Total enrollment at Wisconsin’s public and charter schools declined by 25,000 students from fall 2019 to fall 2020 and has continued to fall in the years since then. While declining birth rates and apparent movement of some students to private or home schools likely account for at least two thirds of the decline, somewhere between 0.5% and 1.2% of the state’s school-age population may be unaccounted for, raising questions about their whereabouts and well-being, the state’s workforce, school finances, and more.

PreK-12 enrollment in Wisconsin’s public schools plummeted in the 2020-21 school year by over 25,000 students, or 2.9%, the largest single-year decline in at least 35 years. Previous Forum research documented this drop occurring alongside a decrease in private school enrollment, an increase in homeschool enrollment, and the apparent choice by many parents and guardians to delay sending their youngest children to pre-kindergarten and kindergarten at the height of the COVID-19 pandemic.

As the immediate impacts of COVID-19 subsided, some may have expected – and indeed hoped – to see enrollment numbers rebound. Instead, fall 2021 and fall 2022 data show that public school enrollment kept falling, by an additional 792 students in 2021-22 and another 6,339 students in 2022-23, down to 822,804. The total three-year decline was 32,155 students (rounded elsewhere in this report to 32,000).

Some decrease each year is expected given the state’s falling birth rate. The degree of actual enrollment declines, however, has exceeded some projections and raises the question of whether the state’s schools have recovered the students lost during the peak pandemic years.

To help answer those questions, we analyzed both public school enrollment data and figures on private school and homeschool enrollment, as well as overall population patterns. (Charter school enrollment is included in the public school data.) Our analysis does not include still-unreleased data from the newly begun 2023-24 school year.

We estimate that the movement of students to private schools and homeschooling, in addition to the declining birth rate, accounts for roughly 20,600 to 27,700 students out of the total decline of more than 32,000 students in public school enrollment from fall 2019 to fall 2022. Assuming we have accurately estimated the effect of falling birthrates, that leaves approximately 4,500 to 11,600 students not captured in current data sets. These students would be missing either because the data on them was not collected or, worse, because they disconnected from the education system in Wisconsin entirely (see Figure 1).

To account for private school and homeschool student populations, we drew on datasets published by the Wisconsin Department of Public Instruction (DPI): detailed enrollment data for both public and private schools available via the agency’s WISEdash tool, as well as DPI’s collection of homeschooling statistics. While data from last year for public and private schools throughout the state was made available in March 2023, homeschool data was not published until June. Full data for the current school year will not be made available until next spring.

Where have Wisconsin public school students gone?

To understand where students no longer in the public school system have gone, we first examined private school and homeschool enrollment data. Like public school enrollment, private school enrollment also fell in fall 2020, albeit to a smaller degree: by 1,843 students or 1.5%, down to 118,862 students. Unlike public enrollment, however, it rebounded in fall 2021, jumping by 2.4% to 121,729 students, and increased again in fall 2022 by 2.2% to 124,447 students (see Table 1). That left it 3,742 students above its fall 2019 pre-pandemic level and represented the largest private school total since fall 2011.

Private school enrollment, therefore, increased by more than 5,000 students statewide in just two years from fall 2020 to fall 2022, which has not happened since at least fall 1984. Some of this increase can be attributed to the rebound from the initial drop that happened concurrently with the public school exodus, with families re-engaging in 2021-22 and beyond. Still, private school enrollment gains after consistent, longstanding declines – enrollment was nearly 154,000 in fall 1984 – are worth a deeper look separate from this analysis.

Homeschool enrollment, meanwhile, soared in fall 2020 by 47.3% or 10,234, up to 31,878 students. While still a relatively small proportion of the total student population in Wisconsin, this increase indicates that some parents and guardians likely chose to educate their children at home rather than via public schools, thereby accounting for a portion of the 25,000-student drop in fall 2020 public school enrollment. Homeschool enrollment has since declined, dropping by 2,476 students or 7.8% in fall 2021, and again by 1.9% or another 549 students in fall 2022.

Still, homeschool enrollment – which had been between 18,000 and 22,000 students in each year from fall 1998 to fall 2019 – sat last year at 28,853 students, or 33.3% above pre-pandemic levels.

In addition, DPI officials note that private and home schools are not held to the same standard as public schools when it comes to reporting enrollment data, meaning that there are likely private and homeschool students who exist in the state but are not captured by DPI’s counts.

Together, private school and homeschool enrollment thus could account for just under 11,000 of the total 32,000-student public school enrollment drop from fall 2019 to fall 2022: 3,742 from private schools and 7,209 from homeschool. To be clear, these numbers might not be wholly composed of students previously enrolled in public schools; some proportion may be new students who never previously attended a public school. There would also have been students in private school and homeschool settings who moved into the public schools during this time period, but those individual transitions are obscured by the larger aggregate trend in the opposite direction.

This lack of specificity in the available public data is also why we tracked “missing students” based off of public school enrollment losses rather than total losses: not because the public school sector is the only one from which students may have disconnected but rather because the net gains in private school and homeschool enrollment from fall 2019 to fall 2022 mean that the state’s net losses are only visible in the public school numbers.

The next step in evaluating the true amount of students unaccounted for is acknowledging Wisconsin’s declining birth rate. Since the early 2000s, fewer births have driven a corresponding decline in total student enrollment in the state that remains ongoing. In the five years prior to the pandemic, the state each year lost an average of 3,416 students, or 0.32% of total enrollment. If we isolate solely the decline in fall 2019 – the last academic year before the first “full” pandemic school year – statewide enrollment decreased by 5,642 students, or 0.56%. We use these two numbers (0.32% and 0.56%) as a possible range for the effect of this pre-existing trend on enrollment in each of the last three school years.

If we assume a loss of 0.32% of the full state student population each year – a modest view of the expected effects of declining enrollment prior to COVID-19 – then the pre-pandemic trend would account for 9,600 of the total students lost from fall 2019 to fall 2022. Added to the approximately 11,000 attributed to private school and homeschool enrollment, this estimate leaves us with a loss of 11,600 students that we have still not explained.

If we instead assume a loss of 0.56% each year – a more aggressive view of the pre-pandemic trend – then it would account for approximately 16,800 of the total enrollment drop from fall 2019 to fall 2022. Added to the approximately 11,000 attributed to private school and homeschool enrollment, this estimate leaves us with a still unexplained loss of 4,500 students. Experts can debate what the trend would have been without a pandemic, but as a rule, the more enrollment loss assumed to be inevitable means fewer missing students to explain.

Figure 1 shows the divergence of these pre-pandemic trends from the state’s actual enrollment. Figure 2 illustrates our understanding of the causes of the divergence.

In total, therefore, we believe that prior to the pandemic we would have expected there to be 4,500 to 11,600 more students in schools in the fall of 2022 than are captured in existing data sets. These students would represent a range of 0.46% to 1.19% of total statewide enrollment in the 2022-23 school year, or 14.0% to 36.1% of the total 32,000-student drop from fall 2019 to fall 2022.

This estimate aligns with a wide-ranging national effort to account for “missing students” across the United States, where the total public elementary and secondary school enrollment drop from fall 2019 to fall 2021 topped 1.4 million. A 2023 collaboration between Stanford University, Big Local News, and the Associated Press used available public, private, and home school data from 2019-20 through 2021-22 to find that approximately a third of the examined public school enrollment decline could not be explained by overall demographic decline or by shifts to private or home schools.

Difficulties in identifying missing students

Publicly available data offer some clues as to the schools and students most affected by this issue. Firm conclusions, however, are not possible since the data do not include individual student identifiers that would have allowed us to track students moving between systems.

For example, the state’s districts with the largest enrollments and the state’s urban districts – two groups with a great deal of overlap – saw disproportionate drops in student enrollment (see Figure 3). While public school enrollment in Wisconsin declined by 3.8% from fall 2019 to fall 2022, 10 of the state’s 12 largest enrollment districts saw declines larger than that, including Milwaukee (-9.6%), Kenosha (-8.1%), and Racine (-7.7%). The state’s 12 largest districts accounted for 26.9% of public school enrollment in Wisconsin in fall 2022 but 52.8% of the enrollment decline since fall 2019.

City districts, as categorized by the National Center for Education Statistics, saw a similar phenomenon: Leaving out independent charter schools, the state’s urban districts accounted for 29.0% of public enrollment in fall 2022 but 52.3% of the decline since fall 2019. This over-representation is not unique to the pandemic years: from 2017 to 2020, city districts (again, excluding independent charters) accounted for 68.8% of lost enrollment but only 29.9% of 2020’s statewide public enrollment.

Notably, the districts with the highest enrollment and those located in cities tend to have more education options available than other districts. Students can move with relative ease between sectors in those districts and, in at least some cases, the data suggest that they did. From fall 2019 to fall 2022, while statewide private school enrollment increased by only 3.1%, enrollment in Milwaukee’s private schools rose from 30,512 to 31,982, an increase of 4.8%. Enrollment in Madison’s private schools rose by 9.6%, from 4,099 to 4,491. Still, these private school increases do not appear to account for the whole of the public school losses.

Another factor to consider is which student groups contributed the most to the enrollment drops. Statewide public school enrollment of economically disadvantaged students dropped 6.3% from fall 2019 to fall 2022, compared to a 1.9% drop for students who are not economically disadvantaged. Because household economic status was not a stable indicator during the pandemic due to emergency federal funds and other relief measures, however, we cannot assume that these are directly comparable years.

Statewide public school enrollment of students with disabilities actually saw an increase of 1.8% from fall 2019 to fall 2022. In contrast, students without disabilities declined by 4.7%. Like household economic status, however, it is difficult to isolate the degree to which changes for this population are due to students entering or leaving school as opposed to students being newly identified as having disabilities.

By race and ethnicity, the largest percentage decreases in public school enrollment from fall 2019 to fall 2022 occurred among students identifying as American Indian (-9.3%), Black (-6.6%), and white (-6.0%). Asian student enrollment also decreased (-1.1%). On the other hand, Hispanic and Pacific Isle student enrollments both increased over the same time period (4.1% and 0.8%, respectively). Enrollment of students identifying as two or more races increased the most (13.5%), which could mean that some of the apparent changes over time in other groups are related to students changing how they identify.

It is also notable that enrollment figures plunged for migrant students – defined as the children of migratory dairy or agricultural workers who move between districts as a result of that work. This is a small group of students – fewer than 900 in each year since fall 2009 – yet they saw a marked drop of 50.2% in enrollment from fall 2019 to fall 2022, reaching their lowest levels since at least fall 2005.

The state may wish to investigate this dramatic decline to understand whether it is due to migratory labor changes causing fewer of these workers and their families reside in Wisconsin or to engage in migratory work, whether fewer students are otherwise getting categorized as migrant students, or whether these are truly “missing students” who have historically been more challenging to count and even more so during the turmoil of the pandemic. In any case, this change is an important one and may hint at a larger factor in the overall enrollment decline, as we will further discuss below.

On the whole, the state’s urban districts and districts with the greatest enrollment may have the highest proportions of students missing. The data do not tell the full story, but state and local leaders attempting to find their missing students may look first among low-income, American Indian, Black, white, or migratory households.

Youngest students may be the key

Our 2021 research identified that pre-kindergarten and kindergarten enrollment saw the greatest declines from fall 2019 to fall 2020, at least in part due to families opting to delay sending their youngest children to school during the pandemic. Attendance in Wisconsin schools is not mandatory until age six, meaning that many of these children could remain out of school without consequence at least for the first year of COVID-19. In addition, some students may have been homeschooled but not included in statewide numbers because homeschool enrollment collection in Wisconsin does not begin until the first grade.

In this research we explored whether the enduring declines remained concentrated in the lowest grades. We found that from fall 2019 to fall 2022, the largest grade-level declines occurred across two broad groups (see Figure 4).

The first was those same PreK and kindergarten students. DPI uses three categories for public preschool: K3 (for three-year-olds), K4 (for four-year olds), and PK (for children under six receiving services for disabilities from a school district who are not otherwise enrolled). Combined, enrollment for these groups declined by 10.2%, from 56,338 in fall 2019 to 50,612 in fall 2022. Meanwhile, kindergarten enrollment fell by 5.9%, from a similar 56,756 in fall 2019 to 53,432 in fall 2022. As our 2021 research indicated, the enrollment drops likely point to a large proportion of students who are either no longer in the state, or who are still present but are either being homeschooled or not being formally educated at all.

National data also suggest the pandemic may have caused parents to delay or avoid enrolling students in pre-kindergarten. Data from the American Community Survey shows that from 2019 to 2021, the proportion of 3-to-6-year-olds enrolled in preschool declined from 51.1% to 41.8%, and that preschool enrollment is now at its lowest rate since at least 2005. Enrollment rates in public preschools dropped from 30.5% to 24.0% across the country, while rates in private preschools declined from 20.6% to 17.8%.

These numbers do not necessarily mean that the children who remained in the state were entirely disconnected from education and care. As mentioned, the Wisconsin enrollment numbers do not include homeschooled children until first grade, and it is also possible that families placed their children in childcare settings offering full-day care rather than PreK if half-day PreK was the only in-school offering available. The pandemic-era Child Care Counts program may have kept these childcare offerings more affordable and accessible in recent years, although this program is currently set to expire in January 2024 unless legislative action is taken.

Still, with early education representing some of the most critical years in a child’s development, any apparent retreat from preschool enrollment would merit attention from families, communities, school officials, and policymakers.

The other grades that saw higher-than-average declines were fifth (-7.5%), sixth (-8.3%), and seventh (-7.9%). These drops have a different explanation: around the time of the Great Recession, there appears to have been a “demographic cliff” in the amount of children born in the state. The high school graduating class of 2027 – born in late 2008 and early 2009 – was a few thousand students larger than the graduating class of 2028 born just one year later as the recession took its toll. The effect of depressed birth rates seems to have extended for two additional years, manifesting in lower enrollment numbers for the middle grades in fall 2020 through fall 2022.

We cannot wholly attribute the decreases in public school enrollment, however, to the low birth rates affecting the middle grades and to the apparent opt-outs in the youngest grades. Throughout the elementary and middle school years, we see fewer students in fall 2022 than expected. (High school projections are more difficult to examine because of the high mobility between private and public school at that age.)

Specifically, prior to the pandemic, a group of public school students starting in one elementary or middle grade and then progressing as an age cohort through later grades would typically gain students over time. For example, there were 58,662 public school third graders in fall 2005. By the time this age cohort reached sixth grade in fall 2008, there were 60,233 students in their statewide class. Cohort sizes might typically increase in later grades as children reach the age of mandatory school attendance or parents move their children out of private or home school.

In the pandemic era, however, this trend reversed: Age cohorts lost students in fall 2020 that they have yet to regain. To take the same example, the statewide sixth grade public school class in fall 2022 had 629 fewer students than that same age cohort’s third grade class in fall 2019. Similar public school cohort loss occurred across kindergarteners through fifth graders in fall 2019, who reached third through eighth graders in fall 2022. In each case, the loss of students from the age cohort between fall 2019 and fall 2020 was not fully made up by fall 2022. As we might expect from our previous analysis, the younger students saw the greatest cohort losses, but the widespread nature of the losses indicates that the public school enrollment drop is not solely due to factors affecting the youngest children. Public schools also lost students who were already well established in the public school setting.

As in our previous analyses, it is likely that the loss of enrollment across age cohorts is due at least in part to students moving to private or home schools. However, we do not have the data to fully account for grade level enrollment outside public schools, as both private and home schools can choose to report their students either by grade level or as “ungraded.”

Possible explanations, Implications, and solutions

A myriad of possible factors may explain any additional drops in student enrollment not tied to movement between school sectors, pre-existing declining enrollment trends, or conditions unique to the state’s youngest students.

First, the pandemic may have prompted or forced families with children in Wisconsin to move to other states or countries. That could include international university students with children of their own as well as immigrant parents working in Wisconsin, either with or without authorization to do so. The huge drop in migrant students previously examined could be indicative of such movement. And, notably, the industries hit hardest by the pandemic – restaurants, hotels, and other leisure and hospitality businesses – relied heavily on immigrant labor, who would have been disproportionately affected by layoffs.

On the other hand, the early months of the pandemic restricted international movement. In addition, the state’s English Learner enrollment has nearly fully recovered (only 0.1% fewer students in fall 2022 than in fall 2019), and this group bears some relation to the state’s immigrant student population. Neighboring states have also seen public school enrollment losses, making outmigration to other states less likely as well though still possible.

Next, emerging national studies have documented a rise in student anxiety, disconnection, and other mental health concerns that could lead youth to disengage with school and perhaps drop out entirely, especially if the pandemic destabilized their family and made district outreach difficult. School closures also may have unintentionally communicated to students that school was optional. Sharp increases in both statewide and national chronic absenteeism rates further highlight the degree to which the pandemic upended students’ relationship to school.

Older students in theory could have left school to join the workforce or care for younger siblings during COVID-19. Yet these hypotheses seem less likely to be major factors since the data do not indicate large scale drop-outs in the higher grade levels. It is possible that high school enrollment would have seen greater volatility were it not for the state’s ENGAGE Wisconsin program, a statewide partnership between DPI and the Graduation Alliance. DPI estimates that between 3,000 and 4,000 students were reconnected to school thanks to the program.

Last, possible inaccuracies in the dataset itself might account for some of the missing students. The majority of the reported enrollment decline occurred between fall 2019 and fall 2020, during a year marked by frequent student illness and quarantines, shifts between learning modalities, and changing guidance on how to report students.

Whatever the true number of missing students, every child disconnected from school – whether they left or failed to enter the state’s formal education system – should be cause for concern. Beyond the crucial impacts to the children themselves, the loss of students also deprives the state of educated workers and future economic growth.

Efforts to identify and, as appropriate, reconnect missing students may happen on a local or state level. DPI has noted that federal pandemic relief funds can be used to pay for “outreach and service delivery to special populations” such as disconnected youth. Those funds are set to expire in September 2024. Leveraging any remaining dollars before then may benefit children as well as district finances in light of the link between enrollment and school revenues.

Existing initiatives to reconnect chronically absent or truant students will also likely overlap with the concerns of missing youth. Wisconsin State Assembly Speaker Robin Vos recently announced a bipartisan task force on truancy. The statewide ENGAGE Wisconsin program remains operational through the 2023-24 school year, and districts can still apply for this free support. DPI officials, drawing on input from the State Superintendent’s Youth Council, have encouraged schools to involve students and their families in their own education as a way of re-engaging those who are disconnected.

For school districts, these efforts to reclaim students have an added urgency. The loss of students will result in substantial negative financial impacts to districts, which can lose state and federal funding when enrollment declines. Each student lost means less available in the long term to pay teachers, procure materials, and improve curricula. Many of these costs do not diminish on a per-student basis, putting pressure on districts’ budgets. Some experts are predicting a wave of disruptive school closures nationally as enrollments fall across much of the country.

Ultimately, it may be difficult at this point to undo the movement of students during the pandemic, whether from one type of school to another or from Wisconsin to other parts of the nation or world. However, there is still work to do to reconnect with any missing students, rebuild relationships with families, and find sustainable and fair solutions to the state’s declining public school enrollment. This demographic shift will continue even if every missing student is recovered. Indeed, according to recently released projections from UW-Madison’s Applied Population Lab, public school districts will lose on average 11,000 students annually over the next five years. Though this reality may be difficult to change, thoughtful action by state and district leaders may help families, communities, and schools to better navigate it.