Passed almost a decade ago, the aptly named Third Grade Reading Guarantee aims to ensure that every Ohio student reads proficiently by the end of third grade. That goal makes perfect sense. All children need foundational reading skills to succeed in middle and high school and beyond.
Passed almost a decade ago, the aptly named Third Grade Reading Guarantee aims to ensure that every Ohio student reads proficiently by the end of third grade. That goal makes perfect sense. All children need foundational reading skills to succeed in middle and high school and beyond. As Governor Kasich said as he signed the bill into law, “Kids who make their way through social promotion beyond the third grade...they get lapped, the material becomes too difficult.” The data support his claim: Children who have difficulties reading early in life tend to struggle academically as they progress through school.
In terms of specifics, the reading guarantee includes a number of elements that help to ensure students receive the attention and supports needed to achieve proficiency. They include annual diagnostic testing in kindergarten through second grade, along with required parent notification and reading improvement plans should students be deemed off track. Yet the policy’s most widely discussed provision is its retention and intensive intervention requirement when third graders fall short of state testing benchmarks.
Retention was a hot topic in a recent House Education Committee hearing, as Ohio lawmakers mulled legislation that would waive mandatory retention for a third straight year due to the pandemic’s impacts on achievement. Extending the holiday may be politically convenient, but sadly it’ll do little to ensure that struggling children are getting up to speed. At the least, the state must return to standard protocols next year, as literacy is no less vital than before the pandemic.
Some committee members, however, seemed skeptical about ever returning to a retention policy—and a few have even sponsored another bill that would permanently drop the requirement. In committee, Representative Catherine Ingram voiced concerns that retention is “detrimental” and said that the likelihood of non-graduation increases “exponentially” if students are retained.
Her comments seem to reflect talking points based on research from the 1990s and earlier that raised questions about the effectiveness of retention. But those studies almost surely produced skewed results due to negative selection bias into retention programs—e.g., lower-achieving or less-motivated students tend to be placed in them—and a lack of adequate statistical controls. Referring to this body of work, economists Brian Jacob and Lars Lefgren write that “prior studies fail to account for the selection of students into these programs...thus potentially exaggerating the harm of retention.” Moreover, many of the older studies were based on small samples—sometimes just a few dozen students—usually drawn from a single district. (This paper describes the methodological problems in the early literature.)
Fortunately, more recent analyses have emerged that offer more apples-to-apples comparisons between thousands of retained and non-retained children. As it turns out, this research directly challenges the received wisdom, showing that retention can improve outcomes for students.
The most relevant study to the Ohio debate is a 2017 report by Guido Schwerdt, Martin West, and Marcus Winters, who analyze Florida’s pathbreaking third grade retention policy. They uncover substantial short-term achievement gains in reading and math for retained students relative to very similar pupils who just barely met the promotion threshold. Those positive results mirror earlier evaluations of the Florida program. Such test-score gains do diminish by eighth grade, but when the analysts track students into high school, retained students post higher GPAs and are less likely to need remedial coursework than their non-retained peers. No effects are found on high school graduation, refuting the notion that early-grade retention puts students at a greater risk of dropping out.
Staying in the Sunshine State, a 2019 study by David Figlio and Umut Ӧzek focuses more specifically on English language learners who were held back in third grade. The analysis also compares the academic outcomes of children just above and below the promotional standard. Similar to Schwerdt et al., they too find significant academic gains for retained English language learners and reductions in later remedial course-taking. While the analysis didn’t track students through twelfth grade and focuses on a particular group of students, it adds more high-quality evidence that retention can benefit kids.
Turning to Chicago, a pair of rigorous studies by Jacob and Lefgren examined the effects of retention in the city’s school district. Their earlier analysis (2002) finds test score gains for retained third graders two years later when compared to children who barely met the promotional target. In a 2007 follow-up study that tracked students over a longer timeframe, the researchers found that retained third graders were no more likely to drop out of high school—countering again the idea that retention increases dropout rates.
The final pieces of evidence are more suggestive than causal but also point to the benefits of retention. A recent Manhattan Institute report by Marcus Winters and Paul Perrault found broad improvements in third grade tests scores—not just among retained children—shortly after Arizona and Florida implemented a third grade retention policy. Interestingly, they propose that the threat of retention encouraged schools to focus on early literacy, yielding improvements across the board. And although no rigorous study of Ohio’s reading guarantee has been conducted, it’s worth mentioning the notable improvement on third grade reading scores prior to the pandemic. In 2015–16, for instance, 55 percent of Ohio third graders reached proficiency, while 67 percent met that standard in 2018–19.
Ohio lawmakers could easily cave to political pressure from school groups that despise the retention requirement. But they can’t justify scrapping the policy on the basis of research. Far from being a detriment, studies now indicate that retention actually gives students an academic boost. Of course, retention can’t just be more of the same. It must be paired with effective supports like high-dosage tutoring, quality summer learning, or individualized attention to give students the best chance of catching up. And—just like a good preschool program—the gains made through the extra learning time need to be sustained through excellent instruction in the years thereafter. Yet at the end of the day, giving young children the “gift of time” to become successful readers is common sense and tracks with the best available evidence.
The recently passed state budget created an off-ramp for districts under the control of an Academic Distress Commission (ADC). As part of these new rules, the three districts currently under ADC oversight—Youngstown, Lorain, and East Cleveland—must develop academic improvement plans that contain annual and overall improvement benchmarks. If they meet a majority of their academic benchmarks by June 2025, their ADC will be dissolved.
All three districts under ADC supervision have submitted their improvement plans and received feedback from the Ohio Department of Education (ODE). In a previous post, I analyzed the plan submitted by the Youngstown City School District. I expressed significant concern over how low district officials set their academic benchmarks, concerns that were shared by ODE in its review letter. The plan submitted by Lorain City School District is close to impossible to analyze. The majority of their benchmarks are incomplete, a fact that was noted by ODE in Lorain’s review letter. The East Cleveland City School District, on the other hand, fared much better. Only two of the district’s goals were flagged by ODE: the benchmark set for the percentage of students who earn an industry recognized credential and the graduation rate target for students with disabilities. Only one benchmark was identified as incomplete or missing: a goal for pre–K.
Of the three plans, East Cleveland’s plan is the most rigorous, detailed, and user-friendly. It focuses on five key areas: turnaround leadership, transformative instruction and support, talent management, transparent finances and operations, and “together we are East Cleveland,” a category which focuses on expanding and improving family outreach and community partnerships. For each area, the plan outlines challenges and areas for growth, a theory of action that will guide the district going forward, and specific objectives and strategies for implementation over the next several years.
The most important part of the plan, of course, is the academic improvement benchmarks that will be used to judge whether the district has made enough progress to dissolve its ADC. East Cleveland identified a total of thirteen. Ten are based on students’ academic performance, two are noncognitive measures (district chronic absenteeism and student participation in co-curricular activities), and one is career-based (the percent of students who have industry-recognized credentials).
For the most part, these benchmarks cover the right ground and set rigorous targets. For example, East Cleveland plans to raise their overall district performance index from 46 percent—the lowest score in the state during the 2018–19 school year—to 60 percent by 2024–25. If the district succeeds in improving that much, it will surpass pre-Covid performance index scores for itself and each of the other ADC-controlled districts, as well as seven of the Big Eight districts that educate a similar student population. In fact, unlike their ADC counterparts, East Cleveland set most of its academic benchmark goals to pre-Covid levels. Rather than using the steep drop in proficiency and growth that occurred during the pandemic to set a new baseline, the district chose to set goals that will improve on their pre- and post-Covid data. That’s important because it indicates that the district is committed to addressing learning loss and truly improving performance.
The district’s growth goals for grades 3–8 in math and reading will be measured by NWEA MAP, a computer adaptive assessment used in schools across the nation. The charts below show the percentage of students in East Cleveland who met or exceeded their expected MAP growth goals during the baseline years of 2018–19 through 2020–21, as well as the district’s target percentages for school years 2021–22 through 2024–25. It’s not ideal that growth is being measured by grade bands, as doing so could mask poor performance in certain grade levels. But each grade band is set to surpass pre-Covid levels before the plan’s final year, and the district is aiming to have a majority of students meet their growth goals in both subjects by 2025.
East Cleveland also deserves credit for including preparedness measures that are nationally-recognized indicators of student readiness. The district aims to have 19 percent of the Class of 2025 earn an ACT composite score of eighteen or higher. That isn’t exactly reaching for the stars—ACT’s college readiness benchmarks are above eighteen for every subtest except English, and 19 percent is far from a majority—but less than 5 percent of students in the district’s Class of 2019 managed to earn a composite score of eighteen. And although ODE dinged the district for setting an industry recognized credential attainment goal that is lower than pre-Covid levels, the district’s goal for the class of 2025 is still nearly 48 percent, more than four times the current statewide average. District officials should ensure that these credentials are meaningful, but the inclusion of credential attainment is a positive step.
All this is good news for East Cleveland kids. But there are a few aspects of the district’s plan that should raise concerns. For instance, it’s understandable why the district chose to use MAP to measure progress. MAP assessments have the state’s stamp of approval, and their data allow teachers and leaders to track individual student growth over time. But MAP assessments aren’t fully aligned to state standards. And opting to use them instead of the value-added measures used on state report cards makes it impossible to draw comparisons between similar schools and districts as well as state averages. These comparisons matter, as they offer a more complete picture of the district’s progress in context.
Unfortunately, East Cleveland didn’t just avoid value-added measures for state tests. It avoided state tests all together. Other than district-wide performance index, the plan doesn’t have a single benchmark for student proficiency measured by criterion-referenced assessments that are aligned to state standards, easily comparable to other districts, and disaggregated by subgroups. That’s a huge deal because it means there are no measures to determine whether the number of students who can read or do math at grade level is increasing, decreasing, or staying flat. Youngstown may have set abysmally low proficiency goals in their plan, but at least they had some. East Cleveland doesn’t have any, and that’s a problem. Growth is important—this is, after all, an academic improvement plan—but proficiency matters, too. If East Cleveland is achieving the meaningful growth its benchmarks seem to promise, then proficiency increases should follow—and district and state leaders should be tracking that improvement.
These are glaring omissions that prevent the East Cleveland plan from being held up as an ideal model. But given the limited parameters for these plans provided by state law, the rigor of the benchmarks the district did include, and the lackluster plans submitted by the other two ADC-controlled districts, East Cleveland deserves praise. If the district implements this plan well and succeeds in meeting all of these benchmarks, kids will benefit. And at the end of the day, that’s all that should matter.
The term “dual enrollment” is often used to refer to young people earning college credits while simultaneously completing their high school coursework. But a recent report from the Institute of Education Sciences (IES) provides an important reminder that there are several versions of dual enrollment. It also provides evidence that they aren’t equals in terms of access, participation, or outcomes.
The analysis looks at data from Rhode Island, which passed legislation to boost access to early college credit in 2013. By 2015, all eligible Rhode Island high school students were given access to college-level coursework free of charge, and dedicated state funding was provided for the purpose. Dual enrollment refers specifically to students taking courses taught by college professors either on campus or virtually. Those courses are primarily provided by the state’s two-year public college. Concurrent enrollment, by contrast, refers to students taking college-level courses taught by high school teachers and held in those buildings. The teachers are trained and certified by—and the credits earned through—Rhode Island’s two public four-year colleges. Additionally, high school students in the state can earn college credit by passing Advanced Placement (AP) tests.
The research focused on 8,726 students who were first-time ninth graders in the 2013–14 school year and enrolled in traditional district schools, alternative schools, charters, and vocational schools. A total of 41 percent of the cohort participated in one or more of the three early college credit programs by grade twelve. The largest proportion (27 percent) took AP tests, while a similar proportion (26 percent) participated in concurrent enrollment. Dual enrollment was a much less common option, being used by just 4 percent of students. Approximately 15 percent participated in more than one early college credit program during high school. Compared with the cohort overall, a lower proportion of participants in early college credit programs were non-White, male, or economically disadvantaged. Interestingly, despite dual enrollment reaching fewer students, its participants were more diverse—including a higher proportion who were non-White or economically disadvantaged—than participants in the other two programs. Students participating in AP and concurrent enrollment registered higher eighth grade achievement levels than did their dual enrollment peers.
Proximity to Rhode Island’s public two-year college campus drove dual enrollment utilization, with 32 percent of dual enrollment students concentrated in just two urban high schools in Providence despite the fact that this option was theoretically open to all via remote access. Concurrent enrollment was available in a majority of schools statewide, but 13 percent did not offer the option and 19 percent enrolled fewer than ten students. The majority of concurrent enrollment students attended suburban schools, while AP test-takers were distributed among urban, suburban, and rural schools in a pattern similar to the cohort overall.
To study the impacts of these programs, the researchers compared early college credit program participants to similar non-participating students, matched based on grade eight characteristics such demographics and state test scores in math. Participation in any of the early college credit programs had a large, positive effect on the probability of students graduating from high school, enrolling in college within a year of graduation, and avoiding developmental education courses during the first year of college. Those positive effects were similar for both economically disadvantaged and non-economically disadvantaged students. Advanced Placement test-taking was associated with a significantly larger increase in college enrollment than dual and concurrent enrollment, but otherwise the effects of all three programs were roughly the same.
The researchers offer a few cautions to keep in mind. Rhode Island is not readily comparable to larger states, and participation patterns and access to early college programs may be different in other places. The research is also non-experimental and is unable to account for unobserved differences between students (e.g., differing levels of motivation) that might bias the effects shown in the study.
However, descriptive data differentiating the three early college credit programs should be taken seriously by anyone interested in building or expanding access to such programs. In states and areas where college campuses are numerous, expansion of dual enrollment might encourage a more diverse early college credit program. States with small and centralized college networks could work with districts to train and certify a larger corps of secondary teachers to make concurrent enrollment more available in distant high schools. And every school could take another look at virtual college classes, in light of pandemic improvements in higher education offerings and expanded Wi-Fi and device access among previously unconnected students. Whatever its structure, policymakers need to continue giving students opportunities to get an inside track on higher education.
SOURCE: Katherine A. Shields, Jessica Bailey, Makoto Hanita, and Xinxin Zhang, “The Effects of Accelerated College Credit Programs on Educational Attainment in Rhode Island,” U.S. Department of Education Institute of Education Sciences (June 2021).
Among its many educational impacts, the pandemic has reenergized efforts to expand private school choice. States like Ohio, where it already existed, have expanded eligibility and increased funding. Meanwhile, Kentucky, Missouri, and West Virginia established their first-ever programs. In total, thirty-one states, as well as the District of Columbia and Puerto Rico, now have private school choice.
It’s possible that even more states will join these ranks over the next few years. As newly established programs grow, the number of participating students may skyrocket—and policymakers will be confronted with plenty of complicated questions. Chief among them is the age-old question: How should private schools that receive public funds be held accountable? (For more on this topic, see Fordham’s previous work.)
Nicole Stelle Garnett of the University of Notre Dame Law School offers some insight on this question in a brief recently published by the Manhattan Institute. She begins by examining the history of private school choice, as well as the current policy landscape. For the most part, private schools are free from the regulations imposed on traditional public and charter schools. Laws vary across states, but almost all are minimal. For example, although forty states regulate curricula in some way, most exempt private schools from those mandates. But the laws that govern private school choice programs tend to be strict, as taxpayer funds come into play. Some states mandate minimum qualifications for teachers and basic curricular requirements, such as teaching civics. Some voucher programs regulate school admissions policies. And all programs manage school finance in some capacity, such as restricting scholarship amounts to the school’s tuition level.
Holding private schools accountable for student outcomes is a little more complicated. Garnett outlines four challenges for accountability in the private school choice context: politics, namely the fierce debates between school choice proponents and opponents; the tension between quality (only high performing schools are permitted to participate) and quantity (parents should have as many options as possible); selection bias, which can make it difficult to compare the performance of students participating in private school choice programs with their public school peers; and the limitations of standardized-test-based accountability, as parents often look beyond test scores to gauge school quality.
According to Garnett, whether or not these challenges can be overcome depends on the purpose of accountability. If the point is to ensure that parents have only “good” choices, then overcoming these complexities will be extremely difficult. But if the goal is simply to help parents make choices, the path forward is smoother and worth pursuing. She offers four suggestions for how to design private school choice accountability regulations so that they give families the information they need, while also expanding the number and variety of options.
Encourage transparency at the school level. In some states, private school choice programs don’t require participating schools to administer any standardized tests. Those that do rarely require the results to be made publicly available. There are a few states that do both, such as Ohio, Wisconsin, Indiana, and Louisiana. Other states would be wise to follow suit, though Garnett notes that only requiring voucher recipients to take the assessment—and not all students—creates an imperfect measure of school quality.
Permit testing flexibility. Private schools shy away from administering state tests for a variety of reasons. Some fear that it’s a slippery slope toward being forced to align their curricula to state exams, while others believe they aren’t an accurate reflection of student achievement and progress. Permitting the use of multiple assessments can assuage some of these concerns, though it also creates a problem: Apples-to-apples comparisons between private and public schools are far more difficult when multiple tests are used. A compromise would be to allow private schools to administer a research-backed test of their choosing, so long as they publicly report the results for all students.
Develop alternative measures of school quality. Families value academic performance, but that’s not the only factor that they consider when making their schooling decisions. If the goal of accountability is to help parents make informed choices, then accountability systems should report on factors that matter to parents. One way to do this would be to require participating private schools to publicly report testing results and at least one noncognitive measure of school quality, such as attendance rates or information about college enrollment and persistence. Doing so would provide families with more information and could help mitigate issues created by allowing test flexibility.
Structure policies to preserve better choices. In the interest of giving families as many choices as possible, accountability policies should be structured to prevent the exclusion of private schools that struggle academically but are higher performing than other schools in their community; the comparative strength of other local schools should matter.
Garnett argues that the ultimate goal of private school accountability should be to ensure that families have access to more and better schools. That means giving parents and guardians the information they need to make wise choices—information that is transparent, easy to interpret, and reflective of the school-quality criteria that matter to them.
SOURCE: Nicole Stelle Garnett, “Accountability and Private-School Choice,” The Manhattan Institute (October 2021).
NOTE: Today, the Ohio House of Representatives’ Primary and Secondary Education Committee heard testimony on SB 229. The bill would, among other provisions, make a number of pandemic-influenced changes to remote learning and test-based accountability. Fordham’s Vice President for Ohio Policy provided interested party testimony on the bill. These are his written remarks.
Thank you, Chair Manning, Vice Chair Bird, Ranking Member Robinson, and House Primary and Secondary Education Committee members for giving me the opportunity to provide interested party testimony today on Senate Bill 229.
My name is Chad Aldis, and I am the Vice President for Ohio Policy at the Thomas B. Fordham Institute. The Fordham Institute is an education-focused nonprofit that conducts research, analysis, and policy advocacy with offices in Columbus, Dayton, and Washington, D.C.
When the 2020-2021 school year finished in June, many people felt a big sigh of relief. It was a challenging year, but there was also a sense of optimism that the coming year would be different and include a return to something resembling normal. Unfortunately, late summer brought a resurgence of Covid cases and many schools were disrupted by both positive cases and quarantines. Most schools exhibited a strong resolve—and a splash of creativity—to keep their students in classrooms whenever possible. Nevertheless, some schools needed to resort to some combination of remote or blended instruction at times.
It’s for that reason that we are largely supportive of the underlying objective of SB 229 which allows Ohio schools increased flexibility to employ remote and blended learning—with some important guardrails—this school year. We don’t offer this support lightly. The data from last school year is clear: Ohio students struggled and those learning remotely struggled even more.
In September, Ohio State University professors Stéphane Lavertu and Vlad Kogan published a first look at 2020–21 student achievement data. Student learning is down one half to a full year in math and one third to one half a year in English. As feared, historically underserved students had the biggest declines. Because Ohio districts followed different paths last year—some opted for virtual school while others chose hybrid or in person learning—the researchers were able to demonstrate that remote learning played a big role in the learning declines.
Despite agreeing with the general intent of SB 229, there are a couple of areas where we have concerns.
The first has to do with a provision related to state testing for online (mostly charter school) students. Existing law provides that a student at an online school who fails to take the state assessment for two consecutive years is automatically withdrawn from the school. SB 229 grants some flexibility to that provision by restarting the clock for those students. On its face, that would be fine. Inexplicably though, it restarts the clock not for this school year but beginning next school year (2022-2023). With 94 percent of students having taken the state assessment last year, there’s no way that online students shouldn’t be expected to take the state assessment this school year. Over the past five or six years, Ohio has made tremendous progress in regards to charter school accountability. This change would be a step backward and is being done without any obvious need. We request that you modify the existing language to restart counting for the two-year automatic withdrawal this school year (2021-2022).
The second concern surrounds changes to the third grade reading guarantee. Given the extraordinarily low levels of student achievement in third grade this year, extending the flexibility granted last year around this provision makes sense. Districts simply couldn’t retain all of the students who haven’t reached the promotion benchmark. That being said, we think it’s critical that parents be given honest, transparent information indicating that their child isn’t reading on grade level and information detailing how important early reading is to later educational success. There is a very real possibility that large numbers of students will be promoted from third to fourth grade without reading at or near grade level, and their parents will believe that everything is okay until they struggle in later grades. While flexibility is needed, we urge you to require an enhanced disclosure to parents empowering them with the information they need to know exactly where their children are in terms of reading ability.
Thank you for the opportunity to provide testimony.
