Counting Zeros: The Every Student Succeeds Act and the Testing Opt-Out Movement
Paul A. Hoversten*
The story begins with threatening letters. In October 2014, the U.S. Department of Education reminded Colorado’s chief state school officer that the department “ha[d], in fact, withheld Title I, Part A administrative funds . . . from a number of States for failure to comply with the assessment requirements” under the Elementary and Secondary Education Act. Given the occasion, the department implied, it wouldn’t hesitate to be ruthless.
At the time, the No Child Left Behind Act was governing law, and states were obligated to ensure that 95 percent of students (and all defined subgroups of students) participated in standardized testing. The 95 percent participation threshold was a compliance criterion—if states or school districts came up short, they risked losing federal funding. As long as students participated in yearly standardized achievement tests, everyone stayed happy. But a state’s failure to reach the 95 percent participation threshold would put federal funds in jeopardy.
So the rise of the testing opt-out movement in 2014 and 2015 put everyone in an awkward position. Students and parents in pockets around the country protested the assessments and “opted out” en masse. In November 2014, students at Cherry Creek High School (near Denver) participated in science and social-studies standardized assessments at a rate of about 3 percent. At Boulder’s Fairview High School, only nine out of 538 seniors participated in the science test; ten participated in the social-studies test. All told, Boulder Valley School District seniors participated in these standardized assessments at a rate of about 16 percent. The opt-out phenomenon was not limited to high schools nor to science and social-studies assessments.
Under pressure to respond, the Colorado Board of Education voted in February 2015 to hold harmless the school districts with low participation rates; ordinarily, such low participation rates would have negative consequences for districts’ state accreditation scores. Shortly thereafter, the Colorado General Assembly frantically moved to pass something, anything, to give its blessing to the opt-out movement. Other states passed similar bills, each with its own nuanced language. Then, in December 2015, only months after the new Colorado policy came into effect, Congress passed the Every Student Succeeds Act (ESSA), reauthorizing the Elementary and Secondary Education Act of 1965 and largely displacing the No Child Left Behind regime. This dizzying series of events left students, teachers, parents, and policymakers understandably confused.
Which brings us back to threatening letters. Less than two weeks after President Obama signed ESSA into law, the U.S. Department of Education reproached the chief school officers of several states: “A few States did not assess at least 95 percent of students in the ‘all students’ group or individual . . . subgroup(s) . . . .” The letter continued, “If a State’s response does not adequately address this problem and meet the State’s assessment obligations . . . [the department] may take enforcement action.”
This tension came to a head in summer 2017. Under ESSA, each state is required to submit an accountability plan to the federal Department of Education, showing how it intends to satisfy its statutory obligations. In its accountability-plan proposal, the Colorado Department of Education bluntly stated, “[Our] accountability participation rate for a school/district/disaggregated group removes [opt-out students] from the denominator.” In other words, students who opt out of assessments would not count against Colorado’s achievement scores. This was an explosive suggestion, and it earned another protesting missive from Washington on August 11, 2017. Under the state’s proposed scheme, the U.S. Department of Education said, “[Colorado] does not meet [the] statutory requirement.” On October 11, 2017, the Colorado Board of Education approved a compromise measure in hopes of conforming with federal demands while maintaining its opt-out policy.
Colorado could be forgiven for assuming it was authorized to craft its own policies in this arena; according to the Wall Street Journal, ESSA represented “the largest devolution of federal control to the states in a quarter-century.” But what precisely had become of No Child Left Behind’s 95 percent participation requirement? In this Essay, I parse the state and federal laws at play to determine whether the federal government really has the legal authority and political will to withhold funding from states that allow students to opt out of standardized assessments. I also evaluate Colorado’s proposed response to the federal threat, gauging its conformity with the letter and the spirit of ESSA.
I. Reform and Confusion in State and Federal Law
This Part narrates the Colorado General Assembly’s efforts to codify legal protections for students and parents who opt out of standardized assessments, and it explains a key misunderstanding that might have occurred along the way. It also discusses the effect of recent federal legislation on standardized-assessment requirements across the country and on Colorado more specifically.
A. Reform in Colorado: SB 15-223 and HB 15-1323
When Colorado State Senators Holbert and Todd set out to enact a testing opt-out bill, Senate Bill 15-223, the end goal was not entirely clear, and the mechanics of the bill were seemingly an afterthought. Introducing the bill in the Senate Education Committee, Senator Holbert emphasized, “Let’s make sure that everyone in the delivery system of education is clear that they do not have the authority, the right, the empowerment from government to stand in the way of” the decision to opt out of standardized assessments. The title and text of the bill made clear that, if passed, it would ensure that a “reduced student participation rate” would not “result in negative consequences.”
But “negative consequences” is a capacious term. Senator Johnston framed the problem concisely:
If [Senator Holbert’s] bill is passed, and thirty percent of students opt out in a district, and those thirty percent are the top performing thirty percent in that district . . . that district, on that assessment, is going to perform less well. . . . Or, let’s say the opposite. Let’s say a district . . . has the lowest thirty percent of their performers that don’t take the test, and [they] suddenly go from “priority improvement” [the lowest rating] to “excellent” because those thirty percent sat out. How are you going to build or maintain a state accountability system that doesn’t have accurate information for either of those two districts?
Senator Holbert responded by painting a grotesque image—an image he would refer to repeatedly—of “a school district, maybe law enforcement, but someone standing there on their doorstep threatening to have the parents or the students arrested if they don’t participate in the assessments.” Whereas Senator Johnston understood the term “negative consequences” to describe any state intervention in underperforming schools and districts, including demotion of their accreditation scores, Senator Holbert appeared to perceive the term in a material, almost corporeal, sense.
Senator Johnston’s point was this: whether or not a school or district reaches its 95 percent participation threshold, at the very least, we should permit ourselves to measure schools and districts based on the results they receive. In other words, beneath the 95 percent threshold, opt-out students’ scores should count as zeros. If a low participation rate inoculates low-performing schools and districts against all “negative consequences,” then there is no accountability system left. At least this way, everyone is always working with a complete denominator.
Whether or not that is good policy, it grasps the distinction between participation and measurement. This approach says we need not punish for punishment’s sake; but there is, of course, a consequence for failure to demonstrate high performance, and standardized assessments are the mechanism that we use for demonstrating high performance. By contrast, Senator Holbert’s preferred approach would have legislatively done away with both the participation requirement (in No Child Left Behind) and the measurement requirement (in ESSA). According to Senator Holbert’s approach, there would be no punishment for schools and districts with low participation and no obligation for schools and districts to measure the performance of all (or nearly all) their students.
Senate Bill 15-223 passed in the full Senate but died in a House committee. A few weeks later, the governor signed into law a similar bill, House Bill 15-1323, prohibiting schools and districts from “impos[ing] negative consequences” as a result of a student or parent’s decision to opt out of standardized assessments. Notably, the language of that statute imposes a prohibitory obligation on schools and districts but apparently not on the State Board of Education or the Colorado Department of Education, as the Senate bill would have done. Thus, although an elementary school principal may not give detention to an individual student who opts out of a standardized assessment, the statute does not preclude the Colorado Department of Education from lowering a school’s accreditation rating based on its overall low participation rate.
B. Reform in Congress: The Every Student Succeeds Act
When the opt-out debate reached Washington, Congress drew precisely the distinction that seemed to elude Senator Holbert in Colorado. Whereas the No Child Left Behind regime would punish schools and districts directly for falling below the 95 percent participation requirement, ESSA does not set a minimum participation threshold. Instead, it sets a minimum measurement threshold. Whether or not they participate in assessments, ESSA demands that 95 percent of students be counted toward the denominator of average student scores. Schools and districts will not be punished for low participation rates, but low participation rates will lead to low achievement scores and, eventually, public accountability and mandatory state intervention. More specifically, ESSA states, “[f]or the purpose of measuring, calculating, and reporting” on average achievement scores, the State shall “include in the denominator the greater of—(I) 95 percent of all [eligible] students . . . or (II) the number of students participating in the assessments.” Below the 95 percent threshold, in other words, nonparticipating students must be scored as zeros.
ESSA was passed in December 2015, but it did not take full effect until the 2017–2018 school year. Colorado had a year and a half to figure out whether and how to comply with the new terms. Recall that SB 15-223 never passed, and the law that took its place uses weaker language. The statute bars “local education provider[s]” from imposing negative consequences based on opt-out decisions, a marked departure from SB 15-223’s proposal which would have barred “[t]he [Colorado] Department of Education” from imposing negative consequences. Moreover, the regulations that accompany Colorado’s opt-out law bear almost no relation to ESSA’s concern with “measuring, calculating, and reporting” achievement scores. Those regulations give examples of “negative consequences” such as “prohibiting school attendance, imposing an unexcused absence, or prohibiting participation in extracurricular activities . . . .” Indeed, Colorado has an “opt-out law” on the books, but its hands are not seriously tied with respect to ESSA’s 95 percent measurement requirement. The Colorado Board of Education and the state Department of Education would be well within the confines of governing state law to count nonparticipating students as zeros, as ESSA requires.
Nevertheless, when it submitted its draft accountability plan to Washington in May 2017, the Colorado Department of Education challenged its federal counterpart to a game of chicken: “[Our] accountability participation rate for a school/district/disaggregated group removes [opt-out students] from the denominator.” In the world of standardized assessments, there is a difference between demanding participation and demanding measurement. No Child Left Behind demanded both; ESSA demands only the latter. Colorado said it would do neither.
When the federal Department of Education responded by taking a hard line, a Colorado Department representative remarked, “It didn’t come as a surprise . . . . There’s a need to reconcile state board, state legislature and federal requirements and policies.” Indeed. This is a direct confrontation between a federal mandate and a state policy, and the federal government is showing its teeth. The question is whether it will bite.
II. Federal Leverage
Whether the federal threat is genuine has two components. First, does the U.S. Department of Education have the legal authority to withhold federal funding based on Colorado’s opt-out policy? And second, does it have the political will to do so? This Part explores both questions.
A. Legal Authority to Withhold Funding
The Constitution authorizes the federal government to place conditions on state grants. There are, however, limitations on that authority. The most familiar articulation of those limitations appears in the landmark case South Dakota v. Dole, but the Supreme Court also articulated some principles about conditional spending and congressional coercion in National Federation of Independent Business (NFIB) v. Sebelius. This Section briefly describes each of those possible arguments and applies them to the testing opt-out conflict between Colorado and the U.S. Department of Education.
In South Dakota v. Dole, the Supreme Court established a four-part test for determining the constitutionality of a conditional spending regime. When Congress purports to place conditions on federal funding, (1) it must do so in the “pursuit of the general welfare,” (2) the conditions of the funding must be stated unambiguously, (3) the conditions must be related to the purpose of the funding itself, and (4) the conditions imposed must not be barred as unconstitutional under another constitutional provision.
With respect to the testing opt-out conflict, the first, third, and fourth Dole factors are easily dispensed with. The “general welfare” test is substantially deferential to Congress’s judgment. It is similarly difficult to challenge the statute on “relatedness” grounds; no matter what standard you apply, ESSA’s 95 percent measurement requirement is clearly related to the policy of educational accountability that animates Title I and similar funding sources. Finally, there is nothing inherent in ESSA’s measurement requirement that would induce or compel grant recipients to behave unconstitutionally.
Dole arguably only comes into play with its requirement that spending conditions be stated “unambiguously,” such that recipient states can “exercise their choice knowingly, cognizant of the consequences of their participation.” Conditions must be set forth ahead of time, just as with a contract, and Congress must furnish the states with “clear notice” that each condition is, in fact, a criterion that must be satisfied in order to receive funding.
ESSA does make a clear statement about its 95 percent measurement condition: “For the purpose of measuring, calculating, and reporting on [average achievement scores, the State shall] include in the denominator the greater of—(I) 95 percent of all [eligible] students . . . or (II) the number of students participating in the assessments.” And ESSA is clear about the consequences of noncompliance, stating that “[i]f a State fails to meet any of the requirements of this section, the Secretary [of Education] may withhold funds for State administration under this part until the Secretary determines that the State has fulfilled those requirements.”
ESSA, however, also nods to the opt-out movement. It features language that, at first blush, might be perceived as something less than “clear notice” of the statutory spending conditions. Section 1111(b)(2)(K) introduces a nonpreemption provision: “Nothing in this paragraph shall be construed as preempting a State or local law regarding the decision of a parent to not have the parent’s child participate in the academic assessments under this paragraph.” This nonpreemption provision could open the door to the argument that ESSA does not make a clear statement about its 95 percent measurement condition at all; rather, it speaks from both sides of its mouth. On one hand, ESSA demands 95 percent measurement, but on the other, it purports not to preempt state and local opt-out policies.
But provided we draw a distinction between a participation requirement and a measurement requirement, this provision is not particularly difficult to reconcile with the remainder of the statute. Congress is telling students and parents the following: if your state says you need not participate in standardized assessments, we will not force you. We will not compel your participation or penalize your nonparticipation. We only require that, if you choose not to participate, your school, district, and state calculate your score as a zero for the purposes of their accountability plan.
ESSA itself is not unclear in that respect. But in March 2017, the federal Department of Education issued a template for states to use in crafting their accountability plans. In that template, it asked states to “[d]escribe how the State factors the requirement for 95 percent student participation . . . into the statewide accountability system.” As explained above, ESSA does not create a participation requirement; it creates a measurement requirement. The federal Department’s characterization of the annual measurement clause was an unforced error.
Despite that error, the linguistic inconsistency between ESSA and the federal department’s template plan is unavailing as a Dole “clear notice” argument. To the degree that the department’s accountability-plan template fostered any confusion about states’ assessment obligations, nothing about that inconsistency suggested that states were not on the hook for measuring at least 95 percent of students’ scores. A participation requirement implies a measurement requirement. The department’s miscue suggested that states had a greater obligation than they really had, not a lesser obligation. Thus a state cannot persuasively argue that it was not on “clear notice” that ESSA obligates it to measure the achievement scores of 95 percent of students regardless of the students’ participation rate. None of the Dole factors gives rise to a compelling legal argument against federal authority to withhold funding based on Colorado’s noncounting policy.
States could conceivably attack the 95 percent measurement requirement as unconstitutionally coercive, using NFIB as a model argument. But ESSA is a bad fit for a coercion claim. Whereas the Medicaid expansion in NFIB dramatically raised the stakes of participation and imposed new funding conditions on states, ESSA is not more lucrative than its predecessor No Child Left Behind, and its conditions are less onerous.
According to Professor Samuel Bagenstos, the Court in NFIB outlined three principles of spending-clause coercion: (1) the “too-big-to-refuse principle,” (2) the “no-new-conditions principle,” and (3) the “no-conditions-about-separate-programs principle.” First, in Colorado’s case, the threatened federal funds constitute somewhere between 0.5 percent and 1.6 percent of the state’s approximately $26.8 billion budget for fiscal year 2018, which is not close to the 10-plus percent at stake for most states in NFIB. Thus the too-big-to-refuse argument is likely off the table. Second, the “no-new-conditions principle” is also a nonstarter here. The expanded Medicaid program in NFIB was coercive in part because it imposed new conditions on a preexisting program. Because ESSA’s measurement requirement was already built in to No Child Left Behind, there are no new conditions. Finally, the Court in NFIB was concerned about Congress’s ability to weave an intricate web of funding conditions across programs, thereby always keeping recipients trapped in a state of perpetual compliance. There are no crossover conditions in the ESSA funding scheme. Each funding source listed in the department’s October 2014 letter independently sets grant conditions based on “[s]tate assessment results,” “valid and reliable information” about states’ provision of free and appropriate public education, or Title I’s accountability-plan section itself. Thus none of the lines of argument in NFIB can be persuasively employed to challenge the federal threat to withhold funding based on ESSA’s measurement requirement.
B. Political Will to Withhold Funding
The federal Department of Education, then, does have the legal authority to withhold funding from states (like Colorado) that count their opt-out student scores as a null set rather than as zeros. But legal authority and political will don’t always align, and “because we can” has never been a particularly compelling justification for aggressive government action. Indeed, withholding federal funding in the context of K–12 education is exceedingly rare.
This is the first line of defense for opt-out supporters. One educational nonprofit organization argues:
The federal government has never penalized a state, district or school for failing to test enough of its students. Due to the successful 2015 and 2016 New York State opt out campaigns, hundreds of districts had less than 95% participation. [The federal Department of Education] said it had no plans to penalize districts or schools by withholding funds.
Despite threats from Washington, this assertion appears to be correct. The federal department’s October 2014 letter to the Colorado Department of Education perhaps sounded more menacing than it really was. The letter stated that the federal department, in the past, had withheld “administrative funds” under Title I, Part A, but it made no such assertion about Title I program funds. Administrative funds comprise up to 1 percent of total funds under Title I, Part A, and program funds make up the other 99 percent. Although the department routinely threatens program funding as well, those threats might not be especially credible. The department has apparently never withheld Title I program funds based on noncompliance with the Elementary and Secondary Education Act, and it has only withheld funding on one occasion since 1990 based on noncompliance with civil rights statutes.
Even if its threats were credible, it seems unlikely that the Trump-DeVos Administration would spend political capital holding states accountable under this interventionist federal scheme. Many were surprised to see Secretary of Education Betsy DeVos take a hard line on Colorado’s noncounting policy, just as many were surprised to see the federal department initially reject Delaware’s accountability plan for setting achievement goals that were not sufficiently “ambitious.” But despite early signs of aggression, the department is still headed by Betsy DeVos, who billed herself as a hands-off, state-and-local-control administrator, and who declined even to acknowledge that schools that receive federal funding should be required to meet the federally-imposed conditions of the Individuals with Disabilities Education Act.
Moreover, although federal law authorizes the department to withhold funding, this is not an area of law with a clear moral right and wrong. By contrast, in the civil rights context, there is often little political downside to punishing bad guys, at least in the easy cases. Society generally agrees that discrimination based on race, sex, or disability is a serious problem and supports efforts to counteract it. Not so in the context of the testing opt-out conflict. There is significant public support for the opt-out movement, and it is a legitimate political force. Any serious adverse action by the department based on disapproval of state opt-out policies seems likely to cause a political firestorm over an issue that initially appeared of little importance to the Trump-DeVos Administration.
Apart from pragmatic considerations, however, perhaps the most compelling rationale against federal withholding of funds is that the punishment doesn’t fit the crime—or rather, it doesn’t punish the criminals. In a January 2016 letter to the director of the State School Boards Association, the New York State Allies for Public Education asserted that “[t]aking Title I money away from the neediest students in order to punish parents who are boycotting a testing system that is out of control is not defensible.” Whether or not you agree that the “testing system” is “out of control” or that parents are the primary victims of funding cuts, the observation here is striking. Title I funds support disadvantaged students in poor, marginalized communities. Yet, by almost all accounts, the students in poor, marginalized communities are not the ones opting out of assessments. During a committee debate, Senator Johnston noted,
[Opting out] has been referred to by some folks tonight as . . . a willful, conscience-driven act of civil disobedience . . . . [T]he power of civil disobedience is to say, “I think that this is a rule that I oppose, and so I’m willing to stand up against this rule and I’m willing to take the consequences of standing up against it.” . . . To say that “I’m going to make this statement, but I want someone else to be punished for the statement that I made” doesn’t seem to be fair . . . or reasonable.
Even if the Department of Education has the legal authority and political will to withhold that funding from Colorado or other states, there would be something perverse about the effects.
III. Colorado’s Revised Proposal and Implications for Other States
On October 11, 2017, the Colorado State Board of Education responded to federal pressure by adopting a compromise policy. Under the new policy, Colorado proposes to split its calculation schemes: for federal accountability purposes, it will count nonparticipating students’ scores as zeros, but for purposes of internal state performance frameworks, it will continue to treat opt-out students as a null set. This might satisfy the strict terms of the federal statute, but it is problematic nevertheless.
The Colorado K–12 performance frameworks utilize standardized-assessment data to (1) “hold districts and schools accountable for performance” and (2) “inform a differentiated approach to state support based on performance and need.” But low participation yields noisy data. For instance, Boulder Valley School District has historically performed above average on standardized assessments. Under the federal scheme (in which nonparticipating students’ scores are calculated as zeros), a low participation rate could plunge that district into the range of low-performing, priority-intervention school districts. Colorado Commissioner of Education Katy Anthes insists that her department can filter out that noise, differentiating between schools and districts with artificially low scores due to low participation and those with genuinely unsatisfactory performance, and manage state interventions appropriately. Based on what? Intuition, apparently. What happens when a traditionally low-performing district presents a low participation rate? What error-correction algorithm will the department use to determine whether that district needs support? What about a middling district? Colorado’s revised plan is grounded in speculation, not data.
This is not how ESSA is supposed to work. Even if Colorado’s revised proposal complies with the letter of the federal statute (it probably does), it subverts the underlying policy. ESSA builds an infrastructure for states to identify, support, and hold accountable the schools and districts that underperform—put another way, to identify and support underserved students. ESSA is designed to encourage high participation in standardized assessments and thus yield meaningful achievement data. By contrast, Colorado’s revised proposal would produce, process, and promptly ignore the data necessary to effectuate the federal policy of accountability and targeted support.
If Colorado’s revised proposal survives federal scrutiny, it will be proof of concept for other states that want to enact or amplify opt-out policies. California and Oregon already allow students and parents to opt out of all state standardized assessments, no questions asked. Utah and Wisconsin permit opt-outs for specific assessments. Idaho, Montana, Nevada, and South Dakota do not have statewide opt-out policies; instead, local schools and districts set their own protocols. Each of those states (and others) already faces a potentially uncomfortable mismatch between the way it calculates achievement data for federal purposes and the way it treats nonparticipation internally.
Faced with that mismatch, states might find it easier to follow Colorado’s lead: that is, to produce data according to the federal scheme but then obscure or ignore that data back home. And if the Colorado plan succeeds, it will turn out to be much easier to subvert ESSA’s policy than originally thought. The Colorado model does not require a direct confrontation with federal statutory requirements; it allows states to submit a plan that facially complies with federal law yet still play fast and loose with that data on the back end. No need for a showdown when misdirection does the job.
ESSA is meant to encourage high participation in standardized assessments, but it does not pursue that goal at all costs. If Congress had wanted to demand high participation in standardized assessments, it could have conditioned federal funds directly on participation rates (as it did in No Child Left Behind), rather than on measurement and reporting schemes. The disruption that might ensue from Colorado’s resistance was predictable—that’s what happens when Congress defangs an accountability statute. States that prefer not to be held accountable will opt out of meaningful accountability. The coming months and years will reveal whether accountability in elementary and secondary education is truly an option.
. Letter from Deborah S. Delisle, Assistant Sec’y, U.S. Dep’t of Educ., to Robert K. Hammond, Comm’r of Educ., Colo. Dep’t of Educ. 6 (Oct. 3, 2014) [hereinafter October 2014 Letter], https://www.cde.state.co.us/cdedepcom/usdoeletter [https://perma.cc/686W-YVBN].
. No Child Left Behind Act of 2001, Pub. L. No. 107-110, 115 Stat. 1425 (2002) (codified in scattered sections of 15, 20, 42, and 47 U.S.C.), amended by Pub. L. No. 114-95, 129 Stat. 1802 (2015) (codified in scattered sections of 20 U.S.C.).
. 20 U.S.C. § 6311(b)(2)(I) (2012).
. Id. § 6311(g)(2). At first blush, only Title I, Part A funds appeared to be at risk for a state’s failure to comply with the participation requirement. See id. But the requirement indirectly affected grant provisions throughout the Elementary and Secondary Education Act as well since those provisions relied on academic achievement data or Title I’s accountability system itself. See infra notes 75–77. In Colorado’s case, that ambiguity meant that noncompliance put the state at risk of forfeiting somewhere between $145 million and $430 million in federal funding. See Colo. Dep’t of Educ., NCLB Final Allocations: Fiscal Year 2014–15, at 8 (2015), http://www.cde.state.co.us/cdefisgrant/nclb-fy14-15-revised-final-allocations [https://perma.cc/7VG3-QK64] (showing that Colorado received approximately $144,844,885 in Title I, Part A funding in fiscal year 2015); U.S. Dep’t of Educ., Funds for State Formula-Allocated and Selected Student Aid Programs 13 (2016), https://www2.ed.gov/about/overview/budget/statetables/18stbystate.pdf [https://perma.cc/KN7G-TJ6S] (showing that Colorado received approximately $430,444,273 in all elementary- and secondary-level funding programs during fiscal year 2015).
. Eric Gorski, Tens of Thousands of Colorado Students Opted Out of PARCC Tests Last Spring, New Data Shows, Chalkbeat (Nov. 12, 2015), https://www.chalkbeat.org/posts/co/2015/11/12/tens-of-thousands-of-colorado-students-opted-out-of-parcc-tests-last-spring-new-data-shows/ [https://perma.cc/HP5U-GTYG]; Elizabeth A. Harris & Ford Fessenden, ‘Opt Out’ Becomes Anti-Test Rallying Cry in New York State, N.Y. Times (May 20, 2015), https://www.nytimes.com/2015/05/21/nyregion/opt-out-movement-against-common-core-testing-grows-in-new-york-state.html?mcubz=3 [https://perma.cc/VDJ3-AWQW].
. See Jenny Brundin, Thousands of Students Protest Colorado Standardized Tests, Colo. Pub. Radio (Nov. 13, 2014), http://www.cpr.org/news/story/thousands-students-protest-colorado-standardized-tests [https://perma.cc/R4XZ-TXDX].
. Eric Gorski, Thousands of Colorado High School Students Refuse to Take State Tests, Denv. Post (Nov. 13, 2014, 4:12 AM), http://www.denverpost.com/2014/11/13/thousands-of-colorado-high-school-students-refuse-to-take-state-tests/ [https://perma.cc/3Q4R-33ZF].
. See Eric Gorski, “Opt-Out” Movement: Just Say No to New Colo. Assessments, Denv. Post (Feb. 28, 2015, 12:04 PM), http://www.denverpost.com/2015/02/28/opt-out-movement-just-say-no-to-new-colo-assessments/ [https://perma.cc/H7FU-K7A9].
. Eric Gorski, Colorado Board of Education: Don’t Punish Districts on Testing Mandates, Denv. Post (Feb. 18, 2015, 4:44 AM), http://www.denverpost.com/2015/02/18/colorado-board-of-education-dont-punish-districts-on-testing-mandates/ [https://perma.cc/RY4J-RKYJ].
. See Colo. Rev. Stat. § 22-7-1013 (2015); Todd Engdahl, Testing Opt-Out Bill Gets Final Senate Approval, Chalkbeat (Apr. 6, 2015), https://www.chalkbeat.org/posts/co/2015/04/06/testing-opt-out-bill-passes-first-floor-test/ [https://perma.cc/6KQM-KFKW]; Todd Engdahl, Testing Opt-Out Bill Passes Senate Education, Chalkbeat (Mar. 26, 2015), https://www.chalkbeat.org/posts/co/2015/03/26/testing-opt-out-bill-passes-senate-education/ [https://perma.cc/7ARH-ZFC9].
. See Sarah-Jane Lorenzo, Opt Out Policies by State, Pol’y Update (Nat’l Ass’n of State Bds. Of Educ., Alexandria, Va.), Aug. 2015, at 1, http://www.nasbe.org/wp-content/uploads/Lorenzo_opt-out-guidance-table.pdf [https://perma.cc/P8WJ-V26E].
. Pub. L. 114-95, 129 Stat. 1802 (2015) (codified in scattered sections of 20 U.S.C.).
. See Press Release, White House, White House Report: The Every Student Succeeds Act (Dec. 10, 2015), https://obamawhitehouse.archives.gov/the-press-office/2015/12/10/white-house-report-every-student-succeeds-act [https://perma.cc/7A2D-MNZQ].
. Letter from Ann Whalen, Assistant Sec’y for Elementary & Secondary Educ., U.S. Dep’t of Educ. 1–2 (Dec. 22, 2015) [hereinafter December 2015 Letter to Several States], https://www2.ed.gov/policy/elsec/guid/stateletters/eseadclpartrate12222015.pdf [https://perma.cc/8YG4-6YYU].
. Id. at 2.
. 20 U.S.C. § 6311(c) (Supp. III 2016).
. Colo. Dep’t of Educ., Consolidated State Plan Under the Every Student Succeeds Act (ESSA), at 63 (2015) [hereinafter Consolidated State Plan], https://www2.ed.gov/admins/lead/account/stateplan17/cocsa2017.pdf.
. See Letter from Jason Botel, Acting Assistant Sec’y, U.S. Dep’t of Educ., to Katy Anthes, Interim Comm’r of Educ., Colo. Dep’t of Educ. (Aug. 11, 2017) [hereinafter August 2017 Letter], https://www2.ed.gov/admins/lead/account/stateplan17/cofeedbackltr882017.pdf [https://perma.cc/66WF-9CGB].
. Id. at 4.
. State Board of Education Regular Board Meeting 9:00 a.m., Colo. Bd. Educ. 12 (Oct. 11, 2017), http://www.boarddocs.com/co/cde/Board.nsf/Public [https://perma.cc/JRJ3-NYLV] [hereinafter Colorado State Board of Education Meeting (October)]; see infra Conclusion.
. No Child Left Behind’s Successor, Wall Street J. (Nov. 29, 2015, 6:12 PM), https://www.wsj.com/articles/no-child-left-behinds-successor-1448838727 [https://perma.cc/RS8Z-EDMA].
. Audio from Senate Education Committee Hearing, Colo. Gen. Assembly 3:30:24 (Mar. 26, 2015), http://coloradoga.granicus.com/MediaPlayer.php?view_id=44&clip_id=7538 [https://perma.cc/7FFC-79DG] [hereinafter Colorado Senate Education Committee Hearing]
(statement of Sen. Holbert, Member, S. Educ. Comm.) .
. S.B. 15-223, 70th Gen. Assemb., 1st Reg. Sess. (Colo. 2015).
. Colorado Senate Education Committee Hearing, supra note 23, at 3:57:32 (statement of Sen. Johnston, Member, S. Educ. Comm.).
. Id. at 3:59:01 (statement of Sen. Holbert, Member, S. Educ. Comm.).
. See Colorado Senate Education Committee Hearing, supra note 23, at 6:14:37 (statement of Sen. Johnston, Member, S. Educ. Comm.) (explaining his proposed amendment to the bill).
. S. Journal, 70th Gen. Assemb., 1st Reg. Sess. 676 (Colo. 2015); H. Journal, 70th Gen. Assemb., 1st Reg. Sess. 1205 (Colo. 2015).
. See Colo. Rev. Stat. § 22-7-1013(8) (2016); 1 Colo. Code Regs. § 3.05(E)(1) (2016); see Governor Signs School Testing Reduction Bill Into Law, CBS Denv. (May 20, 2015 11:43 AM), http://denver.cbslocal.com/2015/05/20/governor-signs-school-testing-reduction-bill-into-law/ [https://perma.cc/UM4W-Y73K].
. See Colo. Rev. Stat. § 22-7-1013(8) (2016); S.B. 15-223, 70th Gen. Assemb., 1st Reg. Sess. (Colo. 2015).
. 20 U.S.C. § 6311(b)(2)(I) (2012); see December 2015 Letter to Several States, supra note 15, at 1–2.
. See 20 U.S.C. § 6311(c)(4)(E)(i) (Supp. III 2016).
. See id. § 6311(d).
. Id. § 6311(c)(4)(E)(ii).
. Id. § 6311(c)(4)(D)(i).
. See Colo. Rev. Stat. § 22-7-1013(8) (2016).
. S.B. 15-223, 70th Gen. Assemb., 1st Reg. Sess. (Colo. 2015).
. Compare § 6311(c)(4)(E)(ii), with 1 Colo. Code Regs. § 3-05(E)(1) (2016).
. 1 Colo. Code Regs. § 3-05(E)(1).
. Consolidated State Plan, supra note 18, at 63.
. See supra notes 31–32 and accompanying text.
. August 2017 Letter, supra note 19, at 3–5.
. Nic Garcia, Feds to Colorado: You Must Count Students Who Opt Out of Standardized Tests, Chalkbeat (Aug. 14, 2017), https://www.chalkbeat.org/posts/co/2017/08/14/feds-to-colorado-you-must-count-students-who-opt-out-of-standardized-tests/ [https://perma.cc/HP83-A68G] (quoting Pat Chapman, the Colorado Department of Education’s executive director of federal programs).
. U.S. Const. art. I, § 8, cl. 1; South Dakota v. Dole, 483 U.S. 203, 206 (1987).
. Dole, 483 U.S. at 205.
. 567 U.S. 519, 587–88 (2012).
. Dole, 483 U.S. at 207–08.
. Id. (citations omitted).
. Id. at 207 n.2 (questioning whether a “general welfare” determination is justiciable in the first place).
. Conditions must “bear some relationship to the purpose of the federal spending.” New York v. United States, 505 U.S. 144, 167 (1992).
. See 20 U.S.C. § 6301(d)(8) (2012).
. Dole, 483 U.S. at 207 (quoting Pennhurst State Sch. & Hosp. v. Halderman, 451 U.S. 1, 17 (1981)).
. See Pennhurst State Sch. & Hosp. v. Halderman, 451 U.S. 1, 25 (1981) (“Though Congress’ power to legislate under the spending power is broad, it does not include surprising participating States with postacceptance or ‘retroactive’ conditions.”).
. See Arlington Cent. Sch. Dist. Bd. of Educ. v. Murphy, 548 U.S. 291, 296 (2006).
. 20 U.S.C. § 6311(c)(4)(E)(ii) (Supp. III 2016).
. Id. § 6311(a)(7).
. See Arlington Cent., 548 U.S. at 296.
. 20 U.S.C. § 6311(b)(2)(K).
. U.S. Dep’t of Educ., Revised State Template Plan for the Consolidated State Plan (2017), https://www2.ed.gov/admins/lead/account/stateplan17/revisedessastateplanguidance.docx [https://perma.cc/DD7R-UPCY].
. Id. at 13 (emphasis added).
. See id.
. Compare id., with 20 U.S.C. § 6311(c)(4)(E) (Supp. III 2016).
. 567 U.S. 519 (2012).
. Compare, e.g., Department of Education Appropriations Act, 2015, Pub. L. No. 13-235, 128 Stat. 2492, 2492 (2014) (appropriating about $15,536,107,000 to carry out Title I and closely related programs in fiscal year 2015), with 20 U.S.C. § 6302 (Supp. III 2016) (appropriating $15,012,317,605 to carry out Title I in fiscal year 2017).
. Whereas the No Child Left Behind Act mandated a 95 percent participation rate and reporting of annual achievement data (out of a denominator that, by definition, included 95 percent or more of all eligible students), 20 U.S.C. § 6311(b)(2)(I)(ii) (2012) (amended 2015); id. § 6311(h)(1)(C)(i)–(ii), ESSA mandates only the latter, 20 U.S.C. § 6311(c)(4)(E)(i)–(ii) (Supp. III, Vol. II 2016); id. § 6311(h)(1)(C)(ii).
. Samuel R. Bagenstos, The Anti-Leveraging Principle and the Spending Clause After NFIB, 101 Geo. L.J. 861, 865 (2013).
. See S.B. 17-254 Narrative, 71st Gen. Assemb., 1st Reg. Sess. 33 (Colo. 2017); John Frank & Brian Eason, Colorado Lawmakers Introduce a $26.8 Billion State Spending Bill, Denv. Post (Mar. 27, 2017, 11:48 AM), http://www.denverpost.com/2017/03/27/colorado-state-budget-proposed/ [https://perma.cc/9PBA-45TG]; supra note 4.
. NFIB, 567 U.S. at 542.
. See Bagenstos, supra note 68, at 881–83.
. See supra note 67.
. See Bagenstos, supra note 68, at 892–98.
. October 2014 Letter, supra note 1.
. E.g., 20 U.S.C. § 7173(a)(14)(A)(ii) (Supp. III 2016).
. E.g., 20 U.S.C. § 1416(b)(2)(B)(i) (2012).
. E.g., 20 U.S.C. §§ 6303(e)(1)(C), 6641(b)(2)(A)(ii)–(iii), 6841(a)(3) (Supp. III 2016).
. Colorado Senate Education Committee Hearing, supra note 23, at 3:50:32 (statement of Josh Abram, Legis. Analyst) (“There are a range of [possible] consequences. . . . It’s my historical knowledge that it is a rare occurrence for the federal government to eliminate Title I dollars – I don’t know of any circumstance in which they have.”); cf. Emma Brown, Yes, the Feds Could Pull North Carolina’s Education Funding for Violating Transgender Civil Rights, Wash. Post (May 9, 2016), https://www.washingtonpost.com/news/education/wp/2016/05/09/yes-the-feds-could-pull-north-carolinas-education-funding-for-violating-transgender-civil-rights/?utm_term=.d5447243f3bd [https://perma.cc/7S35-63KU] (describing only one instance since 1990 in which the Department of Education Office for Civil Rights followed through on its threat to withhold federal funding).
. See, e.g., FairTest, Why You Can Boycott Standardized Tests Without Fear of Federal Penalties to Your School (Jan. 24, 2017, 9:35 PM), http://www.fairtest.org/sites/default/files/ESSA-OptOut-Factsheet.pdf [https://perma.cc/2Y2E-HAXR]; The Truth About Title I Funds and Annual Testing, More Than a Score (Mar. 28, 2016), https://morethanascorechicago.org/2016/03/28/the-truth-about-title-i-funds-and-annual-testing/ [https:// perma.cc/A352-SZZJ].
. FairTest, supra note 79. I could not find an instance in which the federal Department had affirmatively stated that it had no plans to pursue adverse action against New York. FairTest might have meant to refer to an assurance by the chancellor of the New York State Board of Regents that the State itself would not withhold funding from schools and districts. See Monica Disare, Opt-Out Movement Unlikely to Provoke Sanctions from State, This Time Around, Chalkbeat (Feb. 23, 2016), https://www.chalkbeat.org/posts/ny/2016/02/23/opt-out-movement-unlikely-to-provoke-sanctions-from-state-this-time-around/ [https://perma.cc/3JN9-XFVW].
. See supra note 78 and accompanying text.
. See October 2014 Letter, supra note 1.
. 20 U.S.C. § 6304(a) (2012).
. See October 2014 Letter, supra note 1.
. Cf. The West Wing: On the Day Before (NBC television broadcast Oct. 31, 2001) (LEO MCGARRY: “You know what? Bill Russell was getting eaten alive ‘cause they could never get him to throw an elbow. He didn’t want to do it. So, Red Auerbach told him to do it one time. ‘Throw an elbow in a nationally televised game. You’ll never have to do it again.’ ”). To date, the Department of Education has not meaningfully thrown an elbow in the arena of standardized-assessment requirements.
. See supra note 78.
. See Garcia, supra note 45 (“Pushback from the U.S. education department to states has been more stern than many education policy observers expected given DeVos’s support of school choice and local control.”).
. See Erica L. Green, DeVos’s Hard Line on New Education Law Surprises States, N.Y. Times (July 7, 2017), https://www.nytimes.com/2017/07/07/us/politics/devos-federal-education-law-states.html?mcubz=3 [https://perma.cc/FU8Q-D458]; Letter from Jason Botel, Acting Assistant Sec’y, U.S. Dep’t of Educ., to Susan Bunting, Sec’y of Educ., Del. Dep’t of Educ. 3 (June 13, 2017), https://www2.ed.gov/admins/lead/account/stateplan17/deprelimdetermltr.pdf [https://perma.cc/GQ67-82JT].
. Consider the following exchange during Devos’s confirmation hearing:
SEN. KAINE: Should all schools that receive taxpayer funding be required to meet the requirements of the Individuals with Disabilities in [sic] Education Act?
MRS. DEVOS: I think that is a matter that’s best left to the states.
Nomination of Betsy DeVos to Serve as Secretary of Education: Hearing Before the S. Comm. on Health, Educ., Labor, & Pensions, U.S. Senate Committee on Health, Educ., Labor & Pensions 3:31:26 (Jan. 17, 2017, 5:00 PM), https://www.help.senate.gov/hearings/nomination-of-betsy-devos-to-serve-as-secretary-of-education [https://perma.cc/P722-VPLF].
. See Press Release, Pub. Affairs Council, Most Americans Believe Discrimination Is a Serious Problem and Would Support More Efforts by Business to Prevent It (Oct. 17, 2016), https://pac.org/2016pulserelease [https://perma.cc/7JY4-XCM3].
. See Oren Pizmony-Levy & Benjamin Cosman, How Americans View the Opt Out Movement 31–32 (2017); Valerie Strauss, The Testing Opt-Out Movement Is Growing, Despite Government Efforts to Kill It, Wash. Post (Jan. 31, 2016), https://www.washingtonpost.com/news/answer-sheet/wp/2016/01/31/the-testing-opt-out-movement-is-growing-despite-government-efforts-to-kill-it/?utm_term=.c9ae785c8a98 [https://perma.cc/EPL5-JZRF].
. Letter from N.Y. State Allies for Pub. Educ., to Tim Kremer, Exec. Dir., N.Y. State Sch. Bds. Ass’n 2 (Jan. 27, 2016), http://www.nysape.org/nysape-response-to-tim-kremer-nyssba.html [https://perma.cc/QR9K-GDAX].
. Improving Basic Programs Operated by Local Educational Agencies (Title I, Part A), U.S. Dep’t Educ., https://www2.ed.gov/programs/titleiparta/index.html [https://perma.cc/9UNK-87Z5].
. See Oren Pizmony-Levy & Nancy Green Saraisky, Who Opts Out and Why? 6 (2017) (“The typical opt out activist is a highly educated, white, married, politically liberal parent . . . whose household median income is well above the national average.”); Andrew Ujifusa, N.Y. Opt-Out Rate Hits 20 Percent on Common-Core Tests, EdWeek (Aug. 12, 2015, 1:27 PM), http://blogs.edweek.org/edweek/state_edwatch/2015/08/ny_opt-out_rate_hits_20_percent_on_common-core_tests.html [https://perma.cc/8EDX-UJYL] (explaining that, in New York, students who opted out of standardized assessments were “[m]ore likely to be white . . . [l]ess likely to be economically disadvantaged[,] [l]ess likely to come from an economically needy district[,] and [l]ess likely to be an English-language learner”). Despite aberrations in this pattern, see Colorado Senate Education Committee Hearing, supra note 23, at 3:44:53, it seems fair to conclude that students and parents who opt out of standardized assessments, on the whole, do not belong to the communities that benefit most from Title I grant funding.
. Colorado Senate Education Committee Hearing, supra note 23, at 6:53:36.
. Colorado State Board of Education Meeting (October), supra note 21.
. Id. (“I move that [the Colorado Department of Education] update the previously submitted ESSA plan to note that when calculating achievement ratings for identifying schools under ESSA for comprehensive or targeted support, but not for state performance frameworks, the state will count any non-participants in excess of 5%, as non-proficient records, in accordance with the federal requests.”). The motion passed, 7–0.
. District and School Performance Frameworks, Colo. Dep’t Educ., https://www.cde.state.co.us/accountability/performanceframeworks [https://perma.cc/E6DZ-7RD8].
. See SchoolView Data Center, Colo. Dep’t Educ., https://edx.cde.state.co.us/SchoolView/DataCenter/reports.jspx [https://perma.cc/29D8-UABK] (follow “Boulder Valley Re 2” hyperlink under “Browse”; then follow “Performance” hyperlink).
. State Board of Education Regular Board Meeting 9:00 a.m., Colo. State Bd. of Educ. 1:00:52 (Sept. 13, 2017), https://www.cde.state.co.us/sites/default/files/audio//SBE_SEP_13_2017PM.mp3.
. See id. at Part 2, 1:02:10 (“I think we would, we would say we know the ones that are struggling that have the participation right there. The ones that don’t have the participation, we don’t know if they are [struggling] or not. . . . So . . . there are some unknowns. We may be identifying them; we may be misidentifying them. We don’t know.” (emphasis added)).
. Cf. supra note 25 and accompanying text.
. Because the revised plan proposes to measure, calculate, and report data to the federal government according to ESSA’s terms, it technically does not deviate from the statutory requirements. See supra note 35 and accompanying text.
. Nat’l Ass’n of State Bds. of Educ., supra note 12, at 1, 6.
. Id. at 7.
. Id. at 2, 4, 6.
. Id. at 3–4, 6–7.
. See 20 U.S.C. § 6311(g)(2) (2012) (amended 2015).