Low-StakesCompletion-Based

FundingA New Approach to

Financing Competency-Based Education

Introduction

How Revenue is Earned

How Teachers are Compensated

Impact on School Fiscal Health & Sustainability

Discussion

Conclusion

For the past three years, the lead author of this report has been studying the costs of personalized and blended learning, the policy landscape of online charter schools, and the potential of school finance to act as a lever for personalized learning at scale. In most cases, he found that finance models, resource allocation formulas, and school spending patterns showed remarkable resemblances to those established in more traditional school systems. However, the completion-based funding model was a notable exception that embodied a fresh take on school funding.

While many states did not view online charter schools as an area for innovation within existing charter school regulations, the state of New Hampshire leapt into uncharted state policy and experimented with new pathways towards increasing student success. We are pleased to have had the chance to study New Hampshire’s unique funding system and have given it the name of low-stakes completion-based funding. We now share this report with state and school leaders who seek new possibilities for financing student learning in their virtual school sectors.

About this Report About the Authors

Larry Miller is Dean of the School of Education at Florida SouthWestern State College. His current school finance research agenda includes a long-term multi-state evaluation of completion-based funding and an early childhood education cost study. Reach him at larry.miller@fsw.edu and by Twitter @edfiguy.

Megan Just is a Senior Research Analyst at Florida SouthWestern State College. She is currently pursuing a Ph.D. in Educational Psychology at the University of South Florida with specialization in educational measurement and research methodology. Her current research examines three interrelated areas in creativity research: environments that either induce or inhibit creativity/innovation; creativity assessment; and 21st-century educational reforms focused on improving innovation. She may be reached at megan.just@fsw.edu and by Twitter @just4creativity.

Joy Cho is a Research Assistant and student at Florida SouthWestern State College. She possesses research experience in the fields of education finance and political psychology. She is anticipated to graduate in 2018 with a degree in Economics. Reach her at jcho@bucs.fsw.edu.

3

6

8

10

13

13

Table of Contents

1

© 2016 You may at all times use and distribute original copies of this report, or any portion thereof, as it appeared on the CIE website, whether for personal, professional, or teaching purposes. Portions of this document may be reprinted with permission from the Author. Author is the owner of the copyright to this report.

The National Center for Innovation in Education (CIE) was established in 2013 with support from the William and Flora Hewlett Foundation, as well as the Bill & Melinda Gates Foundation. CIE operates out of the College of Education at the University of Kentucky. It strives to be a national voice in building consensus around a new, more coherent vision of education, while putting learners and learning at the center of its mission. The research published by CIE helps state and local systems develop and act on robust theories of change. A unique aspect of CIE’s approach is its cooperative efforts with state education agencies and local districts on innovation priorities, in both policy and in practice, so that lessons learned can inform state policy change and yield a higher likelihood of enabling real transformation.

Florida SouthWestern State College (FSW) is a catalyst for creating an innovative education system that provides accessible educational pathways and prepares students to be enlightened, productive citizens. FSW’s School of Education prepares teachers for careers as early childhood and elementary educators. The college supplements these academic degrees with two on-campus early childhood development centers that integrate clinical education and research. The School of Education also boasts two open admission collegiate charter high schools on its campuses which are ranked 3rd and 10th among all high schools in Florida. We support our high school students to achieve college success through our dual- enrollment program, providing nearly 3,000 students each semester with the chance to take college-level courses for free.

Acknowledgements

This report benefited from the comprehensive feedback of several sources. Scott Milam provided many thoughtful comments and suggestions. Dr. Anne Angstrom gave the text a thorough perusal. Practitioners from more than a dozen states who attended our presentation at iNACOL in San Antonio pressed us to consider funding factors that matter to the schools they operate. We are indebted to Matthew Joseph and Karla Phillips for patiently listening to us present earlier drafts of this paper and pointing out areas

for improvement. Any errors that remain are our own. This report was made possible through the generous support of the Nellie Mae Education Foundation.

Suggested Citation: Miller, L., Just, M., & Cho, J. (2016). Low-stakes completion-based funding: What can we learn from the school that invented it? Lexington, KY: University of Kentucky Center for Innovation in Education.

2

States striving to see improved student outcomes from their schools are taking a fresh look at their funding formulas and asking if changing from an enrollment-based funding system to a performance-based funding system will dramatically improve student performance. A handful of states see their burgeoning online charter school sectors as an opportunity to experiment with new funding formulas.1 New Hampshire has developed a unique approach to performance-based budgeting—called completion-based funding (CBF)2—that seeks to improve student outcomes by funding schools when students complete assignments rather than when they enroll in or attend classes.

Although this is considered a novel approach to funding the American public school system, the fundamental concept is not new to policymakers. These same principles were called into practice in the 1700s over payments to British sea captains for transporting convicted felons to Australia. At the time, the mortality rate of passengers on this voyage stood at roughly 33%. This was seen as a problem, not only because the high death rate was indicative of inhumane travelling conditions, but also because it represented a significant setback in the number of civilians who could cultivate the new settlement. The government responded with a slew of regulatory measures and policy changes, all of which failed to increase the number of surviving passengers. Finally, survival rates increased when the government changed how it funded its sea captains. Instead of paying them at the time of departure for the number of passengers aboard ship, they began paying for the number of passengers alive at the time of arrival in the New World. The survival rate for the same voyage jumped to 99%.3

This story demonstrates the impact incentives can have on the success or failure of publicly funded programs. When education funding is enrollment-based, enrollment levels increase and receive considerable attention from school

1 New Hampshire, Minnesota, and Florida are using CBF to fund students taking online courses either on a part-time or full-time basis. Important differences exist in how states define completion. Florida measures completion by rate of students passing an end-of-course exam. Others, like Minnesota, define completion by when students simply finish their course. (Pazhouh, Lake, & Miller, 2015).2 Governments fund education on enrollment, employment or performance. Completion-based funding is a unique form of performance-based budgeting.3 (Kestenbaum, 2010).

leaders. When completed assignments are funded instead of enrollment, administrator and faculty focus is expected to shift from supporting the recruitment and enrollment process to supporting students completing assignments. The more assignments students complete, the more likely they are to graduate college and career ready.

The growing interest in performance-based funding systems coincides with a push by some education reform advocates to develop a broader set of student success indicators than those specified through standardized testing, such as portfolio assessments and competency-based achievements. A report on performance-based funding and online learning published by iNACOL named New Hampshire the best example of completion-based funding because NH uses competencies, rather than high stakes tests, as performance standards.4 New Hampshire’s sole online school, Virtual Learning Academy Charter School (VLACS)5, is a non-profit operator in a for-profit dominated industry.6 VLACS remains the only school in New Hampshire, district or charter operated, to be completion funded.

VLACS may be the only school in the nation to possess all three of these characteristics: completion-funded, competency-based, and not-for-profit.

New Hampshire’s funding system is thought to result in lower revenues per student compared with attendance-based funding systems used to fund online charter schools in states like Ohio and Pennsylvania, where some very large and profitable, but

4 (Patrick, Worthen, Frost, & Gentz, 2016)5 While VLACS remains peerless at the moment, their model draws from other pioneering organizations in this space. Big Picture Learning, a school whose unique instructional model puts students at the center of an advisor-parent-internship mentor triad, was cited by one school leader, Steve Kossakoski, as an influential pilot in competency-based online education. VLACS also benefited from a content sharing arrangement with Florida Virtual School (FLVS) during its first few years of operation. (Steve Kossakoski, Personal Interview, January 20, 2016).6 Roughly two-thirds of online charter schools are for-profit operators. (Pazhouh, Lake, & Miller, 2015).

Introduction

3

poorly performing schools have emerged.7 Several researchers have called for more nuanced understandings of state-sponsored virtual school operations, especially in regard to funding formulas used to allocate dollars.8 Rather than advocate for the elimination of online charter schools, we saw potential benefits that completion-based funding could offer students enrolled in online charter schools.

A first step toward understanding the potential impact CBF could have on student success and school funding levels

7 (Rich, 2016).8 (Baker & Miron, The business of charter schooling: Understanding the policies that charter operators use for financial benefit, 2015); (Miron & Gulosino, 2016); (Molnar, et al., 2015).

requires a basic understanding of what a completion-based funding formula looks like and how it manifests in state policy. This study also asked whether the VLACS’ version of CBF was associated with changes by school management, such as an adoption of a performance-based compensation for faculty that could be expected to positively impact student success. Finally, our research asked whether completion funding could jeopardize the fiscal health of VLACS, which could prevent the school from meeting basic minimum financial obligations and threaten the long-term viability of the model.9

9 This study is not designed to make causal claims about CBF’s effect on student achievement, though that would be an important study to be performed.

When VLACS first opened its servers in 2008, students could only take courses on a part-time basis. Now, they also offer full-time enrollment opportunities for students in grades 6-12. More recently, VLACS introduced four new learning pathways beyond courses: Projects, Experiences, Teams, and College.10 Full-time status is defined by the number of credits a student is enrolled in per year. At VLACs, a half credit is equivalent to a one-segment course and one credit is equal to a two-segment course. This would be comparable to a brick and mortar school, where a semester-long course would equal a half credit, and a full-year course, one credit. However, the word “semester” is interchanged with the word “segment” at VLACS because there are no time constraints for how quickly a student must finish that course. In order to be considered a full-time student at VLACS, one must be enrolled in a total of six credits, which can be comprised of any combination of half-credit and one-credit courses.2 For example, personal

10 (Programs and Courses, 2016). 11 (Steve Kossakoski, Personal Interview, October 18, 2016).

fitness is a half-credit course equivalent to one segment while algebra is a full-credit course equivalent to two segments.

According to VLACS’ website, the school served about 12,000 (headcount) students in 2014-15. Funding levels do not vary by student full-time status or by student need. For the most recently completed school year (2015-2016), VLACS reported 164 full-time students (151 in high school; 13 in middle school). Female students represented 61.6 percent of the VLACS full-time student body, in contrast to the 48.4 percent of female students statewide. The percentage of White students enrolled full-time at VLACS is 3.2 percentage points higher than the state rate of 86.7 percent. Full-time students dropped out of VLACS at a rate of nearly 3 percent in 2012. The dropout rate fell for two consecutive years until it was just below the statewide rate of about 1 percent, but rebounded in 2015 up to 3.95%.3 As of March 2016, average time employed for faculty was 2.8 years with range of 3 months to 8.2 years (N=156).

12 ([2011-2012 Early exit and dropout rates for grades 9-12], 2013). (2013 - 2014 Early exit and dropout rates for grades 9 - 12, 2015).

4

1211

VLACS Profile

5

Data & Methods

This single-site instrumental case study allowed us to identify and describe relationships between the funding system, school policies and practices, and the financial health of the school. The case study method is well-suited for descriptive research, and one of this study’s primary objectives was to provide a thorough description of the way that CBF works in practice at an online charter school.10

Qualitative data was obtained through on-site interviews and focus groups with school administrators (N=4), teachers (N=12) and full-time students (N=5) over three months. Follow up interviews were conducted on an as needed basis (N=5). Quantitative data was requested from VLACS about teacher and student characteristics, completion rates, and state funding. We augmented these data sources with policy documents, labor agreements, and publicly reported financial records drawn from the IRS 990 database.

10 A comparative case study was not possible because VLACS is the only online charter school in the state of New Hampshire and lacks comparable national peers.

13

13

Defining Our Terms

Competency-based education Competency-based education systems: (1) measure learning rather than seat time, (2) allow students to progress through material at their own pace, earning credit through multiple pathways; (3) improve student engagement by providing students options about the best way for them to learn and master new knowledge and skills.

Completion-based fundingCompletion-based funding compensates schools when students complete predefined milestones. Some states set high stakes milestones such as passing an end of course assessment, while other states set low stakes milestones, such as the successful completion of assignments. Completion-based funding’s acronym (CBF) should not be confused with funding for competency-based education (CBE), which would compensate schools when students demonstrate mastery over competencies. To the best of our knowledge, competency-based funding is not currently practiced in the United States.

Performance-based funding Performance-based funding is a catch-all phrase in public budgeting circles that relates to a number of different models that determine funding levels on the basis of one or more measures of an organization’s performance. Completion-based funding is one type of performance-based funding system. The term completion-based funding is more common in K12 systems, whereas performance-based funding is more commonly found in higher education systems.

Average daily membershipAverage Daily Membership (ADM) is an enrollment-based funding model that allocates funds based on the average number of students assigned to a school district during the school year.

Weighted student fundingWeighted student funding systems allocate higher per pupil funding rates to support students that are, on average, more expensive to educate to state standards. Common weighted student categories include students from economically disadvantaged backgrounds, students with special needs, and English language learners.

How Revenue is Earned

6

At VLACS, time spent learning and completing assignments is variable, meaning students are allowed as much or as little time needed to exhibit mastery over competencies in order to earn credit. In contrast to completion-based funding, enrollment-based funding works with Carnegie units because every student is funded for the same amount of seat time. By eliminating seat time requirements, VLACS needed a new funding mechanism to support its focus on student outcomes. Instead of assuming that the state required all schools to be funded by average daily membership (ADM), VLACS founder and current CEO Steve Kossakoski sought an alternative approach by assuming areas unaddressed by current statute were available. With this perspective, he created a funding system that met VLACS’ needs without violating existing state statute.

VLACS negotiated a memo of understanding (MOU) with the state that converts completions into membership, thus meeting the needs of both institutions. The MOU between New Hampshire and VLACS established that from 2009-2013, the online charter school would be funded at a non-negotiable rate of $5450 per student11, and a second MOU increased the student funding rate to $5498.30 for 2013-201512. According to Kossakoski, “Each biennium all charter schools submit projected enrollment numbers to the DOE. If the state budget is approved and enrollment numbers are accepted then the MOU is created based on the approved budget.”13 Also written into the enrollment agreement is a funding cap on the total amount VLACS can receive from the state. New Hampshire’s biennial budget funds VLACS via a line item allocation for two-year increments.

A potential cash flow problem for states and schools considering adopting a CBF system concerns reconciling the timing of incurred expenses to revenues earned. The timing becomes problematic when a state compensates schools after students complete a predefined milestone because it delays setting the budget until after the school year starts. And by then, schools

11 (Memorandum of understanding between the New Hampshire Department of Education and the Virtual Learning Academy Charter School, 2010) 12 (Memorandum of understanding between the New Hampshire Department of Education and the Virtual Learning Academy Charter School, 2013). Note: This funding rate appears in slight variations among different NHDOE reports and policy documents. However, the differences were deemed insignificant to warrant further mention. 13 (Steve Kossakoski, Personal Interview, October 18, 2016).

have incurred or encumbered most of their instructional expenses for the academic year. New Hampshire solves this dilemma by forward funding VLACS based on predicted completions each year and then reconciles predicted with actual completion rates at the end of the academic year. Any surplus or deficit carries over to the following year’s funding. This approach allows VLACS to operate without a line of credit, reducing operating costs and financial risk for the school.

States must decide from where to draw resources for online schools, especially for students taking only a few classes. They could require funding to follow students, in which case funds would be transferred from the sending school districts to online charter schools. The alternative approach introduces new funding into the system, allowing the sending district to retain the full student allocation while also compensating the receiving school upon successful student completion of the course.14 NH decided to increase funding and not require districts to pay VLACS tuition for part-time students. However, under this model, funding follows full-time students.

As the sole online provider for the state of New Hampshire, VLACS is able to capture the majority of students in New Hampshire who are interested in virtual schooling. The state preserved and protected VLACS’ sole provider status when it implemented the equivalent of a corporation’s “poison pill” in state statute, essentially reimbursing new online charter operators at a rate of more than $2000 less per student.

VLACS funding levels are not influenced by student demographics. A completion is funded at the same rate regardless of who earned it. A weighted completion would provide schools with additional resources for completions from students who qualify for supplemental services, including students with special needs, students from economically disadvantaged backgrounds, and English learners.

CBF is a low-stakes approach since it funds based on assignment completion rather than summative assessments. The low-stakes approach reduces the risk of VLACS losing full funding for students to whom VLACS has provided instruction, but who may be unsuccessful at completing the entire course.

14 This approach is called hold harmless funding (Miller, Roza, & Simburg, 2014)

17

14

15

16

14

15

16

17

Budget Alchemy: Converting Completions into Average Daily Membership

+ + + + =

VLACS earns funding based on the percentage of assignments a student completes, regardless of the amount of time a student spends enrolled in a course. VLACS calculates ‘credits earned’ per student based on that percentage. They then aggregate credits earned across their entire student body. The total number of credits earned by all students is divided by 6 to equal one full-time equivalent student. The resulting quotient is the VLACS’ ADM equivalent enrollment that is then multiplied by NH’s charter school student funding rate ($5,498 in 2015). In talking about how the VLACS funding system connects to school policies and practices, one administrator explained:

“Progress is actually how we make our earnings. So we get paid per student based on progress. [We know] this is when the student logged in. This is when the student submitted this. This is when the student took this quiz. This is when the student communicated with the teacher. All that data is collected” (January 20, 2016).

7

Completions for Funding, Competencies for Credit

To be clear, completion is the funding basis, and competencies are the accountability basis at VLACS. According to VLACS, “Competencies are the ‘big ideas’ that every student is expected to master. A traditional one-credit course is comprised of about eight different competencies. Students can master competencies by choosing a single learning path such as courses, projects, teams, or experiences — or they may blend these options together.”1 Competencies are assessed at periodic intervals in each course. Students must score an 85% or higher on these specified competency assignments to exhibit mastery of content. Class credit is awarded to students if they master every competency and earn at least a 60% average on all of the assignments in the course. Students are “highly encouraged” to move through the course in the sequence of material dictated as a best practice in optimizing learning; however, assignments are not locked or inaccessible to students since flexibility is a cornerstone of the model. Each student moves at a different pace through the material, completing assignments that lead to mastering competency-assessments. Competency-based assessments vary in format (e.g., traditional tests, performance-based, discussion-based, etc.) but are summative in nature. Students could hypothetically master all of the competency assignments and have less than a 100% completion. There is no difference between VLACS and other NH schools in terms of the student earning credit in this example; however, the state would allocate less funding to VLACS than other schools in the state based on its MOU.

This is where the distinction between completion and competency becomes evident. If VLACS were a competency-based funding model, it would bill the state for the full amount for every student who earned credit in their course because earning credit is synonymous with mastery of content. However, we know that VLACS bills the state on completions, which goes hand-in-hand with mastery of competencies most of the time --but not all of the time. A state could choose to fund students based on demonstrated mastery of competencies. The state would fund the student on the proportion of competencies mastered relative to the total number of competencies required by the course. A competency-based funding system would be higher stakes for three reasons: (1) because the funding increments would be larger (i.e., student funded at 20%, 40%, 60%,….) increasing funding uncertainty for the school; (2) a student must earn a score of 85 percent or higher to meet the mastery standard and that standard is higher than the standard used to classify an assignment as complete; (3) if schools were funded on competencies there would be a financial incentive for teachers to pass more students resulting in a conflict of interest between the accountability system and the funding system.

18 VLACS website. http://vlacs.org/middle-high-school/why-vlacs/why-vlacs-for-stu-dents-and-parents/flexible-learning-pathways/

18

How Teachers are Compensated

8

When conceptualizing this study, we hypothesized that VLACS would compensate teachers in a way that was aligned with student completions to ensure a balanced school budget. We were also interested in learning more about faculty compensation and its effect on teacher behaviors. If the new funding system is not positively impacting teachers and students, then why do it?

During the teacher focus group interviews, themes emerged surrounding student information systems and student-teacher relationships that helped to clarify for us how VLACS leveraged its funding to support student learning. In the following sections, we describe how CBF established an environment that influenced compensation policy, information systems, and an overarching instructional strategy to support student completion.

For full-time faculty, teachers are expected to deliver 120 earned credits per year. However, credits are counted differently for teachers when determining reconciliation than how they are counted for students earning credit for their courses. For teachers, one credit is equal to a two-segment course, and “earned” when 100 percent of the assignments in the course are completed by a student. The self-paced nature of VLACS and its rolling admissions means that students may be at various points in a course at the given time. Therefore, crowd-sourcing completed assignments from a larger population may be used to achieve the 120-credit standard and count toward teachers’ reconciliation when eligible. According to VLACS leadership, most instructors achieve the targeted number of credits.

To ensure financial sustainability, VLACS relies heavily on benefits-ineligible part-time instructors. Full- and part-time faculty are eligible to earn additional compensation when the number of student completions exceeds the agreed upon target. These additional payments to teachers come in the form of a reconciliation check rather than a bonus. Leaders conceptualized the payments as a quid pro quo: additional compensation for additional learning. Last year, 41% of part-time instructors earned a reconciliation check of an average $3027, and 27% of full-time faculty earned a reconciliation of an average $6229.15

Senior school administrators were concerned that bonuses incentivize faculty in a way that leads them to put pressure on students to complete competencies. We interpreted the language ‘reconciliation instead of bonuses’ as policy-making by metaphor since the administrators we interviewed explicitly sought to downplay the influence of money. As one leader put it,

“...there’s no part of what we do that ends up saying to the kid, you must do this because our funding is on the line, and instead of just worrying about learning, they’re worried about their school or worried about their teacher losing their job. I don’t think that is the type of pressure I’d ever want to put on a student. It just wouldn’t be right.”

15 (Steve Kossakoski, E-mail Correspondence, October 24, 2016.) 19

19

We asked school leaders what would change if the school was funded based on enrollment like other schools in New Hampshire. They acknowledged a connection between school practices (e.g., monitoring communication, completing homework assignments, etc.) and the funding system, which led them to openly consider ways to mediate this tension. VLACS employs two additional strategies to mitigate negative effects of a performance-based funding system. Faculty salaries are not reduced for missing completion goals; instead, those who significantly missed their targets are referred for professional development. The second strategy is to hire full-time faculty exclusively from the part-time faculty pool to ensure that faculty members have demonstrated successful teaching in VLACS’ unique online learning environment. Of the 156 instructors currently employed by VLACS, 21 (7.4%) instructors are full-time and 135 (91.6%) are part-time. For example, one teacher described VLACS’ compensation plan this way:

“So when we signed a contract, we’re going to get a certain amount of students. . . . I’m a full-time person with other responsibilities. I have a requirement of about between 70-75 students with a certain number of aggregate completions . . . So my goal would be roughly to complete 140 of those completions during the year based on my salary. If I go above that, I would get compensated for that. If I go below that, nothing happens. We’re just encouraged to just keep working along and recognize that there are challenges and that it is just a number.

Although faculty are not financially penalized, nor held accountable for student completion rates, they are held accountable for student success rates. The student success rate (SSR) measures the ratio of students who successfully completed their VLACS course out of the total number of students who enrolled.16 VLACS reports an SSR of 88 percent in the current academic year and has set a target of 90 percent for next year.17

Additionally, teachers reported their draw to working at VLACS due to their perceptions of making a difference through one-

16 (Administrative staff, Personal Interview, October 12, 2016). 17 (Annual report of progress, 2015).

on-one instruction with students, which is encouraged through school policies. One such policy, formed as a direct result of CBF, requires instructors to conduct weekly check-in calls with all students during the first 28 days they are enrolled in courses. After 28 days, instructors move to a monthly check-in call with parents and students together. The importance of these teacher-student relationships was supported in our interviews with faculty members:

“Well, I’ll just say the carrot for me has always been being able to work with students one-on-one. That’s always been the draw for me. Having taught in brick-and-mortar schools, which I’m sure a lot of you have taught in, and having 30 students in a classroom and going home and feeling like I just didn’t get through. I feel like I’m able to get through. That’s the carrot for me.”

“It feels a little bit more fulfilling too only because you can actually cater to what the students’ needs actually are. In that classroom of 30 or 35 kids, you have that group of six that can move a lot faster than that group over there, than this group over here. So you can have 50 kids in one Algebra I class, but the top 16 of them are speeding through as quickly as possible, but they have that straight-A average. Then, you might have a couple over here that started before them. They’re a little bit slower, but they need it, and they’re actually learning it and taking the time to understand it and to help them build on those building blocks. So you feel a little more validated as a teacher as opposed to being looked at as ‘Not everybody’s passing. Why not?’”

Faculty are empowered to leverage VLACS’ rolling admissions structure, which assigns students to courses when “seats” open up. During the first few weeks of class, a student is required to demonstrate adequate progress by completing required assignments determined by the teacher in order to remain enrolled. VLACS’ policy explicitly calls for a 28-day “grace period,” during which time no students are at risk of being dropped. After this cut date, however, teachers are given the discretion to evaluate student progress and to drop a student if they fail to respond to repeated communication attempts and fail to submit work on a regular basis. VLACS does not allow students to be dropped if they are having difficulty mastering competencies.

9

20

21

20

21

Impact on School Fiscal Health & Sustainability

10

Figure 1: A State By State Comparison of Funding Basis for Online Charter Schools

Fund

ing

Basi

s fo

r Onl

ine

Char

ter S

choo

ls

Funding Certainty for Online Charter Schools

High Low

Inpu

tO

utpu

t

High-StakesCompletion-Based

Funding

Low-StakesCompletion-Based

Funding

Average DailyMembership/Average

Daily Attendance

Single Count/DoubleCount Systems

NH

FL, TX

ADM: AZ, AK, NC, OH, OR, PA, SC, WI, IN, UTADA: ME, CA, ID

Single: NV, CO, LA Double: KS, GA, MI

VLACS created and adopted a funding system that increases the chances that they will not generate as much funding as they would have under an enrollment-based funding system. Figure 1 depicts the relationship between allocation basis and funding certainty for four different funding systems. As the allocation basis moves from enrollment, to membership, to

partial completion, and finally completion, funding certainty declines, and funding risk to the school increases. While VLACS did mitigate some of its funding risk by choosing a low-stakes version of CBF, it did not mitigate financial risk on the expense side of the ledger by tightly coupling faculty compensation with student completions.

11

CBF Results in Drastic Reduction of State AidImpact on Student Completion Rates Unclear

Year

Stat

e A

id (i

n M

illio

ns)

2011 2012 2013 2014

$14

$12

$10

$8

$6

$4

$2

$0

VLACS Revenue Under ADM

VLACS Revenue Under CBF

Figure 2: CBF Results in Drastic Reduction of State AidImpact on Student Completion Rates Unclear

Funding certainty and fiscal impact are abstract concepts until you connect real dollars with actual schools. Figure 2 depicts two revenue streams over time. The area in yellow represents VLACS’ actual revenue under CBF. The area above, in blue, represents VLACS’ predicted revenue under ADM. In the school year for which we have the most recent data (2013-2014), VLACS

earned $5.5 million in total state aid, but would have been allocated $12 million by ADM, a $6.5 million dollar difference. For the four-year period captured by this chart, NH allocated more than $21 million dollars less to VLACS under CBF than it would have under ADM. VLACS received less than half the state aid under CBF than it would have received under ADM.

12

Figure 3: Programmatic Investments Rise at the Expense of Overhead Spending

Now we explore VLACS resource allocation patterns. Figure 3 presents VLACS annual expenses by program services, management and general expenses, and other expenses from 2009 – 2015. The data for this chart was drawn from audited financial data drawn from IRS 990 returns. Fiscally healthy schools spend a majority of their resources on program services, including teacher salaries and benefits. VLACS allocated 61 percent of its expenses to program services, on average, over the seven-year period for which data is available. Program services expenses as a percentage of total expenses increased along with enrollment over time. For the first three years of operation (2009-2011), VLACS program expenses represented 56 percent

of total expenses. For the last three years of operation (2013-2015), program expenses represent 68 percent of total expenses, a 12 percentage point increase. To fund a higher relative share of spending on program services, VLACS has steadily reduced spending on management and general expenses. During the first three years of operation, management and general expenses were 20 percent of total expenses, on average. Over the last three years, that percentage fell to 4 percent, a 16 percentage point drop. A separate analysis by the authors finds that VLACS operates on a break-even basis, generating a small surplus or deficit each year.

Program Services

Management & General Expenses

Other Expenses

2009 2010 2011 2012 2013 2014 2015

Year

Shar

e of

Tot

al E

xpen

ses

100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%

13

Future Directions

Discussion

Completion-based funding (CBF) brings with it new funding risks otherwise unseen by traditional brick-and-mortar schools. CBF makes VLACS completely dependent on the New Hampshire legislature and the governor for its budget. This means that VLACS’ funding sources are less reliable than traditional schools, which are supported by both state aid and local property taxes and which default to last year’s funding level if school budgets do not pass. VLACS’ fiscal dependency on the state could be lessened if the hold harmless provision were lifted and districts were forced to share funding with VLACS. The state could reduce spending, but it could also result in a smaller VLACS. For example, Florida Virtual is a much smaller organization today as a result of a similar policy change.18

Whenever the idea of performance management is discussed in education circles, critics will raise concerns over performance pressure driving unintended behavior by teachers and students.19 In a high stakes, high pressure version of CBF, teachers could feel that their jobs are in jeopardy if students do not complete assignments and could then inappropriately make students aware of these concerns. At every stage of this analysis, we found VLACS leaders downplaying, minimizing, and protecting teachers and students from performance pressure.

CBF incentivizes schools to increase student assignment completion rates. For an incentive to be effective, conflicts of interest must be acknowledged and avoided. For example, care must be taken to ensure that teacher assessments of student completion are not tied to personal gain. VLACS accomplishes this objective by aligning funding with completions, and student credits and teacher evaluation with competencies.

Our analysis makes clear that CBF allocates far fewer dollars to VLACS than typical funding formulas. One could argue that it is within VLACS’ power to generate more revenue by increasing student completion rates. But the key question—yet to be completely answered—is whether CBF reduces the cost of education when compared to the cost of education under

18 (Gartner, 2013).

19 (Hill, 2016).

enrollment-based funding systems.20

It could be argued that some of NH’s policies could be strengthened to improve student outcomes and reduce spending. Following in Florida’s footsteps, NH could require a higher stakes, all or nothing, version of CBF that only funds students when a high stakes test is passed.21 NH could reduce education spending by eliminating hold harmless funding. NH could encourage additional online schools to open in NH, using competition and choice to improve completion outcomes for students who choose to learn online. But it must be acknowledged that CBF allocates to VLACS substantially fewer dollars than ADM. It must also be acknowledged that a low stakes, less competitive system may be the reason why VLACS is able to strike a balance between focusing on performance in low pressure way that mitigates unintended consequences of trying funding to student completions.

Conclusion

The most admirable aspect of CBF is the incredible vision exhibited by the leaders in NH’s Department of Education and the founders of VLACS when they orchestrated an agreement that converted student completions into average daily membership. We encourage leaders of school systems, online or not, to be inspired by this innovative approach to designing a funding system that supports the needs of the schools you lead and the students you serve.

CBF provides states with a possible new approach to funding schools with implications for both adults and students. CBF changes the incentives for school administrators, teachers, and students, alike. Administrators are less confident about revenue levels; they must design and implement performance management systems focused on student completion. Teachers no longer view student progress as fixed; they must work to build stronger relationships with students in order to increase their likelihood of completing courses. Students can no longer just show up to school; they must complete assignments to progress. This study describes in detail how VLACS designed and implemented CBF. What is needed now is a better understanding of the relationship between CBF and student outcomes.

20 (Bowen, Lack, Chingos, & Nygren, 2012). 21 (Pazhouh, Lake, & Miller, 2015).

22

23

24

25

22

23

24

25

Arkansas Department of Education. (2015). Arkansas school finance manual 2015-2016. Retrieved from http://www.arkansased.gov/public/userfiles/Fiscal_and_Admin_Services/Publication%20and%20reports/Arkansas_School_Finance_ Manual/Arkansas_School_Finance_Manual_FY16.pdfBaker, B. and Bathon, J. (2013). Financing online education and virtual schooling: A guide for policymakers and advocates. Retrieved from National Education Policy Center website: http://nepc.colorado.edu/files/lb-pb-onlineedfinancing-policy_0.pdfBaker, B. and Miron, G. (2015). The business of charter schooling: Understanding the policies that charter operators use for financial benefit. Retrieved from National Education Policy Center website: http://nepc.colorado.edu/files/rb_baker-miron_charter_revenue_0.pdfBattaglino, T., Haldeman, M., and Laurans, E. (2012). The costs of online learning. Retrieved from Thomas B. Fordham Institute website: https://edex.s3-us-west-2.amazonaws.com/publication/pdfs/20120110-the-costs-of-online-learning_7.pdfBerdik, C. (2016, August). Inside the online school that could radically change how kids learn everywhere. Wired. Retrieved from https://www.wired.com/2016/08/inside-online-school-radically-change-kids-learn-everywhere/Borup, J., Graham, C., and Drysdale, J. (2014). The nature of teacher engagement at an online high school. British Journal of Educational Technology, 45(5).Bowen, W., Lack, K., Chingos, M., and Nygren, T. (2012). Interactive learning online at public universities. Ithaka S+R. doi: http://dx.doi.org/10.18665/sr.22464California Legislative Information. (n.d.). Education code title 2, division 3, part 22, chapter 5, article 4 [37640]. Retrieved from http://leginfo.legislature.ca.gov/faces/codes_displayText.xhtml?lawCode=EDC&division=3.&title=2.&part=22.&chapter=5.&article=4.Code of Colorado Regulations. (2012). Emergency rules 2233104.6-R-1.00 through -9.07. Retrieved from http://www.sos.state.co.us/CCR/GenerateRulePdf.do?ruleVersionId=816Colorado General Assembly. (2003). Session laws of Colorado 2003 first regular session, 64th general assembly. Retrieved from http://tornado.state.co.us/gov_dir/leg_dir/olls/sl2003a/sl_337.htmGartner, L. (2013, August 7) Florida Virtual School lays off hundreds as enrollment plummets. Tampa Bay Times. Retrieved from http://www.tampabay.com/news/education/k12/florida-virtual-school-lays-off-hundreds-as-enrollment-plummets/2135327Haynes, E., Zeiser, K., Surr, W., Hauser, A., Clymer, L., Walston, J., Bitter, C., and Yang, R. (2016). Looking under the hood of competency-based education: The relationships between competency-based education practices and students’ learning skills, behaviors, and dispositions. Retrieved from Nellie Mae Education Foundation website: https://www.nmefoundation.org/getattachment/Resources/Competency- Based-Pathways/Looking-Under-The-Hood-Of-Competency-Based-Educati/CBE-Study-Full-Report-(2).pdf?ext=.pdfHill, P. (2016, September 27). Wells Fargo and the Atlanta schools testing scandal. [Web log post]. Retrieved from http://www.crpe.org/thelens/wells-fargo-atlanta-schoolsIdaho Legislature. (2016). Idaho statutes title 33 chapter 52. Retrieved from https://legislature.idaho.gov/idstat/Title33/T33CH52SECT33-5208.htmIndiana Charter School Board. (n.d.). For schools: Funding. Retrieved from https://www.in.gov/icsb/2431.htmKansas State Department of Education. (2016). Kansas virtual education requirements and monitoring plan. Retrieved from http://www.ksde.org/Portals/0/TLA/Graduation%20and%20School%20Choice/Virtual/Updated%20Virtual%20Education%20 Requirements%20in%20Kansas.pdfKestenbaum, D. (2010, September 10). Pop quiz: How do you stop sea captains from killing their passengers? NPR. Retrieved from http://www. npr.org/sections/money/2010/09/09/129757852/pop-quiz-how-do-you-stop-sea-captains-from-killing-their-passengersKim, P., Kim, H., and Karimi, A. (2012). Public online charter schools students: Choices, perceptions, and traits. American Educational Research Journal, 49(3), 521. Louisiana Department of Education. (2016). FY 2016-17 minimum foundation program formula. Retrieved from https://www.louisianabelieves.com/docs/default-source/minimum-foundation-program/proposed-fy2016-17-mfp-resolution--- submitted-to-legislature-march-15-2016.pdf?sfvrsn=2Maine Department of Education. (n.d.). Aggregate daily attendance reporting. Retrieved from http://www.maine.gov/education/adaform/adadefinitions.htmMichigan Legislature. (2007). The State School Aid Act of 1979 (excerpt) Act 94 of 1979. Retrieved from http://www.legislature.mi.gov/(S(sbauvzfnrazf4mtq33yu43cb))/mileg.aspx?page=getObject&objectName=mcl-388-1606aMiller, L., Gross, B., Maas, T., Hernandez, J., Lu, A., and Lake, R. (2016). Financial personalized learning: What can we learn from first-generation adopters? Retrieved from the Center for Reinventing Public Education: http://www.crpe.org/sites/default/files/blfinance_4.2016_ final_1.pdfMiller, L., Roza, M., and Simburg, S. (2014). Funding for students’ sake: How to stop financing tomorrow’s schools based on yesterday’s priorities. Retrieved from Edunomics Lab website: http://edunomicslab.org/wp-content/uploads/2014/05/chap2_SEAF3_Miller_Roza_Simburg.pdfMiron, G. and Gulosino, C. (2016). Virtual schools report 2016. Retrieved from National Education Policy Center website: http://nepc.colorado.edu/files/publications/RB-Miron%20Virtual%20Schools.pdfMiron, G. and Urschel, J. (2012). Understanding and improving full-time virtual schools. Retrieved from National Education Policy Center website: http://nepc.colorado.edu/files/nepcrbk12miron.pdf

References

14

Molnar, A., Huerta, L., Barbour, M. K., Miron, G., Shafer, S. R., and Gulosino, C. (2015). Virtual schools in the U.S. 2015: Politics, performance, policy, and research evidence. Retrieved from National Education Policy Center website: http://nepc.colorado.edu/publication/virtual-schools-annual-2015National Alliance for Public Charter Schools. (2016). A call to action to improve the quality of full-time virtual charter public schools. Retrieved from http://www.publiccharters.org/wp-content/uploads/2016/06/VirtualReport_Web614.pdfNevada Legislative Counsel Bureau Fiscal Division. (2015). The Nevada Plan for school finance. Retrieved from http://www.leg.state.nv.us/division/fiscal/NevadaPlan/Nevada_Plan.pdfNew Hampshire Department of Education. (2010). Memorandum of understanding between the New Hampshire Department of Education and the Virtual Learning Academy Charter School. New Hampshire Department of Education. (2011). Charter school per pupil aid 2010-2011. Retrieved from http://education.nh.gov/data/documents/charter_school_aid_10-11.pdfNew Hampshire Department of Education. (2012). Charter school per pupil aid 2011-2012. Retrieved from http://education.nh.gov/data/documents/charter_school_aid_11-12.pdfNew Hampshire Department of Education. (2013). [2011-2012 Early exit and dropout rates for grades 9-12]. Retrieved from http://education.nh.gov/data/documents/dropouts11-12.pdfNew Hampshire Department of Education. (2013). Charter school per pupil aid 2012-2013. Retrieved from http://education.nh.gov/data/documents/charter_school_aid12_13.pdfNew Hampshire Department of Education. (2013). Memorandum of understanding between the New Hampshire Department of Education and the Virtual Learning Academy Charter School. New Hampshire Department of Education. (2014). Charter school per pupil aid 2013-2014. Retrieved from http://education.nh.gov/data/documents/charter_school_aid13_14.pdfNew Hampshire Department of Education. (2014). NH pace flexibility waiver insert. Retrieved from http://education.nh.gov/assessment-systems/documents/waiver-insert.pdfNew Hampshire Department of Education. (2015). 2013-2014 Early exit and dropout rates for grades 9-12. Retrieved from http://education.nh.gov/data/documents/dropouts13-14.pdfNew Hampshire Department of Education. (2015). Charter school per pupil aid 2014-2015. Retrieved from http://education.nh.gov/data/documents/charter_school_aid14_15.pdfNew Hampshire General Court. (n.d.). Title XV education chapter 194-b chartered public schools section 194-b:11. Retrieved from http://www.gencourt.state.nh.us/rsa/html/XV/194-B/194-B-11.htmNew Hampshire General Court. (n.d.). Title XV education chapter 198 school money adequate education; Education trust fund section 198:40-a. Retrieved from http://www.gencourt.state.nh.us/rsa/html/xv/198/198-40-a.htmNew Hampshire State Board of Education. (2014). Minimum standards for public school approval. Retrieved from http://education.nh.gov/legislation/documents/ed3062014-min-stands.pdfNoor, S. (2016). Oregon fall membership manual for the 2016-17 school year. Retrieved from Oregon Department of Education website: http://www.ode.state.or.us/wma/data/schoolanddistrict/students/docs/fallmembershipmanual2016-17.pdfNorth Carolina General Assembly. (2015). State and local funds for charter schools. Retrieved from http://www.ncga.state.nc.us/Sessions/2015/Bills/House/PDF/H1111v0.pdfOhio Department of Education. (2010). ADM used in funding calculations under AM SUB HB 1. Retrieved from http://education.ohio.gov/Topics/Finance-and-Funding/School-Payment-Reports/State-Funding-For-Schools/Financial-Reports/ADM- Used-in-Funding-Calculations-under-AM-SUB-HB-1Patrick, S., Myers, J., Silverstein, J., Brown, A., and Watson, J. (2015). Performance-based funding & online learning: Maximizing resources for student success. Retrieved from International Association for K-12 Online Learning website: http://www.inacol.org/wp-content/uploads/2015/03/iNACOL-Performance-Based-Funding-and-Online-Learning.pdfPatrick, S., Worthen, M., Frost, D., and Gentz, S. (2016). Promising state policies for personalized learning. Retrieved from International Association for K-12 Online Learning website: http://www.inacol.org/wp-content/uploads/2016/05/iNACOL-Promising-State-Policies-for- Personalized-Learning.pdfPazhouh, R., Lake, R., and Miller, L. (2015). The policy framework for online charter schools. Retrieved from Center for Reinventing Public Education website: http://www.crpe.org/sites/default/files/crpe-policy-framework-online-charter-schools-final_0.pdfPennsylvania Department of Education. (n.d.). Charter school funding. Retrieved from http://www.education.pa.gov/K-12/Charter%20Schools/Pages/Charter-School-Funding.aspx#tab-1Programs and Courses. (2016). Retrieved from Virtual Learning Academy Charter School website: http://vlacs.org/middle-high-school/programs-courses/Prothero, A. (2016, August 11). Ohio must rethink how online charter schools are funded, says state’s auditor. [Web log post]. Retrieved from http://blogs.edweek.org/edweek/charterschoice/2016/08/ohio_must_rethink_online_charter_school_funding_says_auditor.htmlPublic Schools of North Carolina. (2011). North Carolina virtual public schools (NCVPS) allotment formula. Retrieved from http://www.ncpublicschools.org/docs/fbs/allotments/ncvps/formula.pdf

15

Rice, J. (2012). Review of “The costs of learning.” Retrieved from National Education Policy Center website: http://nepc.colorado.edu/files/TTR-OnLineLrng-Rice_2.pdfRich, M. (2016, May 18). Online school enriches affiliated companies if not its students. New York Times. Retrieved from http://www.nytimes.com/2016/05/19/us/online-charter-schools-electronic-classroom-of-tomorrow.htmlRichmond, G. (2016, July 6). Online charters mostly don’t work. Education Next. Retrieved from http://educationnext.org/online-charters-mostly-dont-work-forum-virtual-schools-greg-richmond/Russ, K. (2015). Charter schools informational paper 27. Retrieved from the Wisconsin Department of Public Instruction website: http://dpi.wi.gov/sites/default/files/imce/sms/Charter-Schools/LFB%20Info%20Paper%2027-Jan%202015.pdfStake, R. E. (2005). Qualitative case studies. In R. E. Stake, N. Denzin, Y. Lincoln (Eds.), The Sage Handbook of Qualitative Research. London, England: Sage Publications.Steve Kossakoski, Personal Interview, January 20, 2016. Steve Kossakoski, Personal Interview, October 18, 2016. Steve Kossakoski, Personal Interview, October 24, 2016. South Carolina Legislature. (1996). South Carolina Charter Schools Act of 1996. Retrieved from http://www.scstatehouse.gov/code/t59c040.phpTexas Constitution and Statutes. (2013). Education code title 2, subtitle f, chapter 30a, subchapter a, section 30a.153. Retrieved from http://www.statutes.legis.state.tx.us/Docs/ED/htm/ED.30A.htmUtah Department of Administrative Services. (2016). R277-419-6 Student membership calculations. Retrieved from http://www.rules.utah.gov/publicat/code/r277/r277-419.htm#T6Vander Ark, T. (2016, July 6). Online charters expand learning options. Education Next. Retrieved from http://educationnext.org/online-charters-expand-learning-options-forum-virtual-schools-tom-vander-ark/Virtual Learning Academy Charter School. (2015). Annual report of progress.Virtual Learning Academy Charter School administrative staff, Personal Interview, October 12, 2016.Virtual Learning Academy Charter School teacher focus group, Personal Interview, January 20, 2016.Waters, L., Barbour, M., and Menchaca, M. (2014). The nature of online charter schools: Evolution and emerging concerns. Educational Technology & Society, 17(4), 379-389. Watson, J., Pape, L., Murin, A., Gemin, B., and Vashaw, L. (2014). Keeping pace with k-12 digital learning: An annual review of policy and practice. Retrieved from Keeping Pace with K-12 Digital Learning website: http://www.kpk12.com/wp-content/uploads/Evergreen_ KeepingPace_2015.pdfWoods, Richard. (2015). FY 2016 GNETS program frequently asked questions. Retrieved from Georgia Department of Education website: https://www.gadoe.org/Technology-Services/Data-Collections/Documents/FTE%20Resources/FY2016/FY2016%20GNETS%20FAQ.pdfYuan, K., Vi-Hhuan, L., McCaffrey, D., Marsh, J., Hamilton, L., Stecher, B., and Springer, M. (2013). Incentive pay programs do not affect teacher motivation or reported practices: Results from three randomized studies. Educational Evaluation and Policy Analysis, 35(2), 3-22.

16