Relationships Between Families and Head Start Staff: Associations with Children’s Academic Outcomes Through Home Involvement and Approaches to Learning

ABSTRACT

Family-program partnerships in early care and education have the potential to influence young children’s development through multiple pathways. Guided by the Head Start (HS) Parent, Family, and Community Engagement (PFCE) framework, the current study uses data from the 2014 HS Family and Child Experiences Survey (FACES) to examine the direct, and indirect, through families’ home involvement and young children’s approaches to learning, associations between family-staff relationships and preschool children’s receptive vocabulary and early math skills. Research Findings: Drawing on data collected from families, teachers, and direct assessments with children, and controlling for multiple child, family, and teacher factors, we found that strong family-family services staff partnerships were directly, and indirectly, through higher family-home involvement, associated with children’s higher receptive vocabulary skills. In addition, we found that stronger family-teacher relationships were associated with children’s stronger approaches to learning, which in turn, were associated with children’s higher receptive vocabulary and early math skills. Practice or Policy: Our findings help substantiate and expand on the HS PFCE framework and help highlight the potential importance of family service staff in HS programs.

Introduction

Quality early care and education (ECE) is associated with children’s positive outcomes and life trajectories (Burchinal et al., 2016; Hong et al., 2019), and family engagement is one critical aspect of quality ECE (Wechsler et al., 2016). Family engagement in ECE programs can be thought of as a partnership between families and program staff, all of whom share responsibility for supporting children’s development and learning (Weiss et al., 2006; Westmoreland et al., 2009). Family-ECE program partnerships impact parenting and children’s developmental outcomes, which is particularly important for children from families with low incomes, as these children are most at risk for lower school readiness skills which can negatively impact their transition to kindergarten and later academic performance (Porter & Bromer, 2013; Zaslow et al., 2011). Aligned with this, Head Start, the key U.S. federal program designed to support school readiness in children from families with low incomes, includes not only a focus on children’s early learning, development, and health, but also on family well-being, recognizing that partnering with and supporting the whole family is critical for children’s success. Indeed, since its inception in 1965, Head Start has been committed to offering comprehensive programs that intentionally partner with families and communities to meet the needs of disadvantaged children with the goal of closing the opportunity gap for children in poverty (Zigler & Styfco, 2010).

Aligned with this commitment, and recognizing the importance of family well-being, Head Start has developed a model to orient this aspect of their work: the Head Start Parent, Family, and Community Engagement (HS PFCE) framework (U.S. Department of Health and Human Services, Administration for Children and Families, Office of Head Start, National Center on Parent, Family, and Community Engagement [NCPFCE], 2018). The HS PFCE framework takes a systems approach and is based upon nascent research identifying potential antecedents (e.g., from program leaders who ensure diverse family voices are included) and outcomes (e.g., children who are ready for school) of strong ECE program-family partnerships (Forry et al., 2012; NCPFCE, 2018). This framework also highlights that HS program-family partnerships are comprised of the relationships that families form with different individuals within the program, from the center directors and family services staff (FSS) who help families meet their needs and goals, to the classroom teachers who care for and educate enrolled children. Although the HS PFCE framework proposes that this network of relationships can enhance both family and child outcomes, there is limited research substantiating this model, nor do we understand the ways in which the different relationships families form with key individuals within the program may directly and indirectly enhance children’s outcomes. We seek to fill this void by examining data gathered from families, teachers, and children in the Head Start Family and Child Experiences Survey (FACES) and investigating how both family-FSS relationships and family-teacher relationships are directly associated with children’s academic outcomes (i.e., vocabulary and math skills), as well as indirectly through families’ home involvement (e.g., educational activities at home) and children’s approaches to learning.

The Importance of Partnerships with Families Within Classrooms and Across the Program

The partnerships between families and ECE programs may form at multiple levels, including the relationships families form with the teachers or providers directly caring for their children (i.e., parent- or family-teacher relationships) as well as with administrators or family service coordinators that work across classrooms to engage and support families (NCPFCE, 2018). Indeed, the HS PFCE framework defines family engagement as the process of working with families and children to build positive and goal-oriented relationships and emphasizes that family engagement practices are most effective when they are systemic, integrated, and comprehensive (NCPFCE, 2018). This means that those working across the program engaging and supporting families 1) know their role in fostering these positive and goal-oriented relationships with families and supporting positive family and child outcomes, 2) can coordinate with one another, and 3) are attuned to each child and family’s strengths, interests, and needs (NCPFCE, 2018).

Most of the research conducted thus far has examined family-ECE program partnerships at 1) the family-teacher level, i.e., examining the relationships between families and the direct care provider who watches their child on a daily basis (teacher or family child care provider), or 2) at the overall program level, analyzing the impact of family-centered practices (Forry et al., 2012) Although there are still many unanswered questions, this accumulated research on family-program partnerships appears to indicate that the relationships formed between the early learning context and children’s homelives can play an important role that influences the child’s daily experiences within the program (e.g., the child-teacher relationship; Lang et al., 2015), individual family member functioning (e.g., parental mental health or parental efficacy; Chazan-Cohen et al., 2007; Guterman & Hahm, 2001), the functioning of the family unit (e.g., parental involvement and the parent-child relationship, Trivette et al., 2010), as well as the child’s development (Forry et al., 2012; Lang et al., 2017, 2020).

Aligned with HS PFCE framework, Head Start programs are unique when compared to other child care settings, in that they typically have FSS who focus on collaborating with families to help them secure the resources necessary to achieve their goals (Head Start Act, 2016). These staff members’ primary role is to reduce other family stressors by helping families access community supports (e.g., healthcare, housing, food access, job training, financial education, English language courses, parenting courses, etc.) that can help them thrive (Ayoub et al., 2011; Rouse & Fantuzzo, 2009). When families form strong partnerships where they can openly express their needs, strengths, and challenges with their FSS, they may also be able to secure the environmental resources necessary to adequately invest in and feel successful within their roles as parents. This kind of support may impact children’s outcomes by ensuring children have their basic needs met to foster strong brain development, and/or by ensuring families have the time and the skill to thrive in their roles as parents or caregivers (Forry et al., 2012). To more deeply understand the potential broad impact of family-program partnerships, it is important to consider the partnerships formed at both the programmatic level, where supporting families’ access to community resources is the focus, and at the direct-care level within classrooms, as the bonds formed and support given within each level may have important implications for family functioning and child wellbeing.

Family Partnerships and Associations with Child Development

Recognizing the critical importance of how families engage with their children within their homes and communities, it is important to understand if and how family-program partnerships may support family involvement. Previous research has demonstrated that high quality family-program partnerships are associated with positive parenting practices (Reid et al., 2007; Trivette et al., 2010), including the quality of parent-child relationships (Forry et al., 2012; Lang et al., 2020) and potentially with family involvement (Lang et al., 2017), as well as reduced rates of maternal depression (Chazan-Cohen et al., 2007), and increased sense of parenting mastery (Guterman & Hahm, 2001). Strong family-program partnerships may increase parents’ self-efficacy and feelings of effectiveness, and parents’ feelings of joy in parenting (Forry et al., 2012), all of which may support parents and other key familial caregivers in spending more time with their children. In addition, when families can access the supports they need to thrive through strong family-program partnerships, they may also experience greater well-being and work-life balance, which again, may facilitate greater home-based involvement (Forry et al., 2012). This being said, we do not yet know which relationships within HS programs may play a critical role in fostering family’s home involvement or children’s developmental outcomes. As noted above, FSS may play a particularly important role in helping families access the resources necessary to flourish, so it is important to examine if these partnerships are also associated with children’s outcomes, both directly and indirectly, through family’s home involvement.

Direct Associations

When families and ECE teachers have strong relationships, this may facilitate greater knowledge among all parties, such that parents or guardians as well as ECE providers know how to provide individualized, developmentally appropriate engagement and support that fosters children’s early skills (Lang et al., 2020). Children may also observe quality interactions between family members and ECE professionals, and, through social learning (Bandura & McClelland, 1977), internalize various actions and behaviors (e.g., clear communication, vocabulary that facilitates strong relationships or joint planning) that positively support their development. Strong family engagement, including both families’ participation in the program and strong family-program partnerships, have been associated with higher vocabulary and math skills in young children (Bulotsky-Shearer et al., 2012; Elicker et al., 2013; Jeon et al., 2020), reductions in internalizing and externalizing behavior, and higher prosocial skills (Forry et al., 2012). Indeed, previous research has found evidence for the direct impact of strong parent-teacher relationships and children’s school readiness skills (Mendez, 2010; Powell et al., 2010). Interestingly, some research has found that teacher-reported attitudes toward families were more strongly predictive of children’s academic and social skills than families’ hands-on involvement in the program which highlights that the quality of the partnerships formed between families and ECE programs may be more critical than the actual amounts of time families are able spend in the ECE setting itself (Rimm-Kaufman et al., 2003). When we consider the role of FSS specifically, strong partnerships between families and FSS may also directly support children’s development in these areas by ensuring children’s basic needs are met. However, as outlined within Forry and colleagues’ (2012) conceptual model and included as potential outcomes of the HS PFCE framework (NCPFCE, 2018), the connection between family-program partnerships and children’s school readiness skills may work through various mediators, including positive changes in family functioning, or through strengthening children’s social emotional assets, such as their approaches to learning; and thus it may be through these mechanisms by which children’s pre- and early academic skills are enhanced.

Indirect Associations

Family/Parental Involvement

Previous research has found that family or parental involvement, particularly when focused on what key family caregivers are doing with their children to promote their learning (e.g., reading with their children, taking them to learning experiences in the community, teaching children about numbers) is associated with children’ early math skills (Guberman, 2004; Pan et al., 2006) and early vocabulary (Fantuzzo, McWayne, et al., 2004; Van Voorhis et al., 2013). In fact, Fantuzzo, McWayne, and colleagues (2004) found that family home-based involvement was “the strongest predictor of child outcomes” (p. 467), when compared to other kinds of family involvement, such as school-based or home-school conferencing, and Van Voorhis and colleagues’ (2013) literature review drew similar conclusions regarding the importance of family’s home involvement for children’s early literacy and math skills. Previous research has also shown that families are more likely to be involved with their children’s early learning when they have strong and positive relationships with their children’s teachers and programs (Forry et al., 2012; Jeon et al., 2018).

Approaches to Learning

Approaches to learning (ATL) “focuses on how children learn. It refers to the skills and behaviors that children use to engage in learning” (Early Childhood Learning and Knowledge Center, 2020). Children are considered to have strong ATL when they demonstrate persistence, motivation, emotional regulation, flexibility, and an openness to – but not a need for – adult support as they engage with materials or in activities (Chen & McNamee, 2011; Fantuzzo, Perry, & McDermott, 2004; Jeon et al., 2021). Previous research has demonstrated that young children’s ATL is significantly associated with their early skill development, including their early language and literacy skills (Fantuzzo, Perry, & McDermott, 2004; Li-Grining et al., 2010; Weiland et al., 2013) and early math skills (Bustamante et al., 2017; Duncan et al., 2007; Li-Grining et al., 2010). Approaches to learning may also be particularly important for children who are economically disenfranchised (Fantuzzo, Perry, & McDermott, 2004). Because ATL appears to be such a critical factor in early development, it is one of the five central domains of development highlighted within Head Start’s Early Learning Outcomes Framework (HS ELOF; Office of Head Start, 2015).

Recognizing the importance of young children’s ATL for concurrent and future development (Chen & McNamee, 2011), it is important we consider the factors that may foster children’s ATL. Strong partnerships between families and ECE programs may be one mechanism to enhance children’s ATL, as adult partners within these relationships may model cooperation, persistence, and flexibility as they collaborate to meet their shared goals (Hughes & Kwok, 2007; Lang et al., 2020). Previous research has demonstrated that strong family-teacher partnerships are associated with children’s ATL, which, in turn, are associated with children’s higher math gains (Jeon et al., 2021), or other key school readiness skills or early academic outcomes (Fantuzzo Perry, & McDermott, 2004; Hughes & Kwok, 2007). Although there is a growing body of research supporting the importance of family-teacher partnerships for fostering young children’s early approaches to learning (e.g., Elicker et al., 2013; Hughes & Kwok, 2007; Lang et al., 2020), documenting ATL as a key mediator between family-teacher partnerships and child outcomes across multiple early childhood samples helps to substantiate and refine theoretical models (Forry et al., 2012), and frameworks (HS PFCE; NCPFCE, 2018) on family engagement. Such information helps to validate the importance of investing in systems and structures to develop and maintain strong family-program partnerships and can also help identify the particular pathways through which family-program partnerships may have the most significant influence on young children’s development. Indeed, identifying the critical mediators between family-program partnerships and children’s outcomes supports the ECE field in knowing what to target in professional development efforts or other resources provided to ECE professionals.

The Current Study

The current study takes a critical next step in examining Head Start family-program partnerships and their association with important kindergarten readiness skills by examining these from multiple perspectives and levels within the Head Start programs. It incorporates the perspective of children’s families in assessing parents’/guardians’ relationships with their program’s FSS, thus providing an indicator of program-level family engagement. It also incorporates the perspective of ECE teachers, assessing their relationships with children’s families and thereby providing an indicator of classroom-level connections. Our research further examines both the direct and indirect associations between family-program partnerships and children’s early vocabulary and math skills. Examining these associations may, as described above, help illuminate potential pathways of influence that can refine both our conceptual models and frameworks of family-program partnerships and family engagement, and the kinds of practices or resources program staff use, invest in, or emphasize when they partner with families. In addition, our study uses a large national data set, the Head Start FACES Survey, which allowed us to control for a number of relevant child, family, and teacher characteristics in our analyses, and to utilize assessor-collected standardized child outcome measures, including children’s early receptive vocabulary and math skills. Understanding the direct and indirect associations between family-program partnerships and child outcomes is particularly important for children within Head Start programs, as children from families with low-incomes face multiple-risks to their educational wellbeing (Rouse & Fantuzzo, 2009). As Rouse and Fantuzzo argue (2009), to support “the educational well-being of young children with disproportionate risks requires intentional, systematic, and comprehensive interventions … ” (p. 1). Aligned with this work, and as noted above, the HS PFCE framework prioritizes the coordination across all roles within HS to help children meet their goals, with an emphasis on helping and empowering families to meet their needs so they, and their children, can thrive. Our study offers an opportunity to examine this framework in action.

Our specific research questions are:

RQ1:

Are parents’ perceptions of family-FSS relationships and teachers’ perceptions of family-teacher relationships directly associated with children’s academic outcomes (i.e., early receptive vocabulary and math skills specifically)?

RQ2:

Are parents’ perceptions of family-FSS relationships and teachers’ perceptions of family-teacher relationships associated with family home involvement (or learning activity) and, in turn, associated with children’s academic outcomes?

RQ3:

Are parents’ perceptions of family-FSS relationships and teachers’ perceptions of family-teacher relationships associated with children’s approaches to learning and, in turn, academic outcomes?

Our first research question is more exploratory in nature. Given the mixed findings within the literature noted previously, with some research finding direct connections between family-program partnerships and children’s outcomes (e.g., Mendez, 2010; Powell et al., 2010), and others finding an indirect association, through critical mediators like family’s home involvement (Fantuzzo, McWayne, et al., 2004), and children’s ATL (Hughes & Kwok, 2007; Jeon et al., 2021), we are unsure if family-FSS or family-teacher partnerships will be directly associated with children’s kindergarten readiness skills. For our second research question, we hypothesize that strong family-FSS relationships will be associated with greater family home involvement, which in turn will result in stronger receptive vocabulary and early math skills (Fantuzzo, McWayne, et al., 2004; Forry et al., 2012). For our final research question, we hypothesize that strong family-teacher relationships will be associated with children’s higher ATL, which in turn will result in strong receptive vocabulary and early math skills (Hughes & Kwok, 2007; Jeon et al., 2021).

Methods

Data Source

This study used data from the nationally representative Head Start Family and Child Experiences Survey 2014 (FACES 2014–2017; U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research and Evaluation, 2014–2017) which includes 2,462 HS children and their families as well as 667 teachers/classrooms and 176 programs from fall 2014 to spring 2015. FACES 2014 provided comprehensive data including information about HS children, families, teachers, classrooms, FSS, and programs. Compared to the earlier series of FACES data which provided more than 2 years of longitudinal data, FACES 2014 aimed to provide data in more effective ways by conducting core and plus studies simultaneously over an academic year. This approach generated rich information for multiple layers of the programs which are associated with the holistic development of HS children. Specifically, FACES 2014 provides information on multiple aspects of relationships among HS families, teachers, and FSS by utilizing a newly developed measure (i.e., the Family and Provider/Teacher Relationship Questionnaire). The FACES 2014 user guide provides further information about survey design and sampling procedures (Klein et al., 2018).

The present study used three datasets: 1) the Classroom + Child Outcomes Core study including child outcomes in spring 2) Classroom Core study including teachers’ reported family teacher relationships, and 3) Family Engagement Plus study including parents’ reported family-FSS relationships. Since the Family Engagement Plus study included a sample of parents taking one-hour interviews and additional surveys related to family engagement, this study included only children and their families who participated in the Family Engagement Plus study so that we could examine multiple aspects of family engagement in HS programs.

Study Sample

In the present study, we selected families who participated in the Family Engagement Plus study and merged teacher and child data from the two core studies. The Family Engagement Plus study provides information about each family’s relationship with their FSS. We merged this data with the Classroom + Child Outcomes Core study using the child ID variable, which exists in both datasets. After we created a merged dataset including child outcomes and family-FSS relationships, we merged this data with the Classroom Core study by using teacher ID to include teachers’ perceptions on relationships with families at the classroom level. After all data merging, the resulting sample consists of 309 children from 195 classrooms in 117 centers and representing 60 HS programs.

To account for variations in the probability selection of the sub-sample that we used for this study, the study employed a sampling weight (PERA12 WT) which is one from the Family Engagement Plus study. This weight can be used for analysis of “parent survey data in fall 2014 or spring 2015 in combination with Family Engagement parent interview data and either teacher child rating or child assessment data in fall 2014 and spring 2015” (Klein et al., 2018, p. 189) to maintain the nationally representativeness of the sample of HS children and families.

Measures

Receptive Vocabulary Skills

Children’s receptive vocabulary skills were assessed using the Peabody Picture Vocabulary Test-4 (Dunn et al., 2006). All children received the PPVT regardless of their dominant language. During the 2014 FACES study, the PPVT-4 was administered via computer rather than the standard paper assessment easel. Children responded verbally or by pointing to images displayed on a computer screen and FACES assessors recorded their responses using a laptop computer. Administering the PPVT in this way reduced the time required for assessors to manage paper assessment materials and improved the assessment flow. These scores had an internal consistency of .97 in Spring 2015 (Klein et al., 2018).

Math Skills

In Fall 2014 and Spring 2015, children completed both the Applied Problems subtest of the Woodcock-Johnson Test of Achievement, third edition (W-J III; Woodcock et al., 2001) and Early Childhood Longitudinal Study birth cohort math assessment (ECLS-B math; Snow et al., 2007). The W-J III Applied Problems measures children’s skills related to mathematical problem-solving using numbers and operations while ECLS-B math assessment adds items that assess children’s understanding of numbers, patterns, counting, shapes, relative quantities, and size. To represent children’s comprehensive math knowledge, we used the FACES-created math scale scores which were computed using item response theory to estimate the full number of math items children would have answered correctly if they had completed all questions on both the W-J Applied Problems and ECLS-B assessments. These scale scores had an internal consistency of .92 in Spring 2015 (Klein et al., 2018).

Children’s Approach to Learning

Teachers rated the sampled children’s ATL using the ECLS-K Approaches to Learning Scale (U.S. Department of Education, National Center for Education Statistics, 2002). On this scale, teachers responded to items asking about how often sampled children displayed behaviors associated with motivation, attention, organization, persistence, and independence such as “How often in the past month did he or she show eagerness to learn new things?” Teachers rated children’s performance across six items on a scale of 0 (never) to 3 (very often). Internal consistency of the ATL scale was .93 in Spring 2015 (Klein et al., 2018). This study uses a mean of the six individual ATL items to represent children’s ATL.

Family and Provider/Teacher Relationship Quality (FPTRQ) Measures

In 2010, the Administration for Children and Families’ Office of Head Start and Office of Planning, Research and Evaluation and the Office sponsored a project to develop strong measures of family program relationships in ECE settings serving children birth to five years of age. This multi-year project yielded multiple assessments of different relationships and from different perspectives, with the goal of providing a set of tools to better understand the quality and impact of family-program partnerships in ECE. Details regarding the extensive development and validation processes for these measures are outlined in Kim et al. (2014). In this study we utilize the FPTRQ FSS parent measure, and the provider/teacher (PT) measure.

Family-FSS Relationships

Within the FPTRQ FSS measure for parents, families are asked to reflect on how comfortable they feel sharing various information (e.g., the number of adult relatives living in the household, their family’s culture and values) with their FSS, how often they have communicated about various aspects related to their child (e.g., child’s goals and progress) and the family’s current context (e.g., problems the parent or caregiver may have with work or school), and how respectful or responsive their FSS is (Kim et al., 2014).

We used a latent variable with three indicators drawn from the spring 2015 parent responses on the FPTRQ short form (Kim et al., 2014). These indicators included parents’ mean scores on each of the three FPTRQ subscales (i.e., knowledge, attitudes, and practice), with scores ranging from 1 (never, none, or strongly disagree) to 4 (very often, all, or strongly agree). Within this measure, families share their perspective of the FSS’s family-specific knowledge, their level of collaboration, responsiveness, communication, and family-focused concern, as well as the staff members’ commitment, understanding of context and respect. This scale was administered to families in either English or Spanish as a part of the Family Engagement Plus wave of FACES data collection between April and July 2015. The full scale had acceptable internal consistency (.79) within the FACES sample.

Family-Teacher Relationships

Family-teacher relations were represented as a latent construct with three indicators drawn from the spring 2015 teacher responses on the FPTRQ short form (Kim et al., 2014). These indicators included teachers’ mean scores on each of the three FPTRQ subscales (i.e., knowledge, attitudes, and practice), with scores ranging from 1 (never, none, or strongly disagree) to 4 (very often, all, or strongly agree). Of the three subscales, knowledge attempts to capture what teachers know about the specific families they serve and the families’ values and practices outside of school; the attitudes subscale asks teachers to rate items related to their beliefs about families and children more generally; and the practice subscale assesses teachers’ actions related to collaboration, responsiveness, and communication with families.

Family Home Involvement

We also used a latent variable (i.e., home involvement), which indicated the frequency of educational activities at home for children. This latent variable included two indicators: 1) the numbers of activities that parents and children did together in the past week (e.g., played with blocks, counting games, discussed or played games with letters, words, and numbers; 14 items yes or no; Cronbach alpha =.67) and 2) frequency of reading with the child in past week (1 = not at all, 2 = once or twice, 3 = 3+times, but not daily, 4 = everyday). We used data collected from parent interviews in spring 2015.

Covariates

We included several child (e.g., gender, English proficiency), family (e.g., poverty, maternal education), and teacher characteristics (e.g., education, experience) as covariates within our analyses. The specific characteristics we included are shown in Table 1 and were selected for inclusion as these characteristics have been shown in past research to be associated with child achievement scores.

Table 1. Descriptive statistics.

	Total (n= 309)
	M (Range)	SD	Factor loading
Child outcomes (level-1)
Receptive vocabulary skills	89.53 (41–134)	16.63
Math skills	20.28 (3.31–40.58)	8.08
Approaches to learning	1.97 (0–3)	.75
Family engagement (level-l)
Parent-reported family-FSS relationships
Knowledge	56.39 (24–68)	9.23	.66
Practice	97.73 (31–124)	17.69	.79
Attitude	55.67 (18–64)	7.63	.80
Family home involvement
Educational activities at home	12.01 (5–14)	1.96	.54
Reading books	3.20 (1–4)	3.20	.59
Child characteristics (level-1)
Gender (1=male)	48%
English proficiency (1= yes)	81%
Returning status (1=yes)	66%
Child Race
European American	27%
African American	23%
Hispanic/Latino	41%
Others	9%
Family characteristics (level-1)
Ratio of income to poverty	2.35 (1–6)	1.42
Mother education	2.27 (1.4)	.91
Family engagement (level-2)
Teacher reported family-teacher relationships
Knowledge	12.32 (5–20)	2.92	.43
Practice	30.66 (18–36)	3.71	.89
Attitude	29.78 (18–36)	2.87	.57
Teacher characteristics (level-2)
Teacher # of years teaching in HS	14.13 (0–30)	8.93
Teacher education level	6 (4–13)	1.16

Children’s English Proficiency

We included children’s language screening results that determined the language of assessment as our indicator of children’s English language proficiency. Children’s parents were asked what language their children use most often at home. Children were also administered two subtests from the Preschool Language Assessment Survey (“Simon Says” and “Art Show;” Klein et al., 2018). Those children whose families reported that they primarily spoke English at home and those who were reported to speak a language other than English at home but who missed fewer than 12 items total between the “Simon Says” and “Art Show” tasks were considered to be proficient in English for the purposes of this study as they were routed to the English-language assessment track (Klein et al., 2018).

Data Analyses

To address the three research questions, we examined direct and indirect models using Mplus 8. Since we included classroom-level teachers’ perceptions of family-teacher relationships, all children in the same classroom had the same scores for family-teacher relationships. To adjust the structure of multilevel data and nesting issues, we used teacher ID in spring 2014 as a cluster ID in research models. We also examined a three-level nested model (child→classroom→center) as FSS work center-wide but the results of the three-level model were not significantly different from the two-level model, so we retained the more parsimonious models as the final models within this study.

For research question 1, we examined the direct associations between parents’ perceptions of family-FSS relationships or teachers’ perceptions of family-teacher relationships and children’s academic outcomes in HS programs. For research question 2, we examined whether parents’ perceptions of family-FSS relationships and teachers’ perceptions of family-teacher relationships are associated with family home involvement (or learning activity) which, in turn, is associated with children’s academic outcomes. For research question 3, we added mediation paths from teachers’ perceptions of family-teacher relationships and parents’ perceptions of family-FSS relationships to children’s academic skills through children’s ATL skills. All models included child/family characteristics, and teacher characteristics as control variables for child outcomes. In the Mplus two-level mediation model, Level-1 variables were included in WITHIN to examine Level-1 mediational paths with two endogenous variables (i.e., home involvement and ATL). Level-2 variables were included in BETWEEN to examine Level-2 mediational paths from a predictor (teacher-reported family-teacher relationships) and teacher characteristics as controls to the intercepts of each Level-1 outcome via home involvement and ATL. A latent construct about home involvement was included in BETWEEN to examine the paths from teacher perceptions of family-teacher relationship to child outcomes via home involvement. We used the model constraint command to test the significance of the indirect paths. More specifically, we add paths from child/family/teacher characteristics to child outcomes for all three of the models. To handle the item-level missing in Level-1 and Level-2 variables, we conducted multilevel multiple imputation (MI) and generated 20 imputed data sets (Rubin, 1987). All the study variables including covariates and the sampling weight as an auxiliary variable were included in the multilevel MI model (Enders et al., 2016).

Results

Direct Models

The direct model for RQ1 was conducted and the model showed acceptable model fit, CFI =.94, RMSEA =.06. Results (Figure 1) showed that parents’ perceptions of family-FSS relationships were directly associated with children’s language skills but not math skills. Teachers’ perceptions of family-teacher relationships were not directly associated with children’s academic skills. We found significant associations between child and family characteristics and children’s academic outcomes. Children’s English proficiency was associated with both language (β = .40, p < .001) and math skills (β = .32, p < .001) and children who were enrolled the previous year in HS showed higher scores in math than children who were newly enrolled in HS in 2014 (β = −.27, p < .001). Mothers’ education level was associated with children’s language skills (β = .12, p < .05). Teachers’ characteristics were not associated with children’s academic skills.

Figure 1. Direct model.

A sampling weight (PERA12WT) was applied; **p < .01; Solid lines indicate significant parameter estimates; dashed lines indicate non-significant parameter estimates; Standardized coefficients were provided. Control variables are included in this model; Model fit: RMSEA =0.06, CFI =.94.

Indirect Models

To address RQ2, we added mediation paths (Model 1-1-1) from parents’ perceptions of family-FSS relationships and teachers’ perceptions of family-teacher relationships to academic outcomes via home involvement (CFI = .95, RMSEA =.03). Results (Figure 2) showed that family-FSS relationships were associated with home involvement which in turn was associated with children’s receptive vocabulary skills and marginally associated with math skills. However, teachers’ perceptions of family-teacher relationships were not associated with home involvement. The indirect path from family-FSS relationship to receptive vocabulary skill via home involvement was significant (b = .14, 95% CI [.05, .27]), while the indirect path from family-FSS relationships to math skills via home involvement was not significant (b = .07, 95% CI [−.02, .16]).

Figure 2. Mediation model.

A sampling weight (PERA12WT) was applied; +p < .10, *p < .05; Solid lines indicate significant parameter estimates; dashed lines indicate non-significant parameter estimates; Standardized coefficients were provided. Control variables are included in this model; Model fit: RMSEA =0.03, CFI =.95.

To address RQ3, we added mediation paths (Model 2-1-1) from parents’ perceptions of family-FSS relationships and teachers’ perceptions of family-teacher relationships to children’s academic outcomes via children’s ATL (CFI = .94, RMSEA =.04). Results (Figure 3) showed that parents’ perceptions of family-FSS relationships were not associated with children’s ATL while teachers’ perceptions of family-teacher relationships were associated with children’s ATL, which in turn were associated with both receptive vocabulary and math skills. The indirect path from family-teacher relationship to receptive vocabulary skill via ATL was significant (b = .73, 95% CI [.04, 1.42]) and the indirect path from family-teacher relationship to math skills via ATL was also significant (b = .59, 95% CI [.15, 1.02]).

Figure 3. Two-level mediation model.

A sampling weight (PERA12WT) was applied; +p < .10, *p < .05, **p < .01, ***p < .001; Solid lines indicate significant parameter estimates; dashed lines indicate non-significant parameter estimates; Standardized coefficients were provided. Control variables are included in this model; Model fit: RMSEA =0.04, CFI =.94.

Discussion

The results demonstrated that Head Start family-program partnerships are both directly and indirectly, through family home involvement and children’s approaches to learning, associated with important kindergarten readiness skills. In addition, the results also highlight the value of examining different kinds of family-staff relationships within Head Start, and the potential importance of both families’ relationships with their child’s classroom teacher (i.e., classroom level), and their relationships with FSS (i.e., program level) for fostering positive child outcomes. Our results not only help to validate previous conceptual models (e.g., Forry et al., 2012) and frameworks on family engagement (e.g., NCPFCE, 2018) that emphasize the impact of family-program partnerships on family functioning, children’s wellbeing and children’s early academic skills, they also enhance this work by 1) documenting key mediators between family-program relationships and child outcomes, and 2) highlighting the specific importance of family-FSS relationships. To our knowledge, our research was the first of its kind to specifically investigate the potential impact of family-FSS partnerships on children’s outcomes, and to examine these associations while also accounting for family-teacher relationships at the classroom level and several relevant child, family, and teacher covariates.

Direct Associations with Children’s Early Academic Outcomes

Within our model, after accounting for covariates, only parents’ assessment of family-FSS relationships were directly associated with children’s receptive vocabulary skills. As noted within the introduction, the previous ECE literature examining direct associations between family-program partnerships and child outcomes has demonstrated mixed findings, with some research finding direct associations (Mendez, 2010; Powell et al., 2010), and others finding indirect associations, with an emphasis on family’s home involvement as the key driver of children’s positive outcomes (Fantuzzo, McWayne, et al., 2004; Van Voorhis et al., 2013). However, previous research primarily focused on parent-teacher relationships at the classroom level, and thus the meaning, and potential spheres of influence that program-level, and specifically family-FFS partnerships, may play in supporting children’s development was less known. Our research provides evidence for the importance and potential direct impact of these partnerships on children’s early developmental outcomes.

Recognizing the role of FSS within HS programs, strong family-FSS relationships may mean that families can more freely share their needs and concerns regarding the broader resources (e.g., housing, nutritional access, health care, mental wellbeing support) they need to thrive. Hence, it may be that these relationships help families secure the broader resources that support children’s healthy development, specifically their early vocabulary skills. This aligns with previous theory (Becker, 1991; Mayer, 1997) and research (Ryan et al., 2005) that young children in poverty are more likely to have lower quality home environments, neighborhoods, as well as health care and nutritional access, all of which may negatively impact their early development, and that interventions that provide high-quality ECE and resource support for families can mitigate these conditions and support positive short- and long-term outcomes for children, including their verbal functioning (Ryan et al., 2005; Yeung et al., 2002). Although our findings lend some support for the important role family-FSS partnerships may play in fostering positive outcomes for children enrolled in HS, future research should examine the mechanisms through which family-FSS partnerships may be fostering children’s early receptive vocabulary skills. For example, it could be that these partnerships help families have greater access to books or other print and language rich materials in the home that facilitate children’s early vocabulary skills. In addition, it could be that these partnerships help families gain access to adequate health care to ward off the experience or effects of ear infections or related issues that may impact children’s auditory experiences. Understanding the specific ways that family-FSS partnerships support families would help in the training, guidance, and support given to family services staff.

Indirect Associations with Children’s Early Academic Outcomes

Family Home Involvement

Our findings align with previous research demonstrating the importance of family home involvement for supporting young children’s early development, particularly for children in families with low income (Fantuzzo, McWayne, et al., 2004), and documenting the importance of family-program partnerships for supporting parents’ involvement (Forry et al., 2012; Lang et al., 2017). Our work helps to validate both Forry et al.’s (2012) conceptual model and the HS PFCE framework (NCPFCE, 2018), by providing additional evidence that higher family involvement and children’s greater kindergarten readiness skills are both potential outcomes of strong HS family-program partnerships. However, our findings also extend this work by documenting that family home involvement may be a key mediator between family-program partnerships and child’s early vocabulary development and math skills (though the significance of the association with math skills must be interpreted with caution, p < .10), and that it was specifically family-FSS partnerships that were associated with greater family home involvement, and in turn, children’s early skills.

Family-FSS partnerships may be critical levers to increasing family home involvement because they may help facilitate families’ access to resources or supports that simultaneously increases their available time to interact and reduces stress that can negatively impact interaction (Forry et al., 2012). Hence, with more available time and less stress, families are afforded more time to spend with their children. In addition, it may be that family services staff help parents and guardians access additional resources such as parenting classes, community play groups, or trips to local community events or experiences that help facilitate their involvement. Although more research is necessary, our finding may indicate that it is the family’s access to resources and support that facilitates their involvement. This substantiates the value of FSS, particularly in HS programs, and provides additional support for HS PFCE framework (NCPFCE, 2018) that emphasizes the role of FSS and the importance of coordination across HS to optimally engage families. Future research should examine how family-FSS relationships specifically enhance family’s home involvement, to identify the strategies and resources that best foster positive partnerships and support families.

Children’s Approaches to Learning (ATL)

While the partnerships between family-FSS demonstrated both direct and indirect (through family home involvement) associations with children’s outcomes, the associations between family-teacher relationships and children’s early vocabulary and math skills were entirely mediated by children’s ATL. This finding aligns with previous research (e.g., Fantuzzo, Perry, & McDermott, 2004; Hughes & Kwok, 2007; Jeon et al., 2021), and further substantiates the important role that family-teacher partnerships may play in fostering the skills and orientations that make children strong learners. As discussed in the introduction, it may be that strong family-teacher partnerships model important skills for children, such as attention, cooperation, and persistence in meeting shared goals (Hughes & Kwok, 2007; Lang et al., 2020), and children internalize these observations (Bandura & McClelland, 1977). However, it may also be that strong-family-teacher partnerships increase parental efficacy and/or the quality of child-guardian interactions by reinforcing or validating the strengths parents or other key family caregivers display in their parenting roles (Forry et al., 2012), and it is these critical factors that ultimately drive children’s greater ATL. More research is needed to identify the discrete mechanisms that underlie these associations, but our findings add to the growing research about the importance of family-teacher partnerships for children’s early social-emotional skills, including their ATL specifically.

In addition, although this particular finding provides additional evidence in support of Forry et al.’s (2012) conceptual model of family-program partnerships and the HS PFCE framework (NCPFCE, 2018) by emphasizing the role that family-teacher partnerships may play in supporting children’s early academic outcomes after accounting for many relevant child, family and teacher factors, it also substantiates the value of investing in children’s ATL. Our findings echo that of previous researchers (Fantuzzo, Perry, & McDermott, 2004), that young children’s ATL appear to be critical in fostering early academic skills. When children are in contexts that enhance their internal motivation as well as their ability to persist, to regulate their emotions, and to be flexible in their thinking or approaches, these orientations and abilities may make it easier for them to attend to and remember new vocabulary they are exposed to in their interactions, and to interact with materials and others in ways that foster their early understanding of patterns and numeracy. The strong association between young children’s ATL and their receptive vocabulary and math skills in our research also further validates HS’s ELOF (ECLKC, n.d.) where ATL is a critical developmental domain of focus in early childhood that should be strategically fostered.

Limitations

Although our study has several strengths as outlined in the introduction, our results must be interpreted within the constraints of our design and particular sample. First, all data was collected at one time point and examined associations within existing conditions (i.e., this was not an experimental design). Though our path models were grounded in theory and aligned with relevant conceptual frameworks (Forry et al., 2012; HS PFCE framework by NCPFCE, 2018), findings are all correlational in nature and we are unable to ascertain causality. Second, although our study included data collected from different perspectives (i.e., parents/guardians, teachers, and children through direct assessment by researchers), the measures for family-teacher relationships and children’s ATL were both collected via survey and by teachers, and hence the findings may be influenced by shared methods variance. Third, the measure for family-teacher partnerships was assessed at the classroom-level, i.e., measuring teachers’ perceptions of their overall, or collective, relationships with families, and thus misses potential nuance across family-teacher dyads. Though our design did capture this dyadic nuance at the program level, by including families’ perceptions of their relationships with the family services staff, future research should examine variation among partnerships at the classroom level as well. Finally, our sample was families, children, and staff within HS center-based programs and thus our findings may not be generalizable to other ECE settings, especially programs that have different structures and supports (e.g., those that lack a funded and trained FSS role).

Implications for Policy, Practice, and Research

Although our findings are specific to HS settings, they do lend some additional evidence in support of HS’s structure and commitment to supporting the family unit. In particular, this research provides some emerging validation for the importance of the FSS role. As more states consider universal ECE or Pre-K models as a critical mechanism to support children’s positive outcomes (e.g., Falgout & McSorley, 2020) and as K-12 school districts and community-based child care centers increasingly become purveyors of universal Pre-K (Morris & Smith, 2021), our research points out that it would be important for these universal ECE models to strategically consider the structures and policies in place to support family engagement and family-program partnerships. Considering our findings document both direct and indirect associations between family-FSS relationships and child outcomes, these models should consider how to integrate FSS within these other settings in ways that facilitate coordination across roles. Although the role of FSS may be less common, or at least less well-documented, outside of HS, our findings may have implications for other childcare settings that serve families who may face greater challenges or vulnerabilities (e.g., family advocacy staff in military-affiliated programs, or family service coordinators who work in centers that serve mainly families receiving publicly funded care). Recognizing the growing body of evidence that it is the holistic support of young children and their families that helps set children up for positive life trajectories (Heckman, 2016; Ryan et al., 2005), FSS can help ensure that families have appropriate access to all the supports necessary to help them, and their children, thrive. Thus, when states, municipalities or other public entities are considering universal ECE models, they should consider that the cost of childcare is not just adequate pay for the teachers within classrooms, but includes the other critical roles, such as FSS, that help facilitate a comprehensive system of care and support.

Our research also helps to further substantiate the importance of professional development (PD) that focuses on family-program partnerships in ECE. Our findings help to identify areas that may be particularly important to emphasize within PD for FSS and ECE teachers. Recognizing the potential importance of positive family-FSS partnerships for increasing family involvement, it may be helpful for FSS to narrow in on the ways that they can help support families’ home involvement, for example, by helping to identify and address barriers that hinder parents’ or caregivers’ available time to be with their children, that help families know about all the rich ways they can be involved with their children, and by validating the great work families do. For ECE teachers, it may be particularly beneficial to focus on the ways their positive partnerships with families can model strong approaches to learning, demonstrating attention, emotional regulation, persistence, and collaboration toward shared goals.

Lastly, in terms of research, although our findings substantiate and extend the important conceptual models (Forry et al., 2012) and frameworks (NCPFCE, 2018) on family-program partnerships, they evoke additional questions. As noted above, more research is needed to understand how family-FSS partnerships may facilitate family involvement and child outcomes, and if the associations we found hold in other childcare contexts (e.g., military childcare programs, other settings serving families with low income). Future research should examine if strong family-FSS partnerships are associated with families’ greater access to other critical social supports (e.g., healthcare, nutritional and housing subsidies, parenting classes) and/or parents’ and caregivers’ greater self-efficacy in their parenting role. In addition, future research should examine the specific actions and behaviors within family-teacher partnerships that may facilitate children’s ATL. We also recognize that both these critical partnerships with families take place within a program, and for HS, within a more structured system of ECE, and thus, future research should examine what within the program environment, such as center-wide leadership practices, or specific structural supports outlined within HS policy, helps to foster positive family-program partnerships at both the program- and classroom-level.

Conclusion

Head Start promises to help close the opportunity gaps that exist for children in families with low income (Ludwig & Phillips, 2007). Although continued investment past ECE is likely necessary to help sustain the early positive skills that HS fosters (Ansari, 2018), our research provides some additional evidence that attention to family-program partnerships within HS programs may be one key mechanism for both directly and indirectly supporting children’s kindergarten readiness skills. Our results suggest that families’ relationships with HS staff may impact children’s learning experiences in the home and their learning behaviors which, in turn, may affect children’s academic outcomes. These results illuminate the potential far-reaching impacts, both in the classroom and at home, of families’ relationships with HS staff and have implications for determining where HS programs should focus PD efforts and resources. Further research should investigate whether these changes in children’s approaches to learning and families’ home learning practices persist past the preschool period.

Author statement

The second author conceived of this study and conducted the data analyses; the third author assisted with the literature review, wrote portions of the methods, and provided multiple rounds of editing. The first author added to and refined the research questions and wrote the introduction and discussion sections. All authors have read and approved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).