Head Start: The Family and Child Experiences Survey (FACES)

Head Start is a national program that helps young children from families with low incomes prepare to succeed in school. It does this by working to promote children’s early learning and health and their families’ well-being. Head Start connects families with medical, dental, and mental health services to ensure that children are receiving the services they need to support their development. Head Start also involves parents in their children’s learning and development and helps parents make progress on their own goals, such as housing stability, continuing education, and financial security (Administration for Children and Families 2020).  Head Start provides grants to local public and private nonprofit and for-profit agencies. The agencies in turn deliver comprehensive services to children and families with low incomes.

The Head Start Family and Child Experiences Survey (FACES) was launched in 1997 as a periodic, longitudinal, descriptive study of Head Start’s performance. There are 12 regions for federal management of Head Start. FACES gathers data on Head Start programs, staff, children, and families from Regions 1 through 10, which are the 10 geographically based Head Start regions nationwide. As of 2014, a study of Region XI (American Indian and Alaska Native; AIAN) was added, or the AIAN Family and Child Experiences Survey (AIAN FACES) (see below). FACES was designed to be a reliable source of data for describing the skills of Head Start children; the experiences of the children and families served by Head Start; the quality of Head Start classrooms; and the qualifications, credentials, and opinions of Head Start staff. The U.S. Department of Health and Human Services, Administration for Children and Families, funded the FACES studies. Mathematica also conducted the 2021—2022 Study of Family and Staff Well-Being in Head Start Family and Child Experiences Survey (FACES) Programs, which built on the FACES 2019 program sample. More information can be found here.

FACES 2006 and 2009

In 2006 and 2009, Mathematica conducted five-year studies, each comprising about 3,400 3- and 4-year-old children newly enrolled in 60 Head Start programs around the country. Our studies followed Head Start children and families from entrance into the Head Start program through one or two years of program participation, with follow-up in the spring of kindergarten. Researchers gathered comprehensive data through direct child assessments in multiple domains; observations of Head Start classrooms; and interviews with Head Start parents, teachers, and administrators.

For both studies, Mathematica analyzed the data and prepared a series of products. We also prepared data files and data file documentation to make the data available to the community of researchers for secondary analyses. Subcontractors for FACES 2006 and 2009 were Educational Testing Service and Juárez and Associates.

FACES 2014-2018

Mathematica redesigned FACES to include a Core Plus design to provide key data more rapidly and with greater frequency (Core studies) and to help researchers examine more complex issues and topics in greater detail and efficiency (Plus studies). Using a Core Plus design provides more flexibility in incorporating special topic or methodological studies. The two FACES 2014 Core studies are the Classroom + Child Outcomes Core and the Classroom Core. The Classroom + Child Outcomes Core took place in the 2014–2015 Head Start year. In fall 2014 and spring 2015, we assessed the school readiness skills of more than 2,000 3- and 4-year-old Head Start children in 60 programs, conducted surveys with their parents, and asked their Head Start teachers to describe their social-emotional skills and diagnosed disabilities. In spring 2015 we visited 176 programs (including the original 60) to conduct surveys with program directors, center directors, and teachers and observe 667 classrooms. Therefore, the Classroom + Child Outcomes Core collected child-level data along with program and classroom/teacher data from 60 programs. The program and classroom/teacher data collected across all 176 programs conducted in spring 2015 represent the first round of the Classroom Core. In spring 2017, FACES conducted a second round of the Classroom Core with an updated sample of programs to ensure that it was nationally representative of all Head Start programs at that time. In spring 2017, 178 programs participated in the study.

FACES 2014-2018 included five Plus studies: (1) the Family Engagement Plus study, (2) the Five Essentials Measurement System for Early Education Educator Survey Pilot study (5E-Early Ed Educator Survey Pilot Study), (3) the first-ever study of the children and families served by Region XI AI/AN Head Start programs, conducted in 21 programs with over 1,000 children during the 2015-2016 program year AI/AN FACES 2015, (4) a Plus topical module on programs’ perspectives on the new Head Start Program Performance Standards (HSPPS Plus module), and (5) a Plus topical module on program functioning using the Early Education Essential Organizational Supports measurement system (Early Ed Essentials Plus module).

Mathematica prepared a series of analytic reports documenting study findings. We also prepared data files and data file documentation so that these data are available to the community of researchers for secondary analyses.

Educational Testing Service and Juárez and Associates served as subcontractors for Mathematica.

FACES 2019

FACES 2019 followed a similar structure as FACES 2014-2018; it had a Core Plus study design that provided data on a set of key indicators more rapidly and with greater frequency (Core studies) and served as a vehicle for studying more complex issues and topics in greater detail and with increased efficiency (Plus studies). It also aligned the timing of the study of Region XI American Indian and Alaska Native Head Start (AIAN FACES 2019) with the Classroom + Child Outcomes Core study in Regions 1-10.

Mathematica and its partners, Juárez and Associates, Educational Testing Service, and consultants Margaret Burchinal and Martha Zaslow, developed the instruments and data collection procedures for two Core studies. We selected and recruited a nationally representative sample of 180 Head Start programs and 360 Head Start centers in Regions 1-10. We assessed the school readiness skills of 2,400 children and survey their parents and Head Start teachers in fall 2019 and spring 2020 (Classroom + Child Outcomes Core from a subsample of 60 of the 180 programs). We observed 720 Head Start classrooms and survey Head Start staff in spring 2020 (Classroom Core in all 180 programs).

Mathematica analyzed the data from each round of the study and prepared a series of products. We also prepared data files and data file documentation so that these data are available to the community of researchers for secondary analyses.

Survey Methodology Highlights

Technological innovations in data collection increase the accuracy and timeliness of data management and analysis. These included web-based computer-assisted direct assessments of the study children, as well as web instruments for teachers to use when reporting on children’s social-emotional, cognitive, and physical outcomes. We also used a dual-screen administration for child assessments, a web option for parent surveys in addition to the computer-assisted telephone survey, and web options for Head Start staff surveys.