabilities Education Act, 20 U.S.C. §§ 1400 et seq. In denying A.M. relief, the District Court concluded that E.H. was afforded a free appropriate public education (“FAPE”) by the DOE for the 2012– 2013 school year. We disagree and hold that the DOE failed to offer E.H. a FAPE because the substance of the individualized education program formulated for E.H. was legally inadequate. Accordingly, we VACATE the judgment of the District Court and REMAND for further proceedings. ______________

JASON HALE STERNE, Cuddy Law Firm, P.C., Auburn, NY, for Plaintiff‐Appellant.

ANDREW A. FEINSTEIN, Andrew A. Feinstein, LLC, Mystic, CT, for Amicus Curiae Council of Parent Attorneys and Advocates, in support of Plaintiff‐ Appellant.

AARON M. BLOOM, Assistant Corporation Counsel (Richard Dearing and Devin Slack, on the brief), for Zachary W. Carter, Corporation Counsel, New York, NY, for Defendant‐Appellee.

2 ______________

WESLEY, Circuit Judge: Before the court is an action brought under the Individuals with Disabilities Education Act (“IDEA”), 20 U.S.C. §§ 1400 et seq., by Plaintiff‐Appellant A.M., on behalf of herself and her autistic son, E.H., against Defendant‐Appellee the New York City Department of Education (“DOE”). In May 2012, the DOE convened a meeting of the local Committee on Special Education (“CSE”) for the purpose of formulating an individualized education program (“IEP”) for E.H. for the 2012– 2013 school year. Believing the program to be inadequate for her son, A.M. continued E.H.’s enrollment at a private special education school. Subsequently, A.M. filed a due process complaint against the DOE, seeking tuition reimbursement and claiming procedural and substantive violations of the IDEA that deprived E.H. of a free appropriate public education (“FAPE”) for the 2012–2013 academic year. Following a three‐day hearing, an impartial hearing officer (“IHO”) denied A.M. that relief, and A.M. appealed that decision to a state review officer (“SRO”), who affirmed. Thereafter, A.M. brought suit in the United States District Court for the Southern District of New York (Oetken, J.), which affirmed the order of the SRO. A.M. v. N.Y.C. Dep’t of Educ., No. 14‐CV‐9224 (JPO), 2015 WL 8180751 (S.D.N.Y. Dec. 7, 2015). A.M. appealed, contending principally that the IEP formulated for E.H. violated the IDEA and deprived him of a FAPE. For the reasons set forth below, we VACATE the District Court’s judgment and REMAND for further proceedings.

3 BACKGROUND I. LEGAL FRAMEWORK “The IDEA requires New York [S]tate to ‘provide disabled children with a [FAPE].’” M.W. ex rel. S.W. v. N.Y.C. Dep’t of Educ., 725 F.3d 131, 135 (2d Cir. 2013) (quoting R.E. v. N.Y.C. Dep’t of Educ., 694 F.3d 167, 174–75 (2d Cir. 2012)). In accordance with the statute, the DOE, through a CSE,1 “must produce, in writing, an [IEP] that ‘describes the specially designed instruction and services that will enable the child to meet’ stated educational objectives and is reasonably calculated to give educational benefits to the child.” Id. (quoting R.E., 694 F.3d at 175) (citing 20 U.S.C. § 1414(d)). Where, as here, a parent believes that the program developed for his or her child for the upcoming school year would deprive the child of a FAPE, the parent may file a due process complaint with the DOE seeking review of “any matter relating to the identification, evaluation, or educational placement of the child, or the provision of a [FAPE] to such child.” 20 U.S.C. § 1415(b)(6)(A). The filing of a due process complaint “triggers an administrative procedure by which the board of education appoints an [IHO] who conducts a formal hearing and fact‐finding. The decision of an IHO may be appealed to [an SRO], and an SRO’s decision may be challenged by filing a civil action in state or federal court.” M.O. v. N.Y.C.

1 “In New York, the state has assigned responsibility for developing IEPs to local [CSEs]. CSEs are comprised of members appointed by the local school district’s board of education, and must include the student’s parent(s), a regular or special education teacher, a school board representative, a parent representative, and others.” R.E., 694 F.3d at 175 (citation omitted) (citing N.Y. Educ. Law § 4402(1)(b)(1)(a)).

4 Depʹt of Educ., 793 F.3d 236, 239 (2d Cir. 2015) (per curiam) (citations omitted) (quoting Hardison v. Bd. of Educ. of the Oneonta City Sch. Dist., 773 F.3d 372, 376 (2d Cir. 2014)) (citing, inter alia, 20 U.S.C. §§ 1415(g), (i)(2)(A); N.Y. Educ. Law §§ 4404(1)–(3)). II. STATEMENT OF FACTS A. The Student ‐ E.H. In May 2012, at the time the CSE convened, E.H. was a six‐year‐old boy diagnosed with autism. E.H. suffers from global developmental delays, significantly impaired communication and social functioning, and substantial language impairments. He also exhibits physical stereotypy (e.g., “[t]apping on surfaces or part of the body”) and vocal stereotypy (i.e., “palilalia,” which is characterized by “[n]on contextual vocalizations or vocalizations emitted in a high pitched tone”). Suppl. App. 873–74. Before E.H. turned five years of age, he attended a private preschool, where he was placed in a special education class with a student‐teacher‐paraprofessional ratio of 6:1:1. Thereafter, once E.H. transitioned to school age, he attended a public community school, where he was placed in a 12:1:1 special education classroom. In 2011, E.H. began attending Manhattan Children’s Center (“MCC”), a private special education school, after A.M. determined that the IEP produced by the DOE for the 2011–2012 school year denied E.H. a FAPE. At MCC, E.H. received schooling in a classroom with six students and six teachers (i.e., a student‐teacher ratio of 1:1), and received, among other things, applied behavioral analysis (“ABA”) therapy, “which is an intensive one‐on‐one therapy that ‘involves breaking down activities into discrete tasks and rewarding a child’s accomplishments.’” R.E., 694 F.3d at 176 (quoting Cty. Sch. Bd. v. Z.P. ex rel. R.P., 399 F.3d 298, 301 (4th Cir. 2005)).

5 In 2011, A.M. filed a due process complaint against the DOE seeking tuition reimbursement for her unilateral placement of E.H. at MCC and recovery of “make up sessions” of physical therapy and instruction time missed during the summer of 2011, for which no educational program was offered to E.H. Suppl. App. 555. An IHO found in A.M.’s favor, concluding that it “c[ould] discern no consistent effort in administering the education of [E.H.],” thereby depriving E.H. of a FAPE. Suppl. App. 559. In addition, the IHO determined that the special education program at MCC in which A.M. unilaterally placed E.H. was appropriately tailored to “me[e]t [E.H.’s] special education needs,” Suppl. App. 560, and that A.M. was entitled to the requested relief, including tuition reimbursement. That decision was uncontested by the parties. B. E.H.’s May 2012 Individualized Education Program In May 2012, a CSE team convened for its annual meeting to develop an IEP for E.H. for the 2012–2013 academic year. The CSE team was composed of the following members: (1) Nessan O’Sullivan, the School District representative and a DOE school psychologist; (2) Judy Sommers Schneid, a DOE special education teacher; (3) E.H.’s mother, A.M.; (4) parent member Marie Wise; (5) Samantha Solow, an assistant educational coordinator at MCC; (6) Marisa Savard, one of E.H.’s lead teachers at MCC; (7) Carrie Friedman, E.H.’s occupational therapist at MCC; and (8) Amy Hunt, E.H.’s speech‐language pathologist at MCC. Prior to the CSE meeting, neither O’Sullivan nor Schneid met with E.H., nor did the DOE conduct any evaluation of E.H. on its own, aside from a psycho‐educational report from March 2010 conducted as part of a mandatory review some two years before the CSE convened to formulate E.H.’s IEP for the 2012–

6 2013 school year. Rather, the evaluative materials present at the meeting were produced largely by MCC educators and medical professionals who had evaluated or treated E.H. Based on the recommendations of the DOE representatives, who relied on the evaluative materials present at the CSE meeting prepared by others, and over the objections of A.M. and MCC staff, the IEP recommended a twelve‐month placement in a public special education program with a student‐teacher‐paraprofessional ratio of 6:1:1. Moreover, despite A.M.’s and MCC staff’s urging that the DOE incorporate ABA therapy into the IEP, the IEP was silent as to the use of any particular methodology for treating A.M.’s educational needs. The IEP did, however, recommend that A.M. receive a number of related services: (1) three weekly, thirty‐minute sessions of individual speech‐language therapy; (2) one weekly, thirty‐minute session of group speech‐language therapy; (3) two weekly, thirty‐minute sessions of individual physical therapy; (4) two weekly, thirty‐minute sessions of individual occupational therapy; and (5) one weekly, thirty‐ minute session of group occupational therapy. The IEP also recommended that E.H. be equipped with a hand‐held tablet computer with specialized software and “voice output in order to increase his independence and social interactions and to serve as a model for language.” Suppl. App. 867, 904. Because the CSE team concluded that E.H.’s behaviors seriously interfered with instruction, the IEP required the development of a behavioral intervention plan (“BIP”), which was incorporated into the IEP. The BIP, which “is generally used to ‘develop[] . . . strategies to deal with . . . problem behavior(s),’” L.O. v. N.Y.C. Dep’t of Educ., 822 F.3d 95, 104 (2d Cir. 2016) (alterations in original) (quoting R.E., 694 F.3d at 190), identified E.H.’s physical and vocal stereotypy and noted that it sought to achieve a “[s]ignificant decrease in target behaviors”

7 through “[t]ime sampling and partial interval recording,” Suppl. App. 873. Although the BIP was based on a functional behavior assessment (“FBA”), which was developed and incorporated into the IEP, the DOE did not conduct an FBA of E.H. An FBA is “an assessment designed to ‘identif[y] . . . the problem behavior, . . . defin[e] . . . the behavior in concrete terms, . . . identif[y] . . . the contextual factors that contribute to the behavior (including cognitive and affective factors) and . . . formulat[e] . . . a hypothesis regarding the general conditions under which a behavior usually occurs.’” L.O., 822 F.3d at 104 (alterations in original) (quoting N.Y. Comp. Codes R. & Regs. tit. 8, § 200.1(r)). In this case, the FBA was “based on . . . the draft [FBA] submitted by [MCC]” to the CSE. Suppl. App. 216. Further, E.H.’s IEP did not provide for parental counseling or training services, nor did it include any transitional support services for E.H. to change schools and programs. C. Administrative Review Upon receipt of a final notice of recommended placement, A.M. advised the DOE that she disagreed with E.H.’s recommended placement and that E.H. would continue his education at MCC for the 2012–2013 academic year. Thereafter, A.M. filed a due process complaint, claiming that the DOE had failed to provide E.H. with a FAPE for the 2012–2013 school year, and she consequently sought private school tuition reimbursement at MCC. As relevant here, A.M. alleged four separate deficiencies in the IEP developed for E.H.: (1) the DOE failed to conduct an FBA and failed to develop an appropriate BIP; (2) the IEP failed to provide parent counseling and training as a related service; (3) the IEP failed to offer transition services; and (4) the IEP failed to provide a placement in a 1:1 classroom that “implement[ed] an [ABA] or substantially similar methodology.” Suppl. App. 571–73.

8 1. Impartial Hearing Officer’s Decision Following a three‐day hearing,2 the IHO denied A.M.’s reimbursement request. The IHO rejected A.M.’s contention that the FBA and BIP developed for E.H. deprived him of a FAPE. Although the FBA for E.H. that was contained in the IEP was based entirely on the FBA prepared by MCC, the IHO determined that this did not result in a FAPE deprivation because “[t]he FBA prepared by the CSE t[ook] the information from the FBA prepared by MCC and use[d] it appropriately.” Suppl. App. 37. The IHO observed that the FBA appropriately listed the target behaviors it sought to eradicate, identified their root causes, and developed adequate strategies to address those problem behaviors. The IHO also concluded that, although no services were included in E.H.’s IEP to assist with his transition from MCC to the proposed public school placement, the DOE did not deprive E.H. of a FAPE. The IHO relied on hearing testimony that E.H. generally did not struggle with transitioning between classrooms in the past, and also pointed to testimony from O’Sullivan, that the DOE “felt that the services that we[re] recommended would enable [E.H.] to make the transition from where he presently was to a public school setting.” Suppl. App. 168.

2 A.M. testified and offered testimony from four members of MCC’s staff: (1) assistant educational coordinator Solow; (2) speech‐language pathologist Adina Haimes; (3) occupational therapist Friedman; and (4) lead teacher Nicole Herz. The DOE offered testimony from O’Sullivan, the DOE school psychologist who also served as the DOE representative on the CSE that formulated the terms of E.H.’s IEP.

9 The IHO additionally considered the IEP’s recommendation of a 6:1:1 classroom and found it to be appropriate. The IHO was “persuaded that . . . O’Sullivan’s observation that [E.H.] need[ed] the interaction with peers would argue for a classroom setting where there [was] an abundance of interaction, rather than primarily one to one instruction.” Suppl. App. 36–37. The IHO also noted that E.H.’s attention had “vastly improved” and that the child was “less in need of that intervention,” and that, in any event, even if poor attending skills persisted, this would not demand one‐to‐one attention, as such skills could be adequately addressed in a 6:1:1 class. Suppl. App. 37. Finally, the IHO rejected A.M.’s view that E.H. required all‐day intensive ABA therapy. In rejecting this argument, the IHO noted that “[t]here [was] no requirement that a particular methodology be set forth on the IEP or provided to [a] [s]tudent, or that the preference as to [the] methodology of any particular evaluator . . . or [p]arent[] be adopted.” Suppl. App. 36. It thus concluded “that the methodology should be left to the discretion of the professionals who w[ould] be working with [E.H.].” Suppl. App. 35. Although the IHO concluded that the IEP complied with the IDEA, the IHO did not address A.M.’s contention that the omission of parental training and counseling services in the IEP deprived E.H. of a FAPE. 2. State Review Officer’s Decision A.M. appealed the IHO’s decision to an SRO, who affirmed. As to the lack of parent counseling and training services in the IEP, the SRO faulted the IHO for “not addressing [its] absence” in the IEP, but nonetheless determined that, although the IEP should have provided for such services, its

10 omission did not result in a FAPE deprivation. Suppl. App. 14 n.3. It based this conclusion on IHO hearing testimony that “‘it was explained to the parent at the outset’ that parent counseling and training was one of the ‘defining [features]’ of the recommended [6:1:1] special class placement.” Suppl. App. 14 n.3 (quoting Suppl. App. 167–68). Further, the SRO reasoned that “the presence or absence of parent training and counseling in an IEP does not necessarily have a direct effect on the substantive adequacy of the plan.” Suppl. App. 14 n.3 (citing R.E., 694 F.3d at 191). Next, the SRO rejected A.M.’s claim that E.H. required a 1:1 classroom with ABA therapy in order to progress and was denied a FAPE when he was assigned to a 6:1:1 classroom that lacked an all‐inclusive ABA program. As to the adequacy of the class size itself, the SRO credited O’Sullivan’s testimony that “the [6:1:1] special class . . . was an ‘intens[ive] behavioral intervention,’” in that it was a rigorous full‐time program in a small special education classroom that employed “a variety of methodologies” to treat each student’s educational needs. Suppl. App. 15 (quoting Suppl. App. 231). The SRO also found persuasive O’Sullivan’s testimony that “he believed [E.H.] could make meaningful progress had the May 2012 IEP been implemented as written because [E.H.’s] special education services needed to be delivered within some social context to allow for modeling and shared instruction.” Suppl. App. 15–16. Moreover, the SRO found A.M.’s methodology argument equally unavailing because, in the SRO’s view, a CSE is generally “not required to specify methodology on an IEP, and the precise teaching methodology to be used by a student’s teacher is usually a matter to be left to the teacher.” Suppl. App. 15. The SRO reasoned that, “although the May 2012 CSE did not recommend ABA [therapy services], it also did not disapprove

11 of the methodology or say ABA should not be used with the student.” Suppl. App. 15. The SRO also observed that O’Sullivan, “[t]he [DOE] school psychologist[,] indicated [that] the May 2012 CSE recommendations incorporated behavioral methodologies, and the [6:1:1] special class recommendation was not contrary to ABA.” Suppl. App. 15. Thus, the SRO concluded that the absence of ABA therapy in E.H.’s IEP did not deprive the child of a FAPE. The SRO did not, however, address A.M.’s challenge to the adequacy of E.H.’s FBA and BIP or the lack of transitional services contained in the IEP. D. District Court Review Thereafter, A.M. brought suit in the District Court, claiming procedural and substantive violations under the IDEA, resulting in the deprivation of a FAPE for E.H. The parties each separately moved for summary judgment, and the District Court granted the DOE’s motion, affirming the SRO’s decision and denying A.M.’s claim for tuition reimbursement. A.M., 2015 WL 8180751, at *6. As to the procedural adequacy of the IEP, the District Court considered and denied A.M.’s challenge to the adequacy of the FBA and BIP contained in the IEP, because the FBA and BIP incorporated into the IEP were “[c]onsistent with the MCC FBA/BIP,” and “adequately identifie[d] the problem behavior and prescribe[d] ways to manage it.” Id. at *4 (quoting R.E., 694 F.3d at 190). The District Court rejected A.M.’s claim that, because the CSE had inadequately reproduced the MCC draft on which the IEP’s FBA and BIP were based, E.H. was deprived of a FAPE. The District Court reasoned that, “so long as the [DOE’s] own FBA/BIP is adequate, it does not matter that the [DOE’s] draft omitted MCC’s charts, citations to academic support, or

12 other information.” Id. The District Court thus concluded that the FBA and BIP were suitable to E.H.’s needs. Id. The District Court also concluded that the failure to provide parent counseling and training in the IEP, although a violation of New York law, was not fatal to the adequacy of the IEP. The District Court observed that “it was explained to the parent at the outset that one of the defining features of this program was that it did have a parent training component to it as well, as part of the program,” id. (quoting Suppl. App. 167–68), and thus A.M. was unable to demonstrate in what manner either she or her son were injured as a result of the omission of parental counseling and training services in the IEP. In addition, the District Court rejected A.M.’s contention that the lack of a formal transition plan for E.H. into his new public school program deprived E.H. of a FAPE, because “A.M. ha[d] ‘not identified any legal requirement that an IEP contain a transition plan, nor ha[d] [A.M.] articulated why the absence of such a plan was so significant as to deny [him]’ a [FAPE].” Id. at *3 (third alteration in original) (quoting R.E., 694 F.3d at 195). The District Court then turned to the IEP’s substantive adequacy and rejected A.M.’s claim that E.H. required a 1:1 classroom that guaranteed ABA therapy all day, because “[t]hat dispute is a debate over educational methodology, based on the record before the SRO,” in which case “courts typically defer to the decision of the SRO.” Id. at *5 (citing M.H. v. N.Y.C. Dep’t of Educ., 685 F.3d 217, 244 (2d Cir. 2012)). The District Court noted that this Court had held in R.E. that, “where ‘almost all of the reports’ before an IHO find that a student needs ABA therapy, and the remaining evidence emphasizes that the student needs ‘a high level of support,’ the [DOE] may not endorse a 6:1:1 classroom with ‘no guarantee of ABA therapy or any meaningful 1:1 support.’” Id. (quoting R.E., 694 F.3d at 193–94). The District

13 Court distinguished E.H.’s case from R.E., however, because this was not, in its view, “a case . . . in which the [DOE] ignored the ‘clear consensus of [the student’s] evaluators.’” Id. (quoting R.E., 694 F.3d at 181). Rather, according to the District Court, although “the SRO and IHO deviated from the consensus of the parent’s evidence, including the representatives from MCC,” id., “their conclusion was consistent with the views of the evaluator on the CSE,” id., who believed that “a 6:1:1 setting would better facilitate [E.H.’s] goals,” id., particularly “developing reciprocity, developing an awareness of peers, and maneuvering himself within the context of other children within a classroom setting,” id. (quoting Suppl. App. 16). Thus, according to the District Court, “the CSE may come to an independent conclusion of the student’s needs from the content of the evaluation reports; the CSE need not simply accept the conclusion of those reports absent an in‐person evaluation with the student.” Id. (emphasis in original). As to the IEP’s silence on the provision of ABA therapy, the District Court observed that, although “[t]he 6:1:1 setting did not expressly include ABA, . . . it also did not exclude ABA.” Id. The District Court credited O’Sullivan’s IHO hearing testimony that the DOE deliberately formulated the IEP in this manner because it “‘didn’t want to tie the hands of the different disciplines that would be working with’ E.H. by requiring a specific methodology for part or all of the day.” Id. (quoting Suppl. App. 168). The District Court also agreed with the IHO’s determination that MCC might have preferred a different methodology, but this preference “d[id] not have to guide the decisions of the CSE.” Id. (quoting Suppl. App. 36).

14 DISCUSSION “We undergo a circumscribed de novo review of a district court’s grant of summary judgment in the IDEA context because the ‘responsibility for determining whether a challenged IEP will provide a child with [a FAPE] rests in the first instance with administrative hearing and review officers.’” M.W., 725 F.3d at 138 (alteration in original) (quoting M.H., 685 F.3d at 240). Our review “requires a more critical appraisal of the agency determination than clear‐error review . . . but . . . nevertheless[] falls well short of complete de novo review.” M.H., 685 F.3d at 244 (alterations in original) (quoting Lenn v. Portland Sch. Comm., 998 F.2d 1083, 1086–87 (1st Cir. 1993)). “Accordingly, our de novo review only seeks to independently verify that the administrative record supports the district court’s determination that a student’s IEP was adequate.” M.W., 725 F.3d at 138. This verification requires that we “engage in an independent review of the administrative record and make a determination based on a ‘preponderance of the evidence.’” Gagliardo v. Arlington Cent. Sch. Dist., 489 F.3d 105, 112 (2d Cir. 2007) (quoting Mrs. B. v. Milford Bd. of Educ., 103 F.3d 1114, 1120 (2d Cir. 1997)). Importantly, our review of the administrative record is “restrained by our lack of specialized knowledge and educational expertise,” which demands that we “defer to the administrative decision [particularly where] the state officer’s review has been thorough and careful.” M.W., 725 F.3d at 138–39 (alteration in original) (internal quotation marks omitted) (quoting R.E., 694 F.3d at 184). Indeed, the Supreme Court has made clear that we are precluded from “substitut[ing] [our] own notions of sound educational policy for those of the school authorities which [we] review.” Bd. of Educ. of Hendrick Hudson Cent. Sch. Dist. v. Rowley, 458 U.S. 176, 206 (1982). “The level of deference granted to the administrative decision, however, is not

15 without limitation. To merit deference, ‘[t]he SRO’s or IHO’s factual findings must be reasoned and supported by the record.’” L.O., 822 F.3d at 109 (internal quotation marks omitted) (quoting M.H., 685 F.3d at 241). Further, although “courts must defer to the reasoned conclusions of the SRO as the final state administrative determination,” should we “f[in]d the SRO’s conclusions unpersuasive even after appropriate deference is paid, [we may] consider the IHO’s analysis, which is also informed by greater educational expertise than that of judges, rather than . . . rely exclusively on [our] own less informed educational judgment.” M.H., 685 F.3d at 246. I. ISSUES FOR JUDICIAL REVIEW In order to determine whether parents of a disabled child are entitled to reimbursement of expenses incurred at a private school in an IDEA challenge to a state‐proposed IEP, we are guided by the three‐step Burlington/Carter test: “(1) the DOE must establish that the student’s IEP actually provided a FAPE; should the DOE fail to meet that burden, the parents are entitled to reimbursement if (2) they establish that their unilateral placement was appropriate and (3) the equities favor them.” M.W., 725 F.3d at 135 (footnote omitted); see also Florence Cty. Sch. Dist. Four v. Carter, 510 U.S. 7 (1993); Sch. Comm. of Town of Burlington v. Dep’t of Educ., 471 U.S. 359 (1985). Here, the parties’ dispute focuses entirely on the adequacy of E.H.’s May 2012 IEP. In reviewing the adequacy of the IEP, we “make a two‐part inquiry that is, first, procedural, and second, substantive,” to determine whether an IEP complies with the IDEA, R.E., 694 F.3d at 189–90, the burden of proof resting with the DOE, see N.Y. Educ. Law § 4404(1)(c).

16 A. Procedural Compliance At step one, we “examine whether there were procedural violations of the IDEA, namely, ‘whether the state has complied with the procedures set forth in the IDEA.’” R.E., 694 F.3d at 190 (quoting Cerra v. Pawling Cent. Sch. Dist., 427 F.3d 186, 192 (2d Cir. 2005)). “The initial procedural inquiry is no mere formality,” Walczak v. Fla. Union Free Sch. Dist., 142 F.3d 119, 129 (2d Cir. 1998), as “[i]t acts as ‘a safeguard against arbitrary or erroneous decisionmaking,’” M.H., 685 F.3d at 245 (internal quotation marks omitted) (quoting Evans v. Bd. of Educ. of Rhinebeck Cent. Sch. Dist., 930 F. Supp. 83, 93 (S.D.N.Y. 1996) (B.D. Parker, J.)). A procedural violation will entitle a parent to tuition reimbursement “if [it] ‘impeded the childʹs right to a [FAPE],’ ‘significantly impeded the parents’ opportunity to participate in the decisionmaking process,’ or ‘caused a deprivation of educational benefits.’” R.E., 694 F.3d at 190 (quoting 20 U.S.C. § 1415(f)(3)(E)(ii); A.C. ex rel. M.C. v. Bd. of Educ. of the Chappaqua Cent. Sch. Dist., 553 F.3d 165, 172 (2d Cir. 2009)). “That is, parents must articulate how a procedural violation resulted in the IEP’s substantive inadequacy or affected the decision‐making process.” M.W., 725 F.3d at 139. “[M]ultiple procedural violations[,] [however,] may cumulatively result in the denial of a FAPE even if the violations considered individually do not.” L.O., 822 F.3d at 109 (alterations in original) (quoting M.W., 725 F.3d at 139). Here, A.M. alleges three procedural errors in the formulation of E.H.’s IEP, which, in her view, independently and cumulatively deprived E.H. of a FAPE.

17 1. Functional Behavior Assessment and Behavioral Intervention Plan New York regulations require the DOE to conduct an FBA “for a student whose behavior impedes his or her learning or that of others.” N.Y. Comp. Codes R. & Regs. tit. 8, § 200.4(b)(1)(v). An FBA must include “the identification of the problem behavior, the definition of the behavior in concrete terms, the identification of the contextual factors that contribute to the behavior[,] . . . and the formulation of a hypothesis regarding the general conditions under which a behavior usually occurs and probable consequences that serve to maintain it.” Id. § 200.1(r). Its purpose “is to ensure that the IEP’s drafters have sufficient information about the student’s behaviors to craft a plan that will appropriately address those behaviors.” R.E., 694 F.3d at 190. In addition, “where, as here, ‘a student’s behavior impedes his learning, a BIP must be developed with strategies to deal with the problem behavior(s).’” L.O., 822 F.3d at 111 (quoting R.E., 694 F.3d at 190) (citing N.Y. Comp. Codes R. & Regs. tit. 8, § 200.22(b)). New York regulation requires that a BIP be “based on the results of a[n] [FBA] and, at a minimum, include[] a description of the problem behavior, global and specific hypotheses as to why the problem behavior occurs and intervention strategies that include positive behavioral supports and services to address the behavior.” N.Y. Comp. Codes R. & Regs. tit. 8, § 200.1(mmm). Where the DOE fails to conduct an FBA of its own, we “must take particular care to ensure that the IEP adequately addresses the child’s problem behaviors,” R.E., 694 F.3d at 190, but the IEP will be rendered legally deficient only where it fails to “adequately identif[y] a student’s behavioral impediments and [fails to] implement[] strategies to address that behavior,”

18 M.W., 725 F.3d at 140. Where the IEP is found to be deficient in this regard, we have held this to rise to the level of “a serious procedural violation because it may prevent the CSE from obtaining necessary information about the student’s behaviors, leading to their being addressed in the IEP inadequately or not at all.” R.E., 694 F.3d at 190. “[S]uch a failure seriously impairs substantive review of the IEP because courts cannot determine exactly what information an FBA would have yielded and whether that information would be consistent with the student’s IEP.” Id. Here, the IEP indicated that E.H.’s behavior seriously interfered with his instruction and therefore required the development of a BIP. As noted, a BIP must be based on the results of an FBA. See N.Y. Comp. Codes R. & Regs. tit. 8, § 200.1(mmm). Rather than conduct his own evaluation of E.H., however, O’Sullivan, the School District representative and DOE psychologist present at the May 2012 CSE meeting, relied on an October 2011 FBA and BIP performed by MCC and submitted to the DOE prior to the CSE meeting in order to develop E.H.’s FBA and BIP for the IEP. A.M. contends that the DOE’s failure to conduct an FBA of its own and develop a BIP from its own assessment of E.H. constitutes a serious procedural violation of the IDEA, which resulted in a FAPE deprivation. She argues further that the BIP and FBA incorporated into E.H.’s IEP lacked certain information contained in the draft versions formulated by MCC. For instance, according to A.M., the “BIP contains no intervention strategies that include positive behavioral supports and services to address the behavior,” and the FBA lacks any “information about the use of functional communication training[,] intensive tact instruction[,] and matched stimuli.” Pl.’s Br. 27 (internal quotation marks omitted). Thus, for A.M., the FBA and BIP

19 incorporated into E.H.’s IEP were inadequate and were thus not reasonably calculated to afford E.H. an opportunity to obtain educational benefits, resulting in a denial of a FAPE. We note at the outset that A.M.’s argument is not without some persuasive force. O’Sullivan described the FBA included in the IEP as more of “a summary of [MCC’s] behavior assessment with the consensus of the meeting at the time,” and stated that “maybe 10–15 minutes” was spent by the CSE creating this summary and developing the BIP at the meeting. Suppl. App. 220–21. Indeed, the barebones nature of the BIP and FBA included in the IEP makes this apparent, particularly when compared with the BIP and FBA produced by MCC. This, of course, is not surprising, given that O’Sullivan never actually met or evaluated E.H. until after the formulation of his IEP, a violation of New York law. See N.Y. Comp. Codes R. & Regs. tit. 8, § 200.4(b)(1)(v) (“[A]n individual evaluation of the referred student shall be initiated by a [CSE] . . . [and] shall be completed within 60 days of receipt of consent . . . . [T]he initial evaluation must include at least . . . a[n] [FBA] for a student whose behavior impedes his or her learning or that of others, as necessary to ascertain the physical, mental, behavioral and emotional factors which contribute to the suspected disabilities.”). As to the BIP incorporated into E.H.’s IEP, we have no difficulty in concluding that it is legally inadequate. Although it was based on the results of MCC’s FBA and includes a description of the problem behavior, it does not provide certain “minimum requirements,” L.O., 822 F.3d at 123, such as “global and specific hypotheses as to why the problem behavior occurs and intervention strategies that include positive behavioral supports and services to address the behavior,” N.Y. Comp. Codes R. & Regs. tit. 8, § 200.1(mmm). Indeed, we recently expressed concern in situations such as this where the IEP

20 contains a BIP that fails to “attempt[] to identify the root causes of . . . behavior deficiencies,” because it deprives the CSE of the ability to adequately “provide effective treatment for these behaviors,” “thereby casting doubt on the adequacy of its provisions for treating them.” L.O., 822 F.3d at 113, 123. Although deficient, we have also stressed that the inadequacy of an FBA or BIP will not necessarily “render an IEP legally inadequate under the IDEA so long as the IEP adequately identifies a student’s behavioral impediments and implements strategies to address that behavior.” M.W., 725 F.3d at 140. Here, the FBA, which was also incorporated into E.H.’s IEP, included the required information that was lacking in the BIP. The FBA hypothesizes that E.H.’s palilalia and physical stereotypy are caused by “[l]ack of active engagement” and that E.H. reacts in such a manner because of “[s]ensory stimulation,” “task avoidance,” and “[a]ccess to preferred attention or toy.” Suppl. App. 874. The FBA further lists several strategies previously attempted, such as the use of “[r]edirection and refocus,” “[r]eplac[ing] verbal stereotyp[y] with communication skills,” and “[e]xpand[ing] [E.H.’s] repertoire of reinforcer and play skills,” and suggests continuing the use of such strategies moving forward. Suppl. App. 874. In addition, the FBA lists positive behavioral techniques that have proven to be successful in the past, including providing E.H. with “[a]ccess to certain toys[,] action figures[,] computer, dinosaur related toys and activities.” Suppl. App. 874. Last, the FBA also identifies the end goal, seeking to “decrease [E.H.’s] verbal and motor stereotyp[y],” and notes that it would measure E.H’s progress and the success of these strategies through “[p]artial interval recording” and “time sampling.” Suppl. App. 874. These descriptions “adequately identif[y] the problem behavior[s] and

21 prescribe[] ways to manage [them],” in accordance with New York law. R.E., 694 F.3d at 190. Although there are undeniable differences between the FBA and BIP included in E.H.’s IEP and the far more detailed and thorough FBA and BIP created by MCC, E.H.’s IEP incorporated the overall material substance of the MCC FBA and BIP, albeit in a more limited form. That the versions ultimately prepared and incorporated into E.H.’s IEP were less comprehensive than those developed by E.H.’s current private school education provider does not mean that E.H. was deprived of a FAPE. Indeed, aside from pointing to certain differences between MCC’s and the DOE’s FBAs and BIPs (such as the lack of any mention of E.H.’s history of crying, which has been successfully treated, at times, through redirection and immediate return to activity), A.M. does not describe how such omissions rise to the level of a FAPE deprivation; that is, in what manner E.H.’s ability to meaningfully progress would be delayed as a result of the deficiencies identified in the IEP’s FBA and BIP. Accordingly, although the BIP developed for E.H. is deficient in a number of important respects, when viewed together with the FBA, the IEP nonetheless “adequately identifies [E.H.’s] behavioral impediments and implements strategies to address th[ose] behavior[s].” M.W., 725 F.3d at 140. Thus, E.H. was not deprived of a FAPE as a result of the FBA and BIP that were included in his IEP. 2. Parental Counseling and Training Next, A.M. contends that the DOE’s failure to provide for parental counseling and training services in E.H.’s IEP constituted a violation of the procedures of the IDEA, thereby contributing to a FAPE denial. “For educational programs for students with autism, New York requires that an IEP include a

22 ‘[p]rovision . . . for parent counseling and training . . . for the purpose of enabling parents to perform appropriate follow‐up intervention activities at home.’” L.O., 822 F.3d at 122 (alterations in original) (quoting N.Y. Comp. Codes R. & Regs. tit. 8, § 200.13(d)). The inclusion of such services in the IEP is intended to “assist[] parents in understanding the special needs of their child; provid[e] parents with information about child development; and help[] parents to acquire the necessary skills that will allow them to support the implementation of their child’s [IEP].” N.Y. Comp. Codes R. & Regs. tit. 8, § 200.1(kk). Although a violation of New York’s regulations, “[w]e have repeatedly held . . . that parental counseling and training omissions are ‘less serious’ procedural violations ‘because the presence or absence of a parent‐counseling provision does not necessarily have a direct effect on the substantive adequacy of the [IEP].’” L.O., 822 F.3d at 122 (second alteration in original) (quoting M.W., 725 F.3d at 141). This is so “because school districts are required . . . to provide parent counseling,” and thus “remain accountable for their failure to do so no matter the contents of the IEP.” Id. (alteration in original) (quoting R.E., 694 F.3d at 191). Here, the SRO found that the DOE erred by failing to provide for parent counseling and training in the IEP, in violation of New York law. It appears that this may not be an anomaly. If the DOE’s failure to comply with § 200.13(d) in formulating the terms of IEPs for students with autism has become a common practice, such a development would be cause for concern. See, e.g., L.O., 822 F.3d at 122; J.C. v. N.Y.C. Dep’t of Educ., 643 F. App’x 31, 32 (2d Cir. 2016) (summary order); R.B. v. N.Y.C. Dep’t of Educ., 603 F. Appʹx 36, 39 (2d Cir. 2015) (summary order); C.F. ex rel. R.F. v. N.Y.C. Dep’t of Educ., 746 F.3d 68, 79–80 (2d Cir. 2014); F.L. ex rel. F.L. v. N.Y.C. Dep’t of Educ., 553 F.

23 App’x 2, 7 (2d Cir. 2014) (summary order); M.W., 725 F.3d at 142; K.L. ex rel. M.L. v. N.Y.C. Dep’t of Educ., 530 F. App’x 81, 87 (2d Cir. 2013) (summary order); P.K. ex rel. S.K. v. N.Y.C. Dep’t of Educ., (Region 4), 526 F. App’x 135, 138 (2d Cir. 2013) (summary order); R.E., 694 F.3d at 193, 195. The SRO nonetheless concluded, in accordance with this Court’s caselaw, that this omission did not result in a FAPE deprivation. See R.E., 694 F.3d at 191, 192–93. Importantly, the SRO also observed that, at the CSE meeting, “‘it was explained to the parent at the outset’ that parent counseling and training was one of the ‘defining [features]’ of the recommended [6:1:1] special class placement.” Suppl. App. 14 n.3 (quoting Suppl. App. 167–68). We defer to that analysis. While its absence from the IEP constitutes a violation of the procedures of the IDEA, evidence that E.H.’s program actually offered parental counseling and training services and that A.M. was made aware of its presence at the CSE meeting makes clear that the omission of parental counseling and training in the IEP itself was nothing more than an immaterial procedural violation. See L.O., 822 F.3d at 123. 3. Transitional Support Services A.M. next argues that E.H. required support services to assist in his transition from his private school classroom at MCC to the less restrictive public school placement in his IEP. A.M. contends that the inclusion of such transitional services in the IEP was compelled by N.Y. Comp. Codes R. & Regs. tit. 8, § 200.13(a)(6). She maintains that, “[a]s a student with autism, [E.H.’s] resistance to environmental change or change in daily routines . . . require[d] transitional support,” and that the DOE’s “failure to address [his] transitional needs [was] likely to lead to regression in his behaviors, and thus in his educational performance.” Pl.’s Br. 25. Thus, in A.M.’s view, because the IEP was devoid of transitional support services for E.H., as required

24 by New York law, E.H. was deprived of a FAPE. The DOE, on the other hand, insists that there is “no[] . . . ‘legal requirement that an IEP contain a transition plan,’” and even if there were, “A.M. did not ‘articulate[] why the absence of such a plan was so significant as to deny’ E.H. an appropriate education.” Def.’s Br. 42 (quoting R.E., 694 F.3d at 195). Although we agree that the absence of a transition plan or services to assist with E.H.’s transition between classrooms did not deprive him of a FAPE, there is some confusion here that bears further explanation. N.Y. Comp. Codes R. & Regs. tit. 8, § 200.13(a)(6), provides that,

[i]n those instances where a student [with autism] has been placed in programs containing students with other disabilities, or in a regular class placement, a special education teacher with a background in teaching students with autism shall provide transitional support services in order to assure that the student’s special education needs are being met.

N.Y. Comp. Codes R. & Regs. tit. 8, § 200.13(a)(6). Although the IHO made no such finding,3 the District Court concluded that there was “no[] . . . legal requirement that an IEP contain a transition plan.” A.M., 2015 WL 8180751, at *3 (quoting R.E., 694 F.3d at 195). While this might be true, the District Court nonetheless failed to appreciate that there remained a violation of § 200.13(a)(6). That is, although the IDEA does not require the inclusion of a formal transition plan in an autistic student’s IEP

3 As noted, the SRO failed to address A.M.’s transition‐plan argument in its decision.

25 that describes services to assist with the child’s transition into a public school classroom or less restrictive educational setting, New York law imposes a separate requirement that, under such conditions, the IEP describe support services to be provided to the student’s public special education teacher in the classroom, which were absent from E.H.’s IEP. See N.Y. Comp. Codes R. & Regs. tit. 8, § 200.1(ddd) (“Transitional support services means those temporary services, specified in a student’s [IEP], provided to a regular or special education teacher to aid in the provision of appropriate services to a student with a disability transferring to a regular program or to a program or service in a less restrictive environment.” (emphases added)); id. § 200.6(c) (“Transitional support services prescribed in the [IEP] shall be provided for a teacher upon the recommendation of the [CSE], and shall be specified in the student’s IEP.” (emphasis added)). Just as with an omission of parental training and counseling services in the IEP, which concerns services that are to be delivered to an individual other than the child itself, the absence of such services that are expressly required by law to be included in a child’s IEP constitutes a violation of the procedures of the IDEA. Similarly, school districts are required by § 200.13(a)(6) to “provide transitional support services,” id. § 200.13(a)(6), to an autistic student’s special education teacher “in order to assure that the student’s special education needs are being met,” id., “no matter the contents of the IEP,” and they thus “remain accountable for their failure to do so,” permitting a parent of an aggrieved child to “file a complaint at any time if they feel the[ir] [child is] not receiving this service.” R.E., 694 F.3d at 191 (citing N.Y. Comp. Codes R. & Regs. tit. 8, § 200.13(d)). Importantly, these regulations contemplate that, where, as here, “a student with a disability transfer[s] to a regular program

26 or to a program or service in a less restrictive environment,” the student’s new classroom teacher may require “temporary services” to ensure that the student’s particular educational needs are being adequately treated in the new learning environment. N.Y. Comp. Codes R. & Regs. tit. 8, § 200.1(ddd). This makes sense, as a classroom teacher may not necessarily be equipped or trained at the outset to treat certain conditions or behaviors that are unique to the new student joining his or her classroom. However, as with the inclusion of any other provision in the IEP, the ultimate beneficiary of its presence, or the individual harmed by its absence, is the student, whose likelihood to progress under the proposed program becomes markedly less certain as terms and conditions imposed by law are wanting from the IEP. We note that this view is not in tension with our holding in R.E., which held only that there was “no[] . . . legal requirement that an IEP contain a transition plan”; that is, that there was nothing in the law that required a description in the IEP of transitional support services to be provided directly to the student. See R.E., 694 F.3d at 195. We do not purport to suggest otherwise. Rather, we hold that New York law requires an IEP to identify temporary transitional support services to be provided to an autistic student’s new classroom teacher when that student is transferring from a private school to a public school program or to a less restrictive classroom setting. See N.Y. Comp. Codes R. & Regs. tit. 8, § 200.13(a)(6); id. § 200.1(ddd); id. § 200.6(c). Moreover, in R.E. we rejected an argument concerning the absence of transitional support services for the student in the IEP because the parents had failed to “articulate[] why the absence of such [services in the IEP] was so significant as to deny [the child] a FAPE.” R.E., 694 F.3d at 195. Here, like there, A.M. has similarly failed to offer a persuasive reason why the absence of

27 transitional services to be provided to E.H.’s public school classroom teacher deprived E.H. of a FAPE. Indeed, the IHO reached the same conclusion, based in part on hearing testimony from Friedman, E.H.’s occupational therapist at MCC, who stated that, although E.H. had initially “had a really hard time with transitions . . . going from the classroom to [occupational therapy] or between different places,” he had “made really nice gains in that where he transitions not only from the classroom to [occupational therapy], but between preferred and non‐ preferred activities without any behavioral overreactions.” Suppl. App. 404. This evidence suggests that, because E.H. had made significant progress in his ability to transition between classroom settings without resulting in any notable problems, the failure to specify transitional support services in the IEP for E.H.’s teacher could not have deprived E.H. of a FAPE. Accordingly, we hold that, although the absence of transitional support services in the IEP for E.H.’s would‐be public school classroom teacher was a procedural violation, it did not deprive E.H. of a FAPE. 4. Cumulative Effect Although “[w]e have previously held that ‘[m]ultiple procedural violations may cumulatively result in the denial of a FAPE even if the violations considered individually do not,’” L.O., 822 F.3d at 123 (quoting R.E., 694 F.3d at 190), that is not the case here. Having identified four procedural errors (the failure on the part of the DOE to conduct an FBA of its own, the development of an inadequate BIP, the failure to include formal parent counseling and training services in the IEP, and the omission of transitional support services in the IEP for E.H.’s would‐be public school teacher), none of which affected the substance of E.H.’s program, this is not the rare case where the violations, when taken together, “impeded the child’s right to a

28 [FAPE] . . . or caused a deprivation of educational benefits.” R.E., 694 F.3d at 190 (internal quotation marks omitted). Accordingly, no FAPE denial resulted cumulatively from the procedural errors committed by the DOE in formulating E.H.’s IEP. B. Substantive Adequacy Unlike procedural violations, which ordinarily will not result in a FAPE denial, “[s]ubstantive inadequacy automatically entitles the parents to reimbursement.” R.E., 694 F.3d at 190. In evaluating whether an IEP was substantively adequate, we assess whether the IEP was “reasonably calculated to enable the child to receive educational benefits.” Rowley, 458 U.S. at 207. “[A] school district fulfills its substantive obligations under the IDEA if it provides an IEP that is ‘likely to produce progress, not regression,’ and if the IEP affords the student with an opportunity greater than mere ‘trivial advancement.’” Cerra, 427 F.3d at 195 (quoting Walczak, 142 F.3d at 130). Importantly, “[i]n order to avoid ‘impermissibly meddling in state educational methodology,’ [we] ‘must examine the record for any objective evidence indicating whether the child is likely to make progress or regress under the proposed plan.’” Id. (internal quotation marks omitted) (quoting Walczak, 142 F.3d at 130). A.M.’s two substantive challenges are inextricably intertwined. She argues that (1) the 6:1:1 classroom setting offered in the IEP and (2) the failure to guarantee any 1:1 ABA therapy in the IEP went against the consensus of the evaluative materials present at the CSE meeting, and that the IEP consequently deprived E.H. of a FAPE. According to A.M., the record “demonstrates overwhelmingly that [E.H.] . . . require[d] [ABA]” and “a significant amount of one‐on‐one instruction” in order to make progress toward his goals. Pl.’s Br. 15. We agree.

29 While attending MCC during the 2011–2012 school year, E.H. was educated in a 1:1 classroom setting with six students and six teachers and received intensive 1:1 therapy using the ABA methodology. The SRO reviewed the evaluative materials present at the CSE meeting and the IHO hearing record, and concluded, relying heavily on O’Sullivan’s testimony, that the IEP was substantively adequate despite the absence of any provision for 1:1 ABA therapy because, “although the May 2012 CSE did not recommend ABA, it also did not disapprove of the methodology or say ABA should not be used with the student,” and that the IEP “incorporated behavioral methodologies . . . not contrary to ABA.” Suppl. App. 15. The SRO was also persuaded by O’Sullivan’s opinion that E.H. “could make meaningful progress had the May 2012 IEP been implemented as written because [E.H.’s] special education services needed to be delivered within some social context to allow for modeling and shared instruction,” which, in O’Sullivan’s view, counseled in favor of a less restrictive placement than a 1:1 classroom setting. Suppl. App. 15–16. The District Court affirmed, concluding that, although “the SRO and IHO deviated from the consensus of the parent’s evidence, including the representatives from MCC,” “their conclusion was consistent with the views of the evaluator on the CSE,” who believed that “a 6:1:1 setting would better facilitate [E.H.’s] goals.” A.M., 2015 WL 8180751, at *5. Thus, the District Court explained that “the CSE may come to an independent conclusion of the student’s needs from the content of the evaluation reports; the CSE need not simply accept the conclusion of those reports absent an in‐person evaluation with the student.” Id. (emphasis in original). Although the IDEA does not speak to methodology, see Assistance to States for the Education of Children With Disabilities

30 and Preschool Grants for Children With Disabilities, 71 Fed. Reg. 46,540, 46,665 (Dep’t of Educ. Aug. 14, 2006), the implementing regulations define a child’s “[s]pecial education” as “specially designed instruction . . . to meet the unique needs of a child with a disability.” 34 C.F.R. § 300.39(a). Rowley teaches that the substantive adequacy of a child’s IEP turns on whether its provisions are “reasonably calculated to enable the child to receive educational benefits.” 458 U.S. at 207. This inquiry requires courts to determine whether “the content, methodology, or delivery of instruction” have been narrowly tailored to “address the unique needs of the child that result from the child’s disability.” 34 C.F.R. § 300.39(b)(3)(i) (emphases added). To be sure, in evaluating the substantive adequacy of a child’s IEP, Rowley directs us to “be careful to avoid imposing [our] view of preferable educational methods upon the States.” Rowley, 458 U.S. at 207. Thus, “once a court determines that the requirements of the [IDEA] have been met, questions of methodology are for resolution by the States.” Id. at 208. As noted, however, deference to the states “is not without limitation.” L.O., 822 F.3d at 109. Our review of a case, such as this, demands that we “engage in an independent review of the administrative record and make a determination based on a ‘preponderance of the evidence.’” Gagliardo, 489 F.3d at 112 (quoting Mrs. B., 103 F.3d at 1120). Our decision regarding R.K., one of the three students whose IDEA challenges we decided in R.E., exemplifies the interplay of this complex set of directives. There, the SRO disagreed with the IHO’s conclusion that “[t]here [was] a consensus that the student need[ed] an ABA program.” R.E., 694 F.3d at 193. Instead, the SRO “concluded that the evidence only indicated that R.K. needed a small, structured setting, which [it] found to be satisfied by a 6:1:1 placement.” Id. The SRO “also

31 found that [the child] did not necessarily need ABA because some evaluations did not specify a teaching method,” and credited testimony from R.K.’s would‐be teacher at the proposed placement that she incorporated 1:1 ABA instruction into the classroom. Id. The SRO determined that the placement was appropriate because “the student would have received individual instruction and that instruction would have been ABA‐based.” Id. (internal quotation marks omitted). We reversed and faulted the SRO’s reliance on the would‐ be teacher’s testimony because “R.K.’s parents had no knowledge or guarantee from the IEP that R.K. would have received a teacher who conducted daily 1:1 ABA sessions with each student.” Id. at 193–94. We went on to explain that the SRO’s conclusion was unsupported by the evidence because those reports that spoke to the size of the child’s classroom either recommended 1:1 instruction or “emphasized that R.K. needed a high level of support.” Id. at 194. We further observed that, “almost all of the reports found that R.K. needed continued ABA therapy,” and we explained that “[t]he fact that some reports did not mention a specific teaching methodology d[id] not negate the clear consensus that R.K. required ABA support.” Id. Because “the plan proposed in [R.K.’s] IEP offered her a 6:1:1 classroom with no dedicated aide and no guarantee of ABA therapy or any meaningful 1:1 support,” we held that the SRO’s conclusion was unsupported by a preponderance of the evidence and was thus “flawed,” that “deference to it [was] not warranted,” and that R.K. was deprived of a FAPE because “the IEP was not reasonably calculated to create educational benefit for R.K.” Id. Thus, R.E. stands for the unremarkable proposition that, when the reports and evaluative materials present at the CSE meeting yield a clear consensus, an IEP formulated for the child

32 that fails to provide services consistent with that consensus is not “reasonably calculated to enable the child to receive educational benefits,” Rowley, 458 U.S. at 207, and the state’s determination to the contrary is thus entitled to no deference because it is unsupported by a preponderance of the evidence. See C.F., 746 F.3d at 81 (holding that the IEP’s failure to provide a 1:1 classroom resulted in the denial of a FAPE because “the testimony and reports indicat[ed] that [the child’s] needs required a 1:1 placement,” thereby rendering “such instruction . . . a necessary component of any plan ‘reasonably calculated to enable the child to receive educational benefits’” (quoting R.E., 694 F.3d at 190)). This remains true whether the issue relates to the content, methodology, or delivery of instruction in a child’s IEP. Although we are mindful that the question of “how best to educate an autistic child is ‘a difficult question of educational policy’ that requires deference to the decisions of administrative experts,” L.O., 822 F.3d at 117 (quoting T.P. ex rel S.P. v. Mamaroneck Union Free Sch. Dist., 554 F.3d 247, 254 (2d Cir. 2009) (per curiam)), where, as here, “it appears plain that, contrary to the findings of the SRO,” the placement was “not adequately designed to address and improve” the child’s educational needs, the administrative officer’s analysis is deserving of no deference, see id. Here, as in R.E., the reports that did address E.H.’s needs as they related to teaching methodology and classroom size specifically recommended the continued need for ABA therapy and 1:1 support in order for E.H. to progress. For example, a report prepared by the Rusk Institute of Rehabilitation Medicine based on psychological evaluations it conducted of E.H. on March 28, April 1, and May 19, 2011 (approximately one year before the CSE meeting in question), recommended that, “[r]egardless of the placement, it is imperative that [E.H.] receive

33 the maximum ABA instruction to allow for carryover and family training in order to make appropriate progress and meet IEP goals.” Suppl. App. 689. The report also recommended that, “[w]ith regard to behavioral management in general,” “[i]ntensive behavioral intervention such as Lovaas’ [ABA] . . . techniques may be particularly useful and will provide a means of increasing communication, attention, and imitation skills, and decreasing his rigidity.” Suppl. App. 690. It further explained that E.H.’s “appropriate behaviors and learning c[ould] be fostered through the development of imitation skills,” and that “ABA is able to focus on the development of these skills.” Suppl. App. 690. Consistent with this report, following a pediatric neurology evaluation of E.H., neurologist Jay E. Selman, M.D. recommended in September 2011 that E.H. be educated using “ABA at least 20 hours per week.” Suppl. App. 749. Earlier that spring, neurologist Ram Kairam, M.D. evaluated E.H. and opined that E.H. “would need ABA therapy at least forty hours a week to see an improvement.” Suppl. App. 752. Dr. Kairam noted that when he evaluated E.H. in January 2011, A.M. “was not totally convinced that [E.H.] should be in a full time ABA program as it m[ight] be too restrictive for [E.H.] and wanted to leave [E.H.] in a [g]eneral education school with special education services,” but he “stress[ed to her] the importance of ABA therapy for an autistic child.” Suppl. App. 751. The hearing record reinforces these doctors’ recommendations. MCC assistant educational coordinator Solow testified that E.H. received primarily one‐on‐one instruction at MCC because he “ha[d] a hard time learning in a group of children.” Suppl. App. 313. She attributed this difficulty to E.H.’s “very poor attending skills.” Suppl. App. 313. She explained that one‐on‐one learning “allow[ed] for more frequent reinforcement,

34 which [E.H.] need[ed].” Suppl. App. 313. In addition, she testified that E.H.’s educators at MCC “believed that [E.H.] needed a program with a one‐to‐one setting, a one‐to‐one classroom, a classroom that uses [ABA], since [they] had seen that that ha[d] been very effective with him.” Suppl. App. 326. She also testified that she told O’Sullivan at the CSE meeting that E.H.’s “needs couldn’t be met in a 6‐1‐1” classroom setting. Suppl. App. 328. E.H.’s lead classroom teacher at MCC, Herz, echoed Solow’s testimony. Herz stated that E.H. “require[d] ABA to make progress” and that “he need[ed] the consistent one‐to‐one support and . . . the consistent expectations and constant positive reinforcement . . . for any and all behaviors.” Suppl. App. 436. Herz explained that a 1:1 classroom was necessary for E.H. “because his attention [was] so fleeting that it can become easy for him to become prompt dependent, so he needs a systematic fading of prompting.” Suppl. App. 436–37. In other words, in Herz’s view, E.H. “continue[d] to need one‐to‐one support to attend consistently to groups” and “he also need[ed] the behavioral component with the research based tactics to target his palilalia and physical stereotypy.” Suppl. App. 437. Importantly, there were no evaluative materials present at the CSE meeting that suggested otherwise, such as that E.H. could benefit from the use of some other methodology rather than 1:1 ABA, or that ABA therapy need not play a crucial role in E.H.’s educational program in order for him to continue to progress.4 Nor did the DOE conduct any evaluation(s) of its own

4 A.M.’s hearing testimony before the IHO that E.H. attended a karate class once each week with other developing children, that he “did not have many of the symptoms of an autistic child,” was “very smart and knows a lot,” and that “it’s just his attention span is what we need to

35 address in order for him to learn” because “he learned his letters; he learned his numbers; he learned his colors,” Suppl. App. 39, does not alter this conclusion. That is, this testimony, although of course highly relevant coming from the mother of the child, did not obviate the clear consensus of E.H.’s evaluators (and their evaluations) that E.H. required meaningful 1:1 classroom support and ABA therapy in order to progress. Further, the IHO’s conclusion, that because E.H. had made improvements while at MCC, he no longer required such a restrictive class setting, is unsupported by the record. First, the quoted language on which the IHO relies is taken out of context. While Solow did testify that E.H. had “really improved” his attending behavior, Solow also testified that E.H. still “ha[d] a hard time learning in a group of children,” that he still had “very poor attending skills,” that E.H. “needed a program with a one‐to‐one setting, a one‐to‐one classroom, a classroom that uses [ABA], since [they] had seen that that ha[d] been very effective with him,” and that E.H.’s “needs couldn’t be met in a 6‐ 1‐1” classroom setting. Suppl. App. 313, 319, 326, 328. Indeed, as Solow testified, the logical inference from the testimony that E.H. had “made huge gains” and “increased [his] social awareness” after a single year of learning at MCC would suggest that the more restrictive academic setting in which he was learning adequately addressed his needs and should thus be continued; not that the program should be discontinued and that he should be transitioned to a less restrictive learning environment that did not necessarily employ ABA therapy techniques—the very kind in which he was previously educated before attending MCC and in which he did not enjoy similar progress or success. Suppl. App. 319, 557. Notably, E.H.’s MCC educators, even despite his improvements, continued to recommend for the 2012–2013 academic year that he “continue in his current placement within a small ABA classroom where he c[ould] receive one‐to‐one instruction throughout the day in order to maintain and generalize skills.” Suppl. App. 654, 800.

36 of E.H. to call into question the opinions and recommendations contained in the evaluative materials. While a private school’s preferred methodology does not bind the CSE in formulating a student’s IEP, where, as here, the consensus of the evaluative materials and “all witnesses familiar with [the child],” C.F., 746 F.3d at 81, specifically recommend the continued need for 1:1 ABA therapy and support for the child to progress, and the DOE does not point to any evidence sufficient to counter these opinions and recommendations, the CSE was bound, at a minimum, to require some level of ABA support in a 1:1 classroom setting in order to establish the adequacy of the IEP. See id. (“The IEP’s substantive inadequacy, therefore, is rooted in the testimony and reports indicating that [the child’s] behavioral needs required a 1:1 placement.”); see also R.E., 694 F.3d at 194. Thus, the District Court and administrative officers’ reliance on the views of O’Sullivan, which were against the clear consensus of the substance of the evaluative materials present at the CSE meeting and the views of E.H.’s evaluators and educational instructors (i.e., “all witnesses familiar with [E.H.],” C.F., 746 F.3d at 81), was error, as this resulted in a proposed program that was not “reasonably calculated to enable [E.H.] to receive educational benefits,” see Rowley, 458 U.S. at 207. This conclusion is inescapable, “particularly when combined with [(1)] the [DOE’s] failure to [conduct] an adequate, individualized [FBA],” (2) the DOE’s failure to prepare an adequate BIP, (3) “the lack of any provision for parent counseling and training,” and (4) the omission of transitional support services in the IEP for E.H.’s would‐be public school teacher at the IEP’s proposed placement. C.F., 746 F.3d at 81. Moreover, we note that “the issue of classroom placement ratio cannot be separated from one of the procedural violations discussed above, namely the [DOE’s] failure to [evaluate E.H. and] conduct an appropriate [FBA of its

37 own] or [BIP].” Id. Indeed, we are left to wonder whether the DOE would have reached the same conclusions and recommended the same deficient services in the IEP that E.H.’s evaluators and instructors believed to be inadequate, had the DOE adequately complied with the IDEA’s procedures in the first instance. See L.O., 822 F.3d at 113; R.E., 694 F.3d at 190. Accordingly, we conclude that these significant deficiencies rendered E.H.’s IEP for the 2012–2013 school year substantively inadequate, thereby depriving E.H. of a FAPE. C. Relief A “substantive violation alone does not entitle [A.M.] to reimbursement for [E.H.’s] education at [MCC]. [A.M.] must still satisfy the second and third parts of the Burlington/Carter test, by showing that [her] alternative placement for [E.H.] at [MCC] was appropriate and that equitable considerations favor reimbursement.” T.M. ex rel. A.M. v. Cornwall Cent. Sch. Dist., 752 F.3d 145, 167 (2d Cir. 2014). Because the District Court erroneously concluded that there was no substantive deficiency in E.H.’s IEP and thus that E.H. was not deprived of a FAPE, the District Court did not address these remaining issues. Accordingly, rather than evaluate these questions ourselves without the benefit of a more fully developed record and briefing on these issues by either party, we believe the preferred course is to remand to the District Court to consider these matters in the first instance. See id. at 170.

CONCLUSION We have reviewed the parties’ remaining arguments and find them to be without merit. For the foregoing reasons, the judgment of the District Court is VACATED and the case is REMANDED for further proceedings consistent with this opinion.