The Future of Families and Child Wellbeing Study (FFCWS, formerly known as the Fragile Families and Child Wellbeing Study) follows a cohort of nearly 5,000 children born in large, U.S. cities between 1998 and 2000. The study oversampled births to unmarried couples; and, when weighted, the data are representative of births in large U.S. cities at the turn of the century. The FFCWS was originally designed to address four questions of great interest to researchers and policy makers: What are the conditions and capabilities of unmarried parents, especially fathers? What is the nature of the relationships between unmarried parents? How do children born into these families fare? How do policies and environmental conditions affect families and children? The FFCWS consists of interviews with mothers, fathers, and/or primary caregivers at birth and again when children are ages 1, 3, 5, 9, 15, and 22. The parent interviews collected information on attitudes, relationships, parenting behavior, demographic characteristics, health (mental and physical), economic and employment status, neighborhood characteristics, and program participation. Beginning at age 9, children were interviewed directly (either during the home visit or on the telephone). The direct child interviews collected data on family relationships, home routines, schools, peers, and physical and mental health, as well as health behaviors. A collaborative study of the FFCWS, the In-Home Longitudinal Study of Pre-School Aged Children (In-Home Study) collected data from a subset of the FFCWS Core respondents at the Year 3 and 5 follow-ups to ask how parental resources in the form of parental presence or absence, time, and money influence children under the age of 5. The In-Home Study collected information on a variety of domains of the child's environment, including: the physical environment (quality of housing, nutrition and food security, health care, adequacy of clothing and supervision) and parenting (parental discipline, parental attachment, and cognitive stimulation). In addition, the In-Home Study also collected information on several important child outcomes, including anthropometrics, child behaviors, and cognitive ability. This information was collected through interviews with the child's primary caregiver, and direct observation of the child's home environment and the child's interactions with his or her caregiver. Similar activities were conducted during the Year 9 follow-up. At the Year 15 follow-up, a condensed set of home visit activities were conducted with a subsample of approximately 1,000 teens. Teens who participated in the In-Home Study were also invited to participate in a Sleep Study and were asked to wear an accelerometer on their non-dominant wrist for seven consecutive days to track their sleep (Sleep Actigraphy Data) and that day's behaviors and mood (Daily Sleep Actigraphy and Diary Survey Data). An additional collaborative study collected data from the child care provider (Year 3) and teacher (Years 9 and 15) through mail-based surveys. Saliva samples were collected at Year 9 and 15 (Biomarker file and Polygenic Scores). The Study of Adolescent Neural Development (SAND) COVID Study began data collection in May 2020 following the onset of the COVID-19 pandemic. It included online surveys with the young adult and their primary caregiver. The FFCWS began its seventh wave of data collection in October 2020, around the focal child's 22nd birthday. Data collection and interviews continued through January 2024. The Year 22 wave included a young adult (YA) survey with the original focal child and a primary caregiver (PCG) survey. Data were also collected on the children of the original focal child (referred to as Generation 3, or G3). Documentation for these files is available on the FFCWS website located here. For details of updates made to the FFCWS data files, please see the project's Data Alerts page. Data collection for the Future of Families and Child Wellbeing Study was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the National Institutes of Health under award numbers R01HD36916, R01HD39135, and R01HD40421, as well as a consortium of private foundations. Below is the citation for use of the FFCWS data accessed through ICPSR. For information on additional citation requirements when

Children by Family Type reports the number and percent of children living in families by child age and by family type. Domain

Since 1960, the U.S. Department of Agriculture has provided estimates of expenditures on children from birth through age 17. This technical report presents the most recent estimates for married- couple and single-parent families using data from the 2011-15 Consumer Expenditure Survey (all data presented in 2015 dollars). Data and methods used in calculating annual child-rearing expenses are described. Estimates are provided for married-couple and single-parent families with two children for major components of the budget by age of child, family income, and region of residence. For the overall United States, annual child-rearing expense estimates ranged between $12,350 and $13,900 for a child in a two-child, married-couple family in the middle-income group. Adjustment factors for households with less than or greater than two children are also provided. Expenses vary considerably by household income level, region, and composition, emphasizing that a single estimate may not be applicable to all families. Results of this study may be of use in developing State child support and foster care guidelines, as well as public health and family-centered educational programs. i

Children in Single-parent Families reports the percent of children living in families that are headed by a single parent. Dimensions Year;Measure Type;Variable Full Description Children are all persons under the age of 18 years, living in families, and related as children by birth, marriage, or adoption to the householder. Children living with married step-parents are not included. Single-parent families may include unmarried couples. The American Community Survey (ACS) collects these data from a sample of households on a rolling monthly basis. ACS aggregates samples into one-, three-, or five-year periods. At this time only state-level annual data are available on CTdata.org. Town-level data aggregated from the five-year datasets (considered to be more accurate for geographic areas that are the size of a county or smaller) can be produced using Census tables currently available on the Census website.

This Child Welfare Information Gateway Podcast is available on Apple Podcasts , GooglePlay , Spotify , Stitcher , SoundCloud , and the Child Welfare Information Gateway website. Subscribe to receive new episodes as they are released. Metadata-only record linking to the original dataset. Open original dataset below.

About the dataset This dataset uses information from the DWP benefit system to provide estimates of children living in poverty for wards in London. In order to be counted in this dataset, a family must have claimed Child Benefit and at least one other household benefit (Universal Credit, tax credits or Housing Benefit) during the year. The numbers are calibrated to the Households Below Average Income (HBAI) dataset used to provide the government's headline poverty statistics. The definition of relative low income is living in a household with equivalised* income before housing costs (BHC) below 60% of contemporary national median income. The income measure includes contributions from earnings, state support and pensions. Further detail on the estimates of dependent children living in relative low income, including alternative geographical breakdowns and additional variables, such as age of children, family type and work status are available from DWP's statistical tabulation tool Stat-Xplore. Minor adjustments to the data have been applied to guard against the identification of individual claimants. This dataset replaced the DWP children in out-of-work benefit households and HMRC children in low income families local measure releases. This dataset includes estimates for all wards in London of numbers of dependent children living in relative low income families for each financial year from 2014/15 to the latest available (2022/23). The figures for the latest year are provisional and are subject to minor revision when the next dataset is released by DWP. Headlines Number of children The number of dependent children living in relative low income across London, rose from below 310,000 in the financial year ending 2015 to over 420,000 in the financial year ending 2020, but has decreased since then to below 350,000, which is well below the number for financial year ending 2018. While many wards in London have followed a similar pattern, the numbers of children in low income families in some wards have fallen more sharply, while the numbers in other wards have continued to grow. Proportion of children in each London ward Ward population sizes vary across London, the age profile of that population also varies and both the size and make-up of the population can change over time, so in order to make more meaningful comparisons between wards or over time, DWP have also published rates, though see note below regarding caution when using these figures. A dependent child is anyone aged under 16; or aged 16 to 19 in full-time non-advanced education or in unwaged government training. Ward level estimates for the total number of dependent children are not available, so percentages cannot be derived. Ward level estimates for the percentage of children under 16 living in low income families are usually published by DWP but, in its latest release, ward-level population estimates were not available at the time, so no rates were published. To derive the rates in this dataset, the GLA has used the ONS's latest ward-level population estimates (official statistics in development). Percentages for 2021/22 are calculated using the 2021 mid year estimates, while percentages for 2022/23 are calculated using the 2022 mid year estimates. As these are official statistics in development, rates therefore need to be treated with some caution. Notes *equivalised income is adjusted for household size and composition in order to compare living standards between households of different types.

This issue of Child and Family Services Reviews Update contains the following sections: Round 4 Year 1 States are Gearing Up, New Mock Case Course Released on E-Learning Academy, Spanish Translations Added to Portal, New Round 4 FAQs Posted, and Child Welfare Reviews Project Upcoming Presentations. Metadata-only record linking to the original dataset. Open original dataset below.

This dataset shows official annual experimental statistics for numbers and percentages of Children age under 16 living in Relative and Absolute low income families, by Local Authority District and Ward. More detailed data breakdowns (such as Age of Child, Family Type and Work Status, plus data for other small area geographies and trend data), are available at the Source link. Percentages are calculated by dividing the number of children age 0-15 living in low income families by resident children age 0-15 from mid-year population estimates. The latest data is marked P for Provisional and is subject to future revision. Data source: Department for Work and Pensions and HM Revenue and Customs. Updates are according to government statistics releases. For more information about this data and its methodology, please see the Source link.

Full Description Children are all persons under the age of 18 years. 'Own children' in a family are sons and daughters, including stepchildren and adopted children, of the householder. 'Single-parent family' means only one parent is present in the home, and is never-married, widowed, divorced, or married, spouse absent. This data originates from the American Community Survey (ACS) 5-Year estimates, table B11003. The ACS collects these data from a sample of households on a rolling monthly basis. ACS aggregates samples into one-, three-, or five-year periods.

This public health factsheet describes facts, assets, and strategies related to families and young children’s health in Camden.

In 1990, the United States Congress authorized a major program designed to enhance the early public school transitions of former Head Start children and their families. Former Head Start children, like many other children living in poverty, were at risk for poor school achievement. This new program was launched to test the value of extending comprehensive, Head Start-like supports "upward" through the first four years of elementary school. This project, administered by the Head Start Bureau of the Administration on Children, Youth, and Families, funded 31 local Transition Demonstration Programs in 30 states and the Navajo Nation from the 1991-1992 school year through the 1997-1998 school year and involved more than 450 public schools. The National Transition Demonstration Study was conducted to provide information about the implementation of this program and its impact on children, families, schools, and communities. More than 7,500 former Head Start children and families were enrolled in the National Study. Thousands of other children and families, however, participated in the Transition Demonstration Program, since supports and educational enhancements were offered to all children and families in the classrooms. The datasets are organized into four broad categories: Family Units -- There are six family unit files. A "family unit" record consists of information about a child or family as the result of source data taken from family interviews, records of child test scores on a child instrument, school archival records, or teacher questionnaires (Part B). If data were available from any combination of these source documents, a family unit record was generated. A broad range of variables are included under this heading. Variables range from simple demographics to standardized scores of social skill ratings as well as neighborhood factor scores and child outcome scores in reading and mathematics. These files are associated with each year of the child's schooling (kindergarten through third grade). School Unit -- There are five school unit files, organized around the year of data collection. A "school unit" record consists of information about a school as the result of source data captured from family interviews, a classroom teacher, or the school principal. The structure of this data file is different from others in that rather than being merged on a common key, the records are actually stacked one upon the other in groups. The first part of the file consists of family data, the middle portion consists of teacher data, and the final portion consists of principal data. A key variable to the construction of this dataset is the REC_SRC (record source) variable. It informs the user as to the source of the data in the record. The abbreviations are "fi," "ta," and "qp" for family interview, teacher questionnaire (Part A), and questionnaire for principals, respectively. The data viewed as the centerpiece of these datasets are the school climate survey variables and their associated factor scores. Classroom Unit -- There are five classroom unit files organized around the year of data collection. The data recorded focus on the classroom and are from the following sources: classroom composition, assessment profile, a developmentally appropriate practice template, and a teacher questionnaire (part a). Some of the data available address the social skills the teacher views as important to his or her particular classroom. Variables addressing diversity of both gender and ethnicity within a single classroom are included when available. Exit Interviews -- There are five Exit Interviews. Exit information was collected from the following groups: experimental and control families, family service specialists, school principals, and classroom teachers. These exit interviews were conducted upon exit from the third grade, and have been combined for both cohorts. Community Characteristics Data -- The community characteristics dataset is a hierarchical file having four distinct levels of data. The type of information available in this file may include data that describe the site, county, school district, or study school. Users of these data are strongly encouraged to consult the accompanying documentation before attempting to use these files.

Users can request data and reports related, but not limited to child abuse, neglect, foster care, and child well-being. Background The National Data Archive on Child Abuse and Neglect collects data on the well-being of children. The archive is a project of the Family Life Development Center, Department of Human Ecology at Cornell University. The archive collects data sets from the The National Survey of Child Health and Well-being, The Adoption and Foster Care Analysis Reporting System, The National Child Abuse and Neglect Data System, and other data related to child abuse, neglect, victimization, m altreatment, sexual abuse, homelessness, and safety. User functionality Users can access abstracts of data sets which discuss the time period and logistics of collecting the data. There are different requirements for accessing different data sets. All requirements are clearly outlined. All data sets must be ordered through the National Data Archive on Child Abuse and Neglect. Application materials must be mailed to the archive for access permission. Requirements for access vary by amount of personal information included in the data set. Data Notes The chief investigator, the years of data collection and a description of the data set is available on the website for every data set. The website does not convey when new data sets will be added.

Healthy Families New York (HFNY), which was based on the Healthy Families America (HFA) model, was established as a strengths-based, intensive home visitation program with the explicit goals of promoting positive parenting skills and parent-child interaction; preventing child abuse and neglect; supporting optimal prenatal care, and child health and development; and improving parent's self-sufficiency.In 2000, a randomized controlled trial was initiated at three sites with the HFNY home visiting program. Families eligible for HFNY at each site were randomly assigned to either an intervention group that was offered HFNY services or to a control group that was given information on and referral to appropriate services other than home visiting. Baseline interviews were conducted with 1,173 of the eligible women (intervention, n=579; control, n=594), and follow up interviews at Years 1, 2, and 3. In addition to data gathered during the follow up interviews, information regarding study participants' involvement in reports of child maltreatment was also extracted and coded from Child Protection Services records.For the current study, mothers in both the intervention and control groups were re-interviewed at the time of the target child's seventh birthday. Interviews (Dataset 1: Mother Interview Data, n=942) included information about parenting, the child, earnings, and household composition. Interviewers also completed face-to-face assessments (Dataset 2: Target Child Interview Data) with 800 of the children who were born and reached the age of 7 at the time of interview. The target child interviews assessed children's receptive vocabulary skills, emotional health, self-regulatory abilities, and problem behaviors. The research team also extracted or obtained administrative data pertaining to Child Protective Service reports, foster care placements, federal and state supported benefits, and programs services and costs (Datasets 3-8).

This webinar was primarily for those states that have incorporated Strengthening Families as part of their child abuse and neglect prevention efforts. Strengthening Families is a strategy for engaging early care and education providers and others working with families with young children in building protective factors that have been shown by research to be connected to child abuse and neglect prevention. The webinar included discussion of new resources; new tools and new directions that can help states sustain and deepen their existing Strengthening Families work.

Presenter: Nilofer Ahsan, Center for the Study of Social Policy

View Webinar (WMV - 46 MB)

Metadata-only record linking to the original dataset. Open original dataset below.

This issue of Child and Family Services Reviews Update contains the following sections: New Resources Added to CFSR Information, Portal Round 4 OMS: Preparing for Launch This Summer, Round 4 Data Profiles Coming in August 2022, CFSR Unit Members and States.

Metadata-only record linking to the original dataset. Open original dataset below.

Background:
The Millennium Cohort Study (MCS) is a large-scale, multi-purpose longitudinal dataset providing information about babies born at the beginning of the 21st century, their progress through life, and the families who are bringing them up, for the four countries of the United Kingdom. The original objectives of the first MCS survey, as laid down in the proposal to the Economic and Social Research Council (ESRC) in March 2000, were:

• to chart the initial conditions of social, economic and health advantages and disadvantages facing children born at the start of the 21st century, capturing information that the research community of the future will require
• to provide a basis for comparing patterns of development with the preceding cohorts (the National Child Development Study, held at the UK Data Archive under GN 33004, and the 1970 Birth Cohort Study, held under GN 33229)
• to collect information on previously neglected topics, such as fathers' involvement in children's care and development
• to focus on parents as the most immediate elements of the children's 'background', charting their experience as mothers and fathers of newborn babies in the year 2000, recording how they (and any other children in the family) adapted to the newcomer, and what their aspirations for her/his future may be
• to emphasise intergenerational links including those back to the parents' own childhood
• to investigate the wider social ecology of the family, including social networks, civic engagement and community facilities and services, splicing in geo-coded data when available

Additional objectives subsequently included for MCS were:

• to provide control cases for the national evaluation of Sure Start (a government programme intended to alleviate child poverty and social exclusion)
• to provide samples of adequate size to analyse and compare the smaller countries of the United Kingdom, and include disadvantaged areas of England

Further information about the MCS can be found on the Centre for Longitudinal Studies web pages.

The content of MCS studies, including questions, topics and variables can be explored via the CLOSER Discovery website.

The first sweep (MCS1) interviewed both mothers and (where resident) fathers (or father-figures) of infants included in the sample when the babies were nine months old, and the second sweep (MCS2) was carried out with the same respondents when the children were three years of age. The third sweep (MCS3) was conducted in 2006, when the children were aged five years old, the fourth sweep (MCS4) in 2008, when they were seven years old, the fifth sweep (MCS5) in 2012-2013, when they were eleven years old, the sixth sweep (MCS6) in 2015, when they were fourteen years old, and the seventh sweep (MCS7) in 2018, when they were seventeen years old.

Safeguarded versions of MCS studies:
The Safeguarded versions of MCS1, MCS2, MCS3, MCS4, MCS5, MCS6 and MCS7 are held under UK Data Archive SNs 4683, 5350, 5795, 6411, 7464, 8156 and 8682 respectively. The longitudinal family file is held under SN 8172.

Polygenic Indices
Polygenic indices are available under Special Licence SN 9437. Derived summary scores have been created that combine the estimated effects of many different genes on a specific trait or characteristic, such as a person's risk of Alzheimer's disease, asthma, substance abuse, or mental health disorders, for example. These polygenic scores can be combined with existing survey data to offer a more nuanced understanding of how cohort members' outcomes may be shaped.

Sub-sample studies:
Some studies based on sub-samples of MCS have also been conducted, including a study of MCS respondent mothers who had received assisted fertility treatment, conducted in 2003 (see EUL SN 5559). Also, birth registration and maternity hospital episodes for the MCS respondents are held as a separate dataset (see EUL SN 5614).

Release of Sweeps 1 to 4 to Long Format (Summer 2020)
To support longitudinal research and make it easier to compare data from different time points, all data from across all sweeps is now in a consistent format. The update affects the data from sweeps 1 to 4 (from 9 months to 7 years), which are updated from the old/wide to a new/long format to match the format of data of sweeps 5 and 6 (age 11 and 14 sweeps). The old/wide formatted datasets contained one row per family with multiple variables for different respondents. The new/long formatted datasets contain one row per respondent (per parent or per cohort member) for each MCS family. Additional updates have been made to all sweeps to harmonise variable labels and enhance anonymisation.

How to access genetic and/or bio-medical sample data from a range of longitudinal surveys:
For information on how to access biomedical data from MCS that are not held at the UKDS, see the CLS Genetic data and biological samples webpage.

Secure Access datasets:
Secure Access versions of the MCS have more restrictive access conditions than versions available under the standard Safeguarded Licence or Special Licence (see 'Access data' tab above).

Secure Access versions of the MCS include:

• detailed sensitive variables not available under EUL. These have been grouped thematically and are held under SN 8753 (socio-economic, accommodation and occupational data), SN 8754 (self-reported health, behaviour and fertility), SN 8755 (demographics, language and religion) and SN 8756 (exact participation dates). These files replace previously available studies held under SNs 8456 and 8622-8627
  
• detailed geographical identifier files which are grouped by sweep held under SN 7758 (MCS1), SN 7759 (MCS2), SN 7760 (MCS3), SN 7761 (MCS4), SN 7762 (MCS5 2001 Census Boundaries), SN 7763 (MCS5 2011 Census Boundaries), SN 8231 (MCS6 2001 Census Boundaries), SN 8232 (MCS6 2011 Census Boundaries), SN 8757 (MCS7), SN 8758 (MCS7 2001 Census Boundaries) and SN 8759 (MCS7 2011 Census Boundaries). These files replace previously available files grouped by geography SN 7049 (Ward level), SN 7050 (Lower Super Output Area level), and SN 7051 (Output Area level)
• linked education administrative datasets for Key Stages 1, 2, 4 and 5 held under SN 8481 (England). This replaces previously available datasets for Key Stage 1 (SN 6862) and Key Stage 2 (SN 7712)
  
• linked education administrative datasets for Key Stage 1 held under SN 7414 (Scotland)
• linked education administrative dataset for Key Stages 1, 2, 3 and 4 under SN 9085 (Wales)
  
• linked NHS Patient Episode Database for Wales (PEDW) for MCS1 – MCS5 held under SN 8302
• linked Scottish Medical Records data held under SNs 8709, 8710, 8711, 8712, 8713 and 8714;
• Banded Distances to English Grammar Schools for MCS5 held under SN 8394
• linked Health Administrative Datasets (Hospital Episode Statistics) for England for years 2000-2019 held under SN 9030
• linked Hospital of Birth data held under SN 5724.

The linked education administrative datasets held under SNs 8481,7414 and 9085 may be ordered alongside the MCS detailed geographical identifier files only if sufficient justification is provided in the application.

Researchers applying for access to the Secure Access MCS datasets should indicate on their ESRC Accredited Researcher application form the EUL dataset(s) that they also wish to access (selected from the MCS Series Access web page).

SN 8156 - Millennium Cohort Study: Age 14, Sweep 6, 2015
The sixth sweep of the Millennium Cohort Study was carried out when the cohort members were 14 years old. As 14 is a key transitional age, the sweep was purposefully ambitious in the breadth and scope of its contents. It included: an interview (CAPI and CASI) with the main parent and their partner (where relevant); a self-completion interview with the cohort members; cognitive assessments for the main parent, the partner and the cohort member; DNA collection of the cohort member and natural parents in the household; physical measurements of the cohort member; placement of a time use diary with the cohort member; placement of an accelerometer with the cohort member.

Latest edition information
For the eighth edition (September 2025), three new harmonised Time Use Diary data files (calendar format, episode level and derived variables) have been added to the study, along with an updated user guide. These files replace previously available data and documentation.

ACF Children Bureau resource

Metadata-only record linking to the original dataset. Open original dataset below.

Content

Household is an occupied housing unit. Householder is a person in whose name the housing unit is rented or owned. This person must be at least 15 years old. Family household is a household in which there is at least 1 person present who is related to the householder by birth, marriage or adoption. Family is used to refer to a family household. In general, family consists of those related to each other by birth, marriage or adoption.

This data uses the householder's person weight to describe characteristics of people living in households. As a result, estimates of the number of households do not match estimates of households from the Housing Vacancy Survey (HVS). The HVS is weighted to housing units, rather than the population, in order to more accurately estimate the number of occupied and vacant housing units. For more information about the source and accuracy statement of the Annual Social and Economic Supplement (ASEC) of the Current Population Survey (CPS) see the technical documentation accessible at: http://www.census.gov/programs-surveys/cps/technical-documentation/complete.html

Context

This is a dataset from the U.S. Census Bureau hosted by the Federal Reserve Economic Database (FRED). FRED has a data platform found here and they update their information according the amount of data that is brought in. Explore the U.S. Census Bureau using Kaggle and all of the data sources available through the U.S. Census Bureau organization page!

• Update Frequency: This dataset is updated daily.

• Observation Start: 1950-01-01

• Observation End : 2019-01-01

Acknowledgements

This dataset is maintained using FRED's API and Kaggle's API.

A. SUMMARY San Francisco offers numerous events and activities tailored for children, youth, and families. However, finding and navigating the disparate sources of information can be a major challenge. Our415.org seeks to simplify this by consolidating all relevant details, ensuring that families can easily find what they need, when they need it. It also encourages discovery of new interests and things to do. This dataset includes data on services targeted for children, youth, and their families in San Francisco.

B. HOW THE DATASET IS CREATED This dataset includes data collected from various sources including DCYF’s Contract Management System, Jotform, and data shares from various departments in the City.

C. UPDATE PROCESS The dataset will be updated on a daily basis, reflecting changes to the source data.

D. HOW TO USE THIS DATASET This dataset can be joined to the Organizations dataset (CHF-0012) using the ID field from the Organizations dataset and Organization ID field in the Services dataset.

E. RELATED DATASETS Organizations serving Children, Youth, and their Families

The Rise of Home Schooling

Data from The Post's analysis of home-schooling enrollment across the US

This repository shares data hand-collected by The Washington Post from individual school districts and states as a whole regarding home-school enrollment from 2017-18 through 2022-23. The data is what is behind this story published on Oct. 31, 2023 by Peter Jamison, Laura Meckler, Prayag Gordy, Clara Ence Morse and Chris Alcantara.

There are two separate data files, both of which cover the same time period: - home_school_district.csv - home_school_state.csv

There is also a data dictionary explaining each file.

Methodology

To measure the growth of home schooling during the pandemic, The Washington Post collected home-school student counts from 6,738 school districts. Together with students from The Washington Post Investigative Reporting Workshop practicum at American University, reporters trawled state websites, contacted education officials in all 50 states and the District of Columbia and submitted multiple public records requests for an annual count of home-schoolers from the 2017-18 school year through 2022-23. The Post ultimately collected data for all public school districts in 29 states and D.C. In all, The Post gathered data from states representing 61% of the American school-age population.

Three states — Pennsylvania, Rhode Island and Tennessee — have not published the number of home-schoolers in 2022-23, and Maine only shared district-level data starting with the 2020-21 school year. In seven states, The Post was unable to obtain usable home-school enrollment figures: In Arizona, Nevada and Oregon, only new home-school registrations are tracked annually at the district level; in North Carolina, home-school registration rolls are not regularly purged as students age out of the system; and in West Virginia, Utah and Alabama, annual enrollment data is unavailable. Eleven additional states do not require any notice when families decide to home-school their children, so enrollment figures in those states are also unavailable. Finally, Montana, Vermont and Nebraska collect data at a county level, not a district level, so there is no district data available - only statewide figures.

The Post made every effort to capture all legal ways to home-school, which vary by state. However, data on home schools established by certain methods, such as registering one’s home-school as a private school, are tracked by some states but not others. That means The Post’s tally is almost certainly an undercount, even in the states from which it gathered data. For instance, Wisconsin and Georgia only provided The Post with tallies of home-schoolers who had submitted required forms electronically. In Kentucky, some districts incorrectly reported zero or one home-schooled students in certain years, which a state education official attributed to an unclear form. The Post excluded those enrollment figures from its analysis. In California, which does not explicitly permit home schooling, many parents operate home-based private schools. The California Department of Education characterizes private schools with five or fewer students as home schools. In Louisiana, many home schools operate as nonpublic schools not seeking accreditation; The Post counted such schools with five or fewer students as home schools as well.

The statewide numbers are not always equivalent to the sum of all district totals in a state. Some states suppress district-level counts of home schoolers below a certain threshold. In Maine, the threshold is 5; in New Mexico, 6; in Mississippi, Ohio and Tennessee, 10; in Wisconsin before 2020-21, 5; and in Wisconsin from 2021-22 on, 20. The Post marked such suppressions as NA within its data. In addition, New Hampshire collects separate data on students who enter home schooling from schools run by the state department of education or from private schools; these additional students are reflected in state data but not district data.

The Post used a variety of methods to match each school district name to an NCES district id. However, this was not always possible. In Georgia, families self-report their school district on home-schooling forms; some report programs which are not school districts, and therefore have no corresponding NCES id. In California, families were only required to report county and school district beginning in 2020-21; in addition, district mergers and name changes mean that some districts could not be matched wi...