Ohio’s opt-out bill skirts the real problem—and four proposals to fix it

Last week, we cautioned that Ohio’s opt-out bill (HB 420) offers a perverse incentive for districts and schools to game the accountability system. The bill has since been amended, but it is no closer to addressing the larger issues Ohio faces as it determines how best to maintain accountability in response to the opt-out movement. 

Current law dings schools and districts when a student skips the exam by assigning a zero for that student when calculating the school’s overall score (opting out directly impacts two of ten report card measures). The original version of HB 420 removed those penalties entirely. Instead of earning a zero, absent students would simply not count against the school. Realizing the potential unintended consequences under such a scenario, including the possible counseling out of low-achieving students and larger numbers of opt-outs overall, the drafters of the substitute bill incorporated two changes.

First, the amended version requires the Ohio Department of Education to assign two separate Performance Index (PI) grades for schools and districts for the 2014–15 school year—one reflecting the scores of all students required to take exams (including those who opt out) and another excluding students who didn’t participate. Second, in a clear move to curb school employees from counseling out low-performers, lawmakers created strict penalties—including criminal sanctions—for any staff who so much as thinks about uttering the words “stay home.” Specifically, it prohibits any public school employee from “negligently suggesting” to a student or parent that the student shouldn’t take the assessment. Penalties include a one-year suspension to the employee’s professional license, grounds for termination, and a minor misdemeanor—the same penalties for cheating. But these changes do nothing to address the heart of Ohio’s opt-out problem.

Ohio is eager for a quick fix where none is to be had. Federal law, thanks in part to strong advocacy from national civil rights groups, still expects 95 percent of students (and all subgroups) to participate in state assessments. Moreover, the Department of Education remains committed to enforcing that provision. Any solution that Ohio develops must work in concert with this requirement. Unfortunately, HB 420 does nothing to discourage opting out and simply attempts to stall rather than addressing the deeper problem. Below are four recommendations that Ohio should consider if it’s interested in a long-term cure rather than a short-term band-aid.

1. Publish only one performance index score, as required under existing law. Legislators need to resist the urge to publish two performance index scores, despite the push by districts that were hit hard by opt-outs. The legislature has already ensured, by extending safe harbor through the 2016–17 school year, that the state will not hold a school’s or district’s test scores against it. But wealthier (and generally higher-achieving) districts with substantial numbers of opt-outs aren’t worried about state interventions; rather, they are worried about the response of their community. By assigning two PI scores, schools and districts will merely point to the higher number as the “real” score. This removes any consequence that the opt-out movement has on a community, like lowered reputation or an impact on long-term property values. The dual PI scores let schools and communities off the hook, making it far less likely that they’ll ever have full student participation. The proposed reporting fix is slated to apply only to the 2014–15 school year. But if Ohio does nothing to actually reduce the number of opt-outs, lawmakers will be under enormous pressure to carry it forward to additional years. If Ohio’s safe harbor measure (initially a one-year reprieve that lawmakers extended to three) is any indication, this won’t be a temporary fix.
2. Develop a formal process for opting out. Taking a cue from Pennsylvania, Ohio could develop a formal process for parents determined to keep their child home from testing. Parents should be required to have an in-person meeting with a principal or school staff member, review a sample of the standardized test in question, hear directly from the school what the adverse impacts from opting out are likely to be, and sign a form explaining why they are choosing to opt out. This gives the state a method to accurately track opt-outs. (Currently, state testing data only show non-participation rates, and there is no way to distinguish intentional opting out from truancy or other types of absence.) This procedure would also add the benefit of ensuring that students and parents are fully informed about opting out and its very real consequences for schools, the reputation of their community, and on the accountability system as a whole.
3. Penalize schools for opt-out students, but only partially. Currently opt outers receive a “zero”, and amended HB 420 essentially gives them a free pass. But what about a middle-ground solution? Starting with the previous year’s score as a baseline, students who miss exams could be assigned a scale score that is, say, 15 percent lower (or an amount that psychometricians could settle on); alternatively, we could use a threshold lower than the student scored the prior year (perhaps dropping from advanced to accelerated in the proficiency categories) for purposes of the school’s and district’s aggregate calculations. Thus, schools would have no incentive to counsel low-performing students out, would still have reason to encourage high-performers to take the assessment, and would not see their performance index score tank completely if opt-out numbers soar. Third graders don’t have scores from the previous year, but because state law requires they pass state exams in order to be promoted (under Ohio’s third-grade reading guarantee), the number of opt-outs at this grade level should be relatively small. This approach also would protect Ohio from federal scrutiny, as it appears to be aligned with guidance from the Department of Education on how states should factor the 95 percent requirement into their accountability systems. (One suggestion from USDOE is for states to lower proficiency ratings for districts that fail to meet the participation benchmark.)
4. Treat teachers fairly. We’re the first to argue that school employees should not, under any circumstances or for any reason, counsel students out. This is not an infringement of First Amendments rights. There is no innate “right” to conspire to undermine the state’s accountability system or change performance data. Making students go away so that your school is rated better is immoral and—while likely extremely rare—should be expressly prohibited. Where HB 420 goes wrong is by criminalizing these actions. Penalties for counseling students out should include grounds for termination and a one-year suspension of one’s educator’s license. There’s no need to go further by making it a crime.

Amended HB 420 softens the blow for districts who report high levels of opt-outs, but it does little to curtail the problem. Ohio needs sensible policy changes that encourage full student participation, not a public relations fix. By continuing to post a single PI score—even one that is lower because of opt-outs—parents and schools alike will feel the pressure to achieve higher participation. A formal process for opting out will ensure that parents are fully informed about the ramifications of their decision. Continuing to penalize schools for opt-outs coheres with federal expectations and removes incentives for schools to counsel out low-performers (or any student) while also responding to those districts where opting out has taken a significant toll. Finally, creating consequences (but not criminal penalties) for teachers will dramatically reduce the likelihood that gaming takes place. In short, a sound policy solution recognizes what motivates community members, parents, schools, and teachers—and aligns incentives accordingly.