Search

Numerous prior studies have shown that high-dosage, in-person tutoring can yield substantial gains in student achievement growth. However, staffing in-person tutoring programs can be challenging. Consequently, virtual tutoring has become an attractive option for school districts seeking to accelerate student achievement growth in the wake of the COVID-19 pandemic. In SY 2021–22, Clayton County Public Schools (CCPS) contracted with a virtual tutoring platform, Tutor.com, to provide one-on-one, on-demand virtual tutoring sessions to students in the district. In this report, Tim Sass and Salma Mohammad Ali use individual tutoring session data from CCPS to highlight patterns of usage among students. We disaggregate usage by school, grade level, subject area, and the number of sessions in which students participated. Additionally, we assess the impact of virtual tutoring use on students’ achievement in math, measured by formative assessment scores (i-Ready or MAP Growth) in SY 2021–22. We find usage of the tutoring platform was modest in terms of the number of students that participated and the frequency of sessions. Usage varied by school and was generally higher among elementary and middle school students. The most popular subject for tutoring sessions was math, where pandemic-era declines in achievement growth have been greater than in reading. While students at all prior achievement levels participated in tutoring sessions, usage was generally lower among higher-achieving students. In general, the extent of Tutor.com usage was not correlated with student achievement gains in math, though there is suggestive evidence that high-intensity use could yield gains in student achievement. Unfortunately, the causal relationship between virtual tutoring usage and student achievement is unclear because students chose their level of Tutor.com use. We find that the relationship between the quantity of non-math tutoring and math test scores is similar to the relationship between math test scores and the number of math tutoring sessions. This suggests that our estimates of the impact of math tutoring on math scores are partly a reflection of the characteristics of students who chose to take advantage of virtual tutoring rather than the tutoring itself. Student usage must increase for virtual tutoring to be an effective tool in promoting student achievement growth. Integrating the use of virtual tutoring into the school day would be the most direct way to promote usage. However, incentivizing students to participate in virtual tutoring before or after school and messaging parents to encourage usage may be helpful as well.

The COVID-19 pandemic affected teachers in myriad, unprecedented ways. In spring 2020, most schools in the United States shut their doors and were forced to switch to remote learning. Changes in instructional mode, coupled with general health and economic concerns brought on by the pandemic, negatively impacted teachers and led to increases in levels of stress and burnout. In this report, Sarah S. Barry and Tim R. Sass use administrative data covering SY 2016–17 through fall of SY 2021–22 from two metro-Atlanta school districts (DeKalb County School District and Gwinnett County Public Schools) to analyze trends in teacher hiring and retention. The data include information on teacher employment status as well as information on teacher experience, certification, subject area taught, school assignment, and teacher demographic characteristics. We aim to provide a better understanding of teacher labor markets in the context of the COVID-19 pandemic and determine appropriate policy interventions in response to changes in teacher mobility and retention. We find that, while there have been changes in attrition and teacher hiring over time, most post-pandemic observations appear to have reverted to pre-pandemic trends. Attrition in both districts initially decreased after the onset of the pandemic. Attrition in Gwinnett remains lower than the pre-pandemic trend, while attrition in DeKalb is slightly higher than before the pandemic. In both districts, the proportion of teachers with 30 or more years of experience who left their respective districts has been on an upward trend since at least SY 2016–17. It does not appear that the pandemic has led to a worsening of teacher attrition in hard-to-staff subject areas. In addition, the proportion of new hires in both districts initially dropped soon after the start of the pandemic. The proportion of new teachers in Gwinnett has since increased relative to the pre-pandemic period. In Dekalb, the proportion of new teachers is well below pre-pandemic levels, though changes have not impacted average student-teacher ratios. Pre-pandemic challenges to recruiting and retaining teachers remain, particularly in “high-need” areas like math and science and special education, and warrant careful analysis of pandemic-era financial incentives (such as those used by the two districts in this study) to determine if these policies could help address teacher shortages in specific subject areas. Further, districts should carefully consider pre-pandemic conditions as they look to better understand any attrition and hiring issues they face.

Two years into the pandemic, little information has been available about how students in the metro-Atlanta area are performing academically. Metro districts have adopted different strategies to help students navigate and recover from the pandemic, including small-group tutoring, expanded summer school programs, and additional supports for technology use and student health and wellness. This report continues the Metro Atlanta Policy Lab for Educations research agenda analyzing how the pandemic has affected student achievement in math and reading in metro-Atlanta districts.In this updated report, Tim Sass and Thomas Goldring used administrative data from three districts: Clayton County Public Schools, Fulton County Schools, and a third district that has chosen to remain anonymous. Our outcomes of interest were math and reading scores on the iReady and MAP Growth formative assessments. We used national percentile rankings to measure student progress, comparing student achievement during the pandemic to that of students throughout the U.S. from before the pandemic.We find that the impact of the pandemic on student achievement growth has been greater in math than in reading; reading achievement has rebounded more than math. Students who were in elementary school when the pandemic hit have fared worse than students who were in middle school. Average performance in metro-Atlanta districts was similar to many other districts around the country. Many students have started to recover academically, but improvement is uneven. We find differences in achievement trends by geography and by economic status within at least one district and by race and ethnicity within multiple districts.Our findings have several policy implications. Recovery efforts ought to target students who experienced the greatest declines in national rankings and have been the slowest to recover. Research-supported recovery strategies include high-intensity, small-group tutoring; extended learning time; and extensive summer academic learning programs. Acceleration efforts are more effective if offered during the regular school day (where possible). Frequent participation, which may need to be mandated or incentivized, is key to the programs success.

In this report, Henry T. Woodyard, Tim R. Sass, and Ishtiaque Fazlul estimated the effects of attending an oversubscribed school-based Georgias Pre-K Program on achievement, attendance, and discipline in elementary school. Using waitlist data for oversubscribed school-based pre-K sites in one metro-Atlanta school district, we compared students who gained a seat through an enrollment lottery and attended a school-based site in Georgias Pre-K Program to students who did not gain a seat through a lottery and did not go to any (school-based or non-school-based) site in Georgias Pre-K Program.We find that students selected in a lottery enter kindergarten significantly more prepared academically, scoring around six national percentiles higher than their non-selected peers on nationally-normed formative assessments. However, these gains fade by the end of kindergarten, and some negative effects on achievement emerge by Grade 4. Measured benefits are larger when we only consider students who qualify for free or reduced-price meals (FRPM), suggesting that attending pre-K may be more beneficial for students from families experiencing low income. While selected students were no less likely than non-selected students to commit a disciplinary infraction in any grade, they did miss about one fewer day of instruction in each grade after kindergarten. The broad patterns we find are consistent with previous studies of the efficacy of universal pre-K programs elsewhere.The limitations of our analysis make us cautious in providing policy recommendations. However, given that one distinguishing characteristic of school-based pre-K sites is that they offer transportation, offering students with limited transportation options priority at sites which offer transportation could be impactful. Similarly, providing additional funding could help non-school-based sites overcome the cost of providing transportation. Finally, informing parents of non-selected students of next steps and other options within Georgias Pre-K Program may reduce the chance that their child does not attend any formal pre-K.

In March 2020, the COVID-19 pandemic forced schools to close and abruptly switch to remote learning. Since that time, schools have slowly transitioned back to in-person learning, and school year (SY) 2021–22 was the first full year where most students returned to in-person learning. In this report, we continue the Metro Atlanta Policy Lab for Education’s (MAPLE) research agenda analyzing how the pandemic has affected student achievement in three metro-Atlanta school districts—focusing on recent pandemic-era student achievement growth during SY 2021–22. This report is the third in a series of reports where we measure pandemic-era student achievement growth. In this updated report, Tim Sass and Salma Mohammad Ali utilize administrative data from three metro-Atlanta school districts: Clayton County Public Schools, DeKalb County School District, and Fulton County Schools. We measure student achievement growth in math and reading using scores from the i-Ready and MAP Growth formative assessments. Using national percentile rankings, we compare student achievement during the pandemic to that of students throughout the United States from before the pandemic. We find that the return to near-universal in-person learning in SY 2021–22 did not yield substantial improvements in average math or reading achievement growth. Similar to our previous report, we find that students who were in Grades 1–3 when the pandemic hit fared worse than students who were in middle school, though this pattern is more pronounced for math than reading. The pandemic has impacted students differently, and the subsequent recovery has been uneven across student groups. We find that differences in achievement by race and ethnicity and economic status have grown in some districts but not in others. We also find a narrowing of place-based average achievement differences for one school district. While our findings confirm that some students are now at or near their pre-pandemic achievement levels, many other students remain substantially below their pre-pandemic achievement levels. We suggest that recovery efforts focus on students who experienced the greatest declines in national rankings and have been the slowest to recover. High-dosage small-group tutoring (either in-person or online) is the most promising strategy for promoting academic recovery.

It has been over two years since the COVID-19 pandemic led to school closures and an abrupt shift to remote learning in March 2020. We continue MAPLE’s research agenda analyzing how the pandemic has affected student achievement in math and reading in three metro-Atlanta school districts. We focus on how students fared in school year 2021–22, the first full year of in-person instruction for students since the pandemic began.

In response to the COVID-19 pandemic, school districts across the United States were forced to close their physical schools and rapidly switch to virtual learning. In this study, Jennifer Darling-Aduana, Tim Sass, and Henry Woodyard examine patterns in student engagement and achievement associated with the enactment of crisis-driven virtual schooling in a metro-Atlanta school district during the fall 2020 semester. To accomplish this, they used data from student records, standardized assessments, and course progress information gathered by the digital learning systems used throughout the district. Overall, students logged substantially fewer hours a week in virtual learning applications than expected based on schedules for traditional face-to-face instruction prior to the pandemic. Despite generally deflated weekly achievement growth compared to expected pre-pandemic trends, students who logged more hours virtually experienced higher rates of student achievement growth throughout the semesterwith time logged in programs with student-directed, interactive assignments associated with the largest gains in growth. They also identified variability in usage patterns by subgroup with students identified as female logging significantly more time across many applications than students identified as male. Students identified as Black also logged more time in synchronous meetings than students identified as White. Consequently, the pandemic learning gap might be more accurately portrayed as an opportunity gap due to less access to educational experiences. Through this research, they aim to provide a benchmark and insights into how to best provide students equitable access to quality virtual engagement opportunities. Moreover, ensuring students have equitable access to the resources needed to learn in this way is an important policy consideration.

Dual Language Immersion (DLI) programs are schools or classrooms within schools that educate students in two languages: English and a target language. DLI programs in Georgia have proliferated in recent years to include 66 programs that reach across 14 school districts as of school year (SY) 2018-19. We partner with two metro-Atlanta school districts to study the impact of access to early-grade DLI programs using data from five cohorts of DLI admission lotteries across 10 DLI programs. This is the largest rigorous analysis of the academic impacts of access to DLIs in Georgia. On average, we find that students with access to DLI programs via enrollment lotteries have higher reading and math test scores by approximately 0.1 standard deviations compared to students who applied but did not gain access. Our results on the impact of DLI access on test scores are comparable in magnitude to the effect of having an experienced teacher.

The Lucy Calkins Units of Study Program is a widely-used program that consists of materials and methods for teaching reading and writing in Grades K8. A metro-Atlanta school district (the district) has used the Lucy Calkins Program for many years in the elementary grades with the timing and extent of adoption at the discretion of individual school leaders. In this report, Bla Figge and Tim R. Sass provide evidence on the impact of the Lucy Calkins Program on student achievement in reading/language arts relative to alternative programs in use. We analyze student test scores on summative (Milestones) assessments in Grades 35 from school year (SY) 201213 to SY 201819 and scores on nationally-normed formative assessments in Grades K3 over the period SY 2017-18 to SY 2018-19. We compare student test scores in a school and grade that had implemented the Lucy Calkins Program to students with similar characteristics and prior-year scores from the same school in different grades and years who were not exposed to the program. The estimated average impacts of the Lucy Calkins Reading and Writing Programs on annual student achievement gains in Grades 4 and 5 are positive but small. In Grades 13, the use of the Reading Program is estimated to have a small, negative effect on student achievement. In neither set of grades can we confidently rule out that the effect is zero. We cannot say with confidence that the impacts differ significantly across implementation levels. Impacts of the Reading Program in Grades 4 and 5 appear to be higher for students experiencing poverty than for students from more affluent families and higher for English language learners than for students proficient in English.

The COVID-19 pandemic forced schools across the U.S. to switch to virtual learning. In this project, we studied trends in technology usage and virtual learning during fall 2020 in a metro-Atlanta district. Our analysis elucidates promising strategies and highlights areas for additional attention, which can help improve the effectiveness of virtual learning.

Two years into the COVID-19 pandemic, little information has been available about how students in the metro-Atlanta area are performing academically. Metro districts have adopted different strategies to help students navigate and recover from the pandemic, including small-group tutoring, expanded summer school programs, and additional supports for technology use and student health and wellness. This report continues MAPLE’s research agenda analyzing how the pandemic has affected student achievement in math and reading in metro-Atlanta districts.

In response to the COVID-19 pandemic, school districts in metro Atlanta closed physical schools and quickly shifted to remote instruction in mid-March 2020 for the remainder of the school year. Many efforts have been used to mitigate the effects of pandemic-induced school closures, including virtual summer school. Prior to the pandemic, studies indicate that participation in summer programs tends to reduce learning loss in mathematics; however, in reading, the impact is mixed. Interventions have ranged from summer enrichment programs to online programs targeting students across all grade levels. In this report from Alexa Prettyman and Tim Sass, we seek to determine whether elementary and middle school students benefited from a virtual summer school program in a metro-Atlanta school district during the COVID-19 pandemic. We used spring assignment completion records, summer school files, and administrative records from the spring of school year 2019-20 to the fall of school year 2020-21 to compare achievement growth between summer school participants and non-participants. Overall, virtual summer school participation was low. Only about 1 in 4 students who were supposed to participate actually did. We find that participants performed better in math, on average, than non-participants, but there was no difference in reading. Most of the gains are concentrated among elementary school students and not middle school students. In addition, students who were more engaged during the spring semester seemed to benefit more from summer school. Districts may want to mandate or incentivize participation in future programs offered in the summer or during breaks in the academic calendar. Lastly, use of clear and objective assignment rules and better documentation of who is expected to attend summer school will provide for causal analysis in the future

Research conducted prior to the COVID-19 pandemic suggests that summer programs tend to improve student achievement growth in math and have a mixed impact for reading. In this project, we analyzed whether elementary and middle school students in a metro-Atlanta district benefited from a virtual summer school program during the pandemic.

The Lucy Calkins Units of Study Program consists of materials and methods for teaching reading and writing in Grades K-8. We provide evidence on the impact of the Lucy Calkins Program on student achievement in a metro-Atlanta district. Achievement gains overall for students in Grades 4 and 5 are no higher when exposed to the Lucy Calkins Program rather than other reading and writing programs. Similarly, in Grades 1-3, we find no evidence of a positive effect on reading achievement gains.

We estimate the academic impacts of eligibility for and participation in an intensive English program aimed at English Learners (ELs) with very low English proficiency. The program is a short-term intervention to introduce students to basic English skillsacademic and socialbefore they transfer to their neighborhood school. Our results indicate that ELs who are eligible for the program have lower English Language Arts (ELA) achievement in the short term, relative to ELs who receive traditional English as a Second Language (ESL) instruction. These findings are concentrated among older ELs. In subsample analyses, however, we find that refugee ELs who are eligible for the program have higher ELA and math achievement in the short term.

A snapshot of the student population in the school district in our study reveals that in 2016 there were approximately 3,300 refugee students across all grades (K-12). Thus, refugee students made up roughly 3 percent of the total student population in the district. On average, there were about 250 refugee students per grade, with significant variation across grades. For example, while only 2 percent of first-graders were refugees, this proportion climbs to 5 percent among ninth-graders. Refugee students in the district come from a variety of countries and enter with varying language skills. In 2016, over 18 countries were represented among the refugee population. The top country of birth for refugee students in the district was Burma (Myanmar), followed by Ethiopia, The Democratic Republic of Congo, Iraq, and Bhutan. Approximately 26 percent of refugee students did not receive English as a Second Language (ESL) services.

Of the various school-based determinants of student achievement, teachers are the most important (Rivkin, Hanushek, and Kain, 2005; Aaronson, Barrow, and Sander 2007; Kane, Rockoff, and Staiger, 2008). Further, teachers who boost student test scores the most are also able to improve adult outcomes of their students (Chetty et al., 2011; Chetty, Friedman, and Rockoff, 2014). There are a number of potential mechanisms for increasing teacher quality, such as improving current teachers skills, enhancing teachers incentives to maximize their performance, and differentially retaining superior teachers. This brief focuses on another promising avenue for enhancing teacher qualityimproving the teacher hiring process. The Metro Atlanta Policy Lab for Education investigated the characteristics principals emphasize and undervalue during the teacher hiring process and found some characteristics were better than others at predicting teacher performance.