Record: Gary Miron and Charisse Gulosino, Virtual Schools Report 2016: Directory and Performance Review. Boulder, CO: National Education Policy Center, (2016). [Available Here]
Introduction: This report is the fourth in an annual series published by the National Education Policy Center (NEPC). The first report in 2013 was followed by the 2014 report (reviewed here) and the 2015 report (reviewed here). I recommend reading the previous reviews first since this post will only cover the information that is new for 2016.
The NEPC report received a massive overhaul this year due to the addition of blended schools (schools that combine traditional face-to-face instruction with virtual instruction) alongside their discussion of full-time virtual schools. To briefly summarize the report’s findings, they say that, “Although the evidence base is becoming stronger and more convincingly negative, an increasing number of parents and students are opting for full or part-time online options” (p. 7).
The 2016 report analyzes data from the 2014-2015 school year. Private virtual or blended schools are excluded from the analysis because the data is not available for them. Also excluded are schools enrolling less than 25 students and schools that offer a mix of face-to-face and virtual options that make it impossible to separate the data of the full-time virtual and blended learning students.
Enrollments: During the 2014-2015 school year, 33 states had full-time virtual schools, and sixteen had blended schools. Many of these schools were charter schools. Two states (New Jersey and Rhode Island) had blended schools but not full-time virtual schools. For this report, 457 full-time virtual schools met the selection criteria. Since last year’s report, 67 schools were opened, and 13 were closed. These 457 schools enrolled 261,868 students. As for blended schools, there were 87 which met selection criteria. They enrolled 26,115 students. Virtual schools operated by for-profit Education Management Organizations (EMOs) enrolled an average of 1,027 students, while those operated by non-profit EMOs enrolled an average 286 students. Public virtual schools enrolled an average of 266 students.
Schools operated by the two largest EMOs, K12 Inc. and Connections Academy, enrolled 57.4% of the students enrolled in full-time online schools in the 2014-2015 school year as compared to 53% in the previous year. While K12’s enrollments decreased slightly during 2014-2015, Connections Academy’s market share grew from 17% to 23%. Overall, virtual charter schools represent about 82.5% of enrollments in full-time virtual schools, which means that district-controlled virtual schools receive a relatively small enrollment. Except for 16 schools, almost all virtual charter schools are run by for-profit organizations. A company called Rocketship Education controls 22.3% of blended school enrollments.
Student characteristics: 69.9% of students in full-time virtual schools were White compared to 49.8% in schools nationally. Blacks and Hispanics were thus underrepresented in virtual schools. Blended schools were closer to the national averages, but Hispanics made up 39.3% of enrollments in blended schools (compared to 15.5% nationally) probably due to the fact that blended schools are concentrated in three states with large Hispanic populations: Arizona, California, and Colorado. Overall, this distribution of race and ethnicity was largely unchanged from the previous year.
The proportion of students in full-time virtual schools who qualified for free or reduced-price lunch was 17% lower than the average in all public schools in 2013-2014. There was a higher proportion of students qualifying for free or reduced-price lunch in blended schools than in full-time virtual schools. This report does not include new information about the proportion of students with disabilities or those who are English language learners, but previous reports suggest that students with these characteristics are also underrepresented in virtual schools.
Full-time virtual schools enroll a disproportionate amount of girls (52.5%). Furthermore, students are more likely to enroll in a virtual school as they age. Interestingly, blended schools enroll noticeably fewer students in middle school than they do in elementary and high schools.
Student-teacher ratios: The available data is incomplete and erratic, but the national student-teacher ratio for all schools was about 16 students to 1 teacher. The average in virtual schools was almost twice as much: 35 to 1. Schools operated by for-profit EMOs had the highest ratio at 44 students to 1 teacher. In blended schools, the ratio was 32:1.
School performance data: In 2014-2015, most states underwent sweeping policy changes with the introduction of the Common Core and its new systems of accountability. For this report, only 19 of the 31 states with virtual schools had published school performance data, which is lower than usual due to the transition to new standardized tests. We must look at the data in this section extremely cautiously due to the limited amount of data and the new assessments; however, it appears that virtual schools continue to underperform academically.
Of the 121 virtual schools for which data was available, 18.2% had proficiency rates above the state average on standardized tests, and 82% had proficiency rates below the state averages. Only 16.7% of independent virtual schools and 14.3% of schools operated by for-profit EMOs had proficiency rates above state averages. Of ten virtual schools operated by nonprofit EMOs, five (50%) had proficiency rates above state averages.
Not many states still report adequate yearly progress (AYP), but among the three states with full-time virtual schools who did, the results were not positive. For example, in Washington, online students met the standard at a lower rate (35.3%) than the rates of their statewide 4th grade (61.8%) and 7th grade (58.6%) counterparts. However, they performed very comparably in reading. Many virtual schools did not report AYP data, even if their state still used AYP.
Many states have replaced AYP with some sort of school performance. Only 13.6% of the 457 full-time virtual schools received a school performance score, but among those that did, 30.6% were rated acceptable. Virtual schools not operated by EMOs were more likely to receive an acceptable rating than virtual schools operated by EMOs: 40.7% compared with 23.5%.
Graduation information was available for 28% of virtual schools and 32% of blended schools. Nationwide the average graduation rate was 81%; however, the on-time graduation rates for virtual and blended schools were 40.6% and 37.4% respectively. The graduation rates for virtual schools have worsened by 3% over the past few years, even though graduation rates overall have been improving about 1% each year.
Discussion & recommendations: There are several key findings from this report. Most importantly, this report deepens the findings from the past three years of NEPC reports that virtual schools do not perform as well academically as traditional schools. Blended schools were included in the study for the first time, and contrary to the authors’ expectations, they did not perform better than full-time virtual schools. However, this may be explained by the fact that blended schools enrolled more students living in poverty. Finally, it is interesting to note that district-operated virtual and blended schools both outperformed their charter school counterparts by upwards of 15% on some measures.
Given the poor performance of virtual and blended schools, the authors recommend that the growth of these schools be slowed or stopped until the reasons for their under-performance are identified and addressed. They also desire the implementation and enforcement of sanctions on virtual and blended schools that fail to prove that they are doing well by their students. Finally, they naturally wish for more data and fixed student to teacher ratios.
Appraisal: While the NEPC report provides the best compilation of the available information on virtual (and now blended) schools, it must be remembered that this data has severe limitations that the authors of the report seem to disregard at times, especially regarding academic achievement. For example, while it is true that all of the school performance measures such as test scores and graduation rates seemed stacked against virtual schools, it is troubling that these measures come from such a small percentage of virtual schools (13.6% for school performance scores; 28% for graduation rates, and approximately half of the schools in 3 states for AYP data). This is not NEPC’s fault since they merely aggregate the available data, which was scarcer than usual due to the shift to the Common Core. However, it is something that must be remembered when working with this data. Although the NEPC report continues to show that virtual schools are underperforming in spite of their student population which includes fewer minorities, fewer students in poverty, and fewer students with special needs, there are still questions to be answered about the efficacy of virtual schools. What are the reasons for their underperformance, especially their dismal graduation rates? Is it a function of the schools themselves or do for profit virtual schools somehow attract a disproportionate number of students who, though white, are nevertheless at risk of failure? Are the results reported here an accurate snapshot of virtual schools as a whole?
Disclaimer: The views expressed in reviews are not the official views of ICHER or of its members. For more information about ICHER’s Reviews, please see the «About these Reviews» Section.