Predictors of Parent-Teacher Relationship Quality in Head Start Classrooms

The ability for parents and teachers to work together serves as a foundation for a collaborative and effective parent-teacher relationship (Clarke et al., 2009). The quality of this relationship contributes to children’s school readiness and social competence (Halgunseth et al., 2009). However, limited research has examined the role of teacher’s professional characteristics, or the role of the classroom racial/ethnic composition on parents' perceptions of family-centered practices. The current study aims to identify teacher and classroom characteristics that may be related to components of effective parent-teacher relationships by exploring predictors of parents’ comfort in disclosing family information and parents’ perceptions of their classroom teachers’ responsiveness. This study utilizes data from the Head Start Family and Child Experiences Survey (FACES), 2014-2015 cohort. The sample includes survey data from parents (n = 747) and teachers (n = 227). Results from this study suggest a congruence in parents’ and teachers’ perception of family-centered practices. These perceptions remained consistent across a variety of classroom profiles.


INTRODUCTION
Family engagement is a key component of early childhood education and has long been a focus in the performance standards of our nation's largest publicly funded preschool program, Head Start (U.S.Department of Health and Human Services, 2016).Family engagement is a multidimensional construct that encompasses a variety of behaviors in which parents and the family unit play an active role in facilitating and enhancing their children's' learning experiences (Epstein, 1995).There's a robust literature base around the benefits of family engagement on children's academic, as well as, social and emotional outcomes (Barnard, 2004;Rimm-Kaufman et al., 2003;Sheridan et al., 2011).

RESEARCH ARTICLE
Although family engagement is frequently defined within the scope of school-based behaviors (e.g.parents volunteering in the classroom), conceptual models have moved towards incorporated home-based activities (e.g.parents working on learning activities at home) highlighting parents' active role in facilitating the home-school connection (Fantuzzo et al., 2000).At the same time, there has been an increasing recognition of the role that teachers play in cultivating a strong and positive relationship with families that can further promote parents' engagement and comfort in interacting with the school system, therefore enhancing children's learning (Herman & Reinke, 2017;Nzinga-Johnson et al., 2009).These family-centered practices include family-specific knowledge, attitudes towards families, family-centered practices, and welcoming environmental features (Kim et al., 2015).Prior research has shown associations between teachers' use of these practices and children's positive outcomes such as academic school readiness, and lower observed problem behavior (Bromer et al., 2011;Forry et al., 2012;Mendez, 2010;Rimm-Kaufman et al., 2003).
Key elements of establishing effective family-provider relationships (or, parent-teacher relationships hereafter) include a focus on family-centered knowledge, attitudes, and practices as well as the environmental features in the classroom that are welcoming and inviting for families.These elements have emerged as important factors that serve the foundation for building effective partnerships (Bromer et al., 2011;Forry et al., 2012).Drawing from the early intervention literature, the provision of family-centered care and practices are associated with positive outcomes for both children and their families (Hughes-Scholes & Gavidia-Payne, 2019).This research has, for the most part, paid little attention to the backgrounds and characteristics of teachers who are more skilled in utilizing family-centered practices and has wholly disregarded parents' perspectives.Instead, most studies of family engagement and family-centered practices rely on teachers' perspectives and self-reported behaviors.Family-centered practices are also often described as a static component of the classroom and universally implemented for all children and families (Trivette et al., 2010).In practice, however, family-centered practices are supposed to be individualized, flexible, and responsive to each family's unique circumstances (Ramos et al., 2015).Given the increasing diversity of families served by early care and education, it is unclear how similar families' experiences of family-centered practices are within the same classroom and whether and how the racial and ethnic composition of a classroom is associated with parents' and teachers' reports of family-centeredness.This study aims to address this gap in the literature by focusing on identifying teacher professional characteristics that relate to how parents perceive their relationships with their classroom teachers across a variety of classroom contexts.
Of similar importance, another factor related to how parents and teachers form effective relationships is the level of knowledge parents share about their own culture and family rearing practices (i.e.family-specific knowledge) with their classroom teachers.When parents and teachers engage in open communication, it encourages teachers to engage in more responsive practices and to be more attentive to family needs (Bromer et al., 2011).Establishing a solid parentteacher relationship promotes ongoing interactions and communication that empowers both teachers and parents (Swick, 2004).Therefore, it's critical for teachers to cultivate high-quality interpersonal relationships with parents in order to support parent engagement and foster trust with families.
However, little is known about the ways in which teacher and classroom characteristics relate to parents' comfort in engaging in this type of communication.Despite the potential benefits of a high-quality parent-teacher relationship, parents and teachers both report levels of dissatisfaction with their interactions (Lawrence-Lightfoot, 2003).It's critical that we examine these interactions in order to best support the ongoing development of these home-school relationships.Using nationally representative data, this study aims to identify specific teacher and classroom characteristics related to the use of family-centered classroom practices as well as how these characteristics relate to parents' comfort in knowledge-sharing.

Parent-Teacher Relationships in Early Childhood
Family engagement is a key component of early childhood education and highlighted in the performance standards of our nation's largest publicly funded preschool program, Head Start (U.S.Department of Health and Human Services, 2016).Head Start's two-generational model simultaneously addresses the needs of the child and family, but also recognizes the critical role of parents as a child's first and most influential teacher.A guiding principle of Head Start is the importance of building and maintaining high-quality relationships between program staff and participating families in order to support children's school readiness.
Characteristics that contribute to a positive parent-teacher relationship in early childhood include building trust, having open and ongoing communication, and collaborating (Elicker et al., 2013).When teachers foster strong ties with families by being responsive and open the lines of communication, families feel more welcome at school and become more engaged in the teacherparent relationship.Specifically, important elements of the parent-teacher relationship have emerged through prior research and serve as a foundation for defining parent-teacher relationship quality: family and background knowledge, attitudes about families, and teacher practices with families (Bromer et al., 2011;Forry et al., 2012).The two elements that this study focuses on are: family and background knowledge and family-centered practices.

Family Specific Knowledge.
Family-specific knowledge is defined as the knowledge and information teachers gain and learn about families in their care.This includes theoretical knowledge and understanding of factors that promote healthy family functioning, as well as knowledge about the specific families they are serving.Family-specific knowledge includes teachers' knowledge about families' culture, parenting beliefs, as well as an awareness of families' strengths and their needs.When parents and teachers engage in information sharing, it allows teachers to have an acute understanding of individual families and can allow teachers to engage in more responsive practices (Bromer et al., 2011).Family-specific knowledge provides teachers with a foundation for understanding individual children and families, provide the necessary information to be more responsive, and aid in the development of a high quality parent-teacher relationship (Forry et al., 2012).In this study, we are interested in examining parents' perception of their teachers' understanding, and responsiveness to this knowledge.Moreover, we're interested in understanding the role of teacher and classroom characteristics that may influence these perceptions.

Family-Centered Practices.
Family-centered classroom practices refers to an approach to working with families that honors and respects their values and choices, emphasizes the importance of information sharing in order for families to make informed decisions, and is individualized and responsive to families needs (Dunst, 2002).According to Dunst and Trivette (1996), family-centered practices have both relational and participatory components that are hypothesized to facilitate parent-teacher relationships.For example, teachers with strong intrinsic relational skills engage in more effective communication strategies that are responsive to families' goals and are more culturally responsive (Halgunseth et al., 2009).Important participatory skills include engaging in advocacy, connecting families to resources and information that is relevant to them, and engaging in joint decision-making (Bromer et al., 2011;Forry et al., 2012).
In reviewing the literature, researchers also found that when family-centered practices are implemented, parents' report positive feelings and satisfaction with the services provided for their children (Dempsey & Keen, 2008).Studies have also found that when providers engage and implement family-centered practices, this results in increased family engagement despite external barriers to participation (Brookes et al., 2006).These outcomes are not only linked to positive family outcomes, but also to improved child outcomes (Dunst et al., 2007), and providers' selfefficacy and competence (Brown et al., 2009;Trivette et al., 2010).
However, there is wide variability in the degree to which family-centered approaches are implemented across program types (McWilliam et al., 1995).Parents in infant/toddler programs rated their education programs as providing more family-centered practices compared to parents of preschool children (age 3 to 5), highlighting the importance of gathering parents' perceptions of implementation across early education programs.

Importance of Parent-Teacher Relationships in Head Start
Despite recognizing the importance of establishing strong partnerships with parents, many educators acknowledge that developing a parent-teacher relationship is challenging (Ratcliff & Hunt, 2009), especially if parents' and teachers' ethnic, cultural, or linguistic backgrounds differ.For families of children from ethnic minority backgrounds, parent-teacher relationships are particularly important such that fostering partnerships early on can have positive effects on children's outcomes throughout their school experience (Nzinga-Johnson, Baker, & Aupperlee, 2009).
With the current ongoing demographic shift and changes to Head Start teacher education mandates, majority of preschool-aged children in Head Start classrooms are taught by a White majority teaching workforce (Bassok et al., 2013), and families are more likely to experience cultural and linguistic differences, creating an additional barrier towards fostering strong parentteacher partnerships (Howard & Lipinoga, 2010).Parents' differing values, miscommunication, and feelings of "not being valued" are prominent, especially by Latinx parents (Carreón et al., 2005;Chrispeels & Rivero, 2001).Therefore, it's critical to identify factors that may support the development of high-quality parent-teacher relationships.

Measuring Multiple Aspects of the Parent-Teacher Relationship
The Family and Provider/Teacher Relationship Quality (FPTRQ) suite of measures (Kim et al., 2015) was developed to examine both parent and teacher perspectives on their relationship.The FPTRQ measures the constructs of Family-Specific Knowledge, Attitudes towards Families, Family-Centered Practices, and Environmental Features, from the perspective of both parents and providers.Therefore, this measure allows researchers to capture and examine the teacher-family relationship from multiple perspectives and comprehensively examine knowledge, practices, and attitudes using one reliable and valid questionnaire.Specifically, we are interested in understanding how parents perceive this relationship by examining parents' perceptions of their classroom teachers' family-specific knowledge and parents' perception of family-centeredness (as measured by family-centered practice) by their classroom teacher.

Factors Related to Teacher's Family-Focused Knowledge and Family-Centered Practices
Why is it important to examine parents' perceptions of their teachers' family-focused knowledge, and more importantly, how is this related to childrens' outcomes?First, when teachers have an acute awareness of the families they serve and/or have an awareness of barriers to engagement, teachers are better able to respond to families' needs by providing resources and engaging in more responsive ways.As such, parents develop positive feelings towards a child's early care and education teacher, which then are associated with positive academic outcomes among children.Conversely, researchers hypothesize that increased family engagement may lead to enhanced provider understanding of parent perspectives and of parents' goals for children in care (Powell et al., 2010).Therefore, it is necessary to examine teacher and classroom factors that may be related to how parents perceive these relationships, in order to better understand the phenomenology of the relationship.

Teachers' Personal and Professional Characteristics.
Preparing teachers and offering training focused on effective family engagement can positively impact teachers' attitudes about families, improve knowledge regarding families' roles in their children's education, and increase family engagement practices (Amatea et al., 2012;Ratcliff & Hunt, 2009).Research has indicated that new teachers, despite receiving training focused on working with families, still struggle with family engagement and are skeptical of families' roles in their children's education (Epstein & Sanders, 2006;Evans, 2013;Markow & Martin, 2005).Teachers' education and years of experience have both been found to be related to the quality of teacher-child relationships (Kontos & Wilcox-Herzog, 2001).However, little research has examined the role teacher education, experience, and training play in components of the development of teacher-parent relationships.
Another factor of interest is the teacher-child race match.When teachers and parents are similar with regard to their ethnic backgrounds, parents are more likely to report higher levels of parent engagement compared to mismatched dyads (Mundt et al., 2015).Although some researchers have highlighted that parent and teacher ethnic match might increase the parent's comfort in interacting with the classroom teacher, which results in more effective parent-teacher relationships, Head Start serves over 1 million children a year from a variety of backgrounds and contexts.Therefore, we were interested in further examining teacher's professional characteristics such as education, years of experience, and training, and how these factors are related to parents' perceptions of teachers' family-focused knowledge and family-centered practices, as well as the role of racial ethnic match in fostering family-centered practice.

Teachers' Perceptions of Family-Centered Practices.
Teacher perceptions of their own practice are extremely important and may highlight discrepancies between both parties in the parent-teacher relationship.If teachers' rate themselves highly, we expect parents to also agree (and therefore predict high parent-reported family-centered practices).If teachers are engaging in high-quality family-centered practices, we expect parents to both rate teachers highly in respect to teachers' family-centeredness but also in regards to their family-specific knowledge.
In a study conducted by Minke and colleagues (2014), researchers highlighted the importance of parents and teachers sharing the similar beliefs of their shared relationship.Their results suggest that shared perceptions of relationship quality may be important in understanding reports of child behavior and finding ways to support positive student outcomes, especially in the presence of problem behavior.Considering the context of Head Start, it's critical to examine how teachers' perceptions of family-centeredness aligns with parents' perceptions of their practice.

Classroom Context.
Researchers have suggested a critical need in teacher's preparation programs in terms of their limited in-depth knowledge and understanding of children's home context (Smith & Sheridan, 2019).With ongoing classroom demographic shifts, we must examine how the classroom context also plays a role in parents' perceptions of teachers' engagement in family-centered practices, and demonstrates family-specific knowledge.Prior studies have also suggested differential effects of classroom composition on classroom management practices (Klapp & Cliffordson, 2008), and have examined the teachers' linguistic responsiveness as a function of classroom composition (Sawyer et al., 2016).Linguistically and culturally responsive practices may take different approaches, such as training and professional development for duallanguage learners (DLLs) or emergent bilingual children.
In summary, when teachers foster strong ties with families by being responsive and sensitive with children and families, parents feel more welcome at school and become more involved in their children's education.Thus, it is imperative to explore additional variables that can be improved (such as teacher training and education on cultural sensitivity and responsiveness) so that children and families receive the support they need to fully engage.

The Present Study
This study aims to identify teacher and classroom characteristics related to teachers engagement in family-centered practice and utilization of family-specific knowledge as perceived by Head Start parents.Using a large, diverse, and nationally representative dataset, the current study aims to explore the following research questions: 1. What teacher personal and professional characteristics influence parents' perceptions of family-centered practices?2. To what extent is classroom composition related to parents' perceptions of family centered practices, and is this relationship moderation by teachers' self-reported family-centered practices?3. What teacher personal and professional characteristics are related to parents' comfort in disclosing family-specific knowledge?4. To what extent is classroom composition related to parents' comfort, and is this relationship moderation by teachers' self-reported family-centered practices?
Based on prior literature, it is expected that teacher education, years of experience, training, and ethnic/racial match will be associated with parents' perceptions of the utilization of these practices.As discussed earlier, teachers who identify with families either with similar values, expectations, or cultural backgrounds may be more likely to communicate more effectively, driving parents' abilities to feel comfortable in engaging more frequently teachers who identify with families either with similar values, expectations, or cultural backgrounds (Trumbull et al., 2001).
Current research has addressed the potential for teacher beliefs and practices to be influential, fostering or hindering positive parent-teacher relationships and parents' involvement, but research has not examined the extent to which the classroom composition may influence these processes.Thus, an examination of the teacher-related characteristics and practices is warranted to inform future examinations of parent-teacher relationship models.

METHOD Participants
Data from the Head Start Family and Child Experiences Survey (FACES) 2014-2015 cohort, a nationally representative sample of Head Start programs, centers, classrooms, children and their families will be used for this analysis.The FACES 2014 study employed a multistage sampling approach, first sampling Head Start programs across the country who served children between the ages of 3 and 5 years old, selecting up to two centers per program, and two classrooms per center.The total sample included 176 total programs, 346 centers, and two classrooms within selected centers.
The present study will only use data from classrooms from which teacher and family data were collected.The analysis sub-sample, therefore, will include data from 116 centers, 227 teachers, and 747 parents who completed the Spring survey.Majority teachers identified as female (96.5%), were either White (22.9%),Black (31.7%), or Hispanic/Latino (36.0%) and on average had taught for 13.54 years (SD = 8.95).Parents indicated the majority of households were single parent households (40.4%), used English as their primary language (75.1%), parents were not employed (33.1%), and majority maternal education was some college (31.8%).

Measures
Data for this study come from two parallel versions of the Family and Provider/Teacher Relationship Questionnaire (FPTRQ; Kim et al., 2015;Ramos et al. 2014) which was shortened for use in the FACES studies (Forry et al., 2012).
Teacher-Reported Family-Centered Practices.
The teacher measure of the FPTRQ (Kim et al., 2015;Ramos et al. 2014) is intended for early childhood education providers in center-and family-based care.This subscale of the FPTRQ asks teachers general questions about how they work with all parents of children in their care (e.g."how often have you met with or talked to parents about goals you have for their child").This family-centered practices subscale is intended to describe teachers' perceptions of family-centered practices (e.g."How often are you able to set goals with parents for their child").
Parent-Reported Relationship Quality Components.The parent version of the FPTRQ asks parents general questions about how they work with their child's (i.e."how often have you met with or talked to your childcare provider or teacher about goals you have for your child").
These two subscales measure: (i) parents' perceptions of family-specific knowledge (e.g."How comfortable would or do you feel sharing the following information with your childcare provider or teacher?") (ii) parent-rated classroom family-centered practices: (e.g."My childcare provider or teacher uses my feedback to adjust the education and care provided to my child").Parents reported their perceptions using a 4-point Likert scale (1 = Never, 2 = Rarely, 3 = Sometimes, 4 = Very Often).

Teacher Demographics.
Each selected teacher was asked to complete a survey on their teaching experience (number of years teaching in Head Start), educational background (the highest grade or year of school completed), teaching credentials (e.g.Child Development Associate), salary, and demographics (age, gender, race/ethnicity, see Table 1).The following demographics will be included in the analyses to attend to the primary aims of the study: teacher education, and training to support dual-language/diverse learners (DLLs).As cited earlier, teachers may use more culturally responsive practices, depending on the composition of the classroom.Therefore, we use the DLL-training variable, as a proxy for culturally responsive training.

ANALYSIS
Descriptive statistics were used to examine the means and standard deviations of the variables of interest.Before proceeding with planned hierarchical linear modeling (HLM; Bryk & Raudenbush, 1992) analyses, we examined the unconditional models to compute intraclass coefficients (ICCs; James et al., 1984).Hierarchical Linear Models (Bryk & Raudenbush, 1992) were then computed for two outcome variables: parents' perceptions of teacher's family-specific knowledge, and parents' perceptions of family-centered practices.These models included teacher's education, years of experience, DLL training, racial/ethnic match and classroom composition as predictors.Analyses were conducted in R (R Core Team, 2019).Parent/Child Demographics.In the parent survey, parents were able to report their educational backgrounds, income level, family structure, and demographic variables including race/ethnicity, age, and marital status.Similarly, they reported their child's race/ethnicity (if different or mixed), first language, age, and gender (See Table 2).The following child demographics of the classroom will be included in the analyses to attend to the primary aims of the study: classroom composition.The proportion of children of ethnically diverse backgrounds (including African-American, Latinx/Hispanic, Asian-American, American-Indian, and mixed-race) was calculated with the average Head Start classroom containing 76% of children from diverse backgrounds relative to their White peers.

RESULTS
When examining descriptive analyses of the dependent variables, we discovered non-normal distributions for both perceived family-centered practices and family-specific knowledge variables.The distribution of scores suggests that most parents report either high levels or extremely low levels of either family-centered practices and comfort in disclosing family-specific information, with both variables resembling a u-shape distribution.After examining, we transformed both outcome variables to reduce the skewness to a better fit model using a power transformation using the powerTransform package in R.This method uses the maximum likelihood-like approach of selecting a transformation of a univariate or multivariate response for normality, linearity and/or constant variance (Box & Cox, 1964).
In examining the correlations among key variables via Pearson's product-moment correlation, we found that teachers who reported greater levels of education also reported having more teaching experience (r = .09,p <.01), have DLL training (r = .33,p <.01), and report utilizing a greater number of family-centered practices (r = .26,p <.01; see Table 3).Parents' perceptions of familyspecific knowledge were associated with perceptions of family-centered practices (r = .45,p <.01).Meaning, parents who felt teachers comfortable in sharing cultural and familial background also endorsed experiencing responsive and family-centered classroom practices by their classroom teacher.Before proceeding with planned hierarchical linear modeling (HLM; Bryk & Raudenbush, 1992) analyses, we examined the unconditional models and found that intraclass coefficients (ICCs; James et al., 1984) were not significant, meaning that classroom membership did not explain a significant amount of the variance in each dependent variable (see Table 4).Only 4% of perceptions of family-centered practices were attributable to classroom membership and 1% of parents' comfort was explained by who their teacher was (James et al., 1984).

Examining Perceptions of Family-Centered Practices
Although nesting was not significant, there's an inherent hierarchical structure of the data, therefore, we decided to continue with the originally planned HLM analysis method to explore how the teachers' characteristics and racial/ethnic match is associated with perceived familycentered practice.
To address RQ1, to what extent does teacher personal and professional characteristics predict parents' perceptions of family-centeredness, we constructed an HLM model using teacher education, experience, DLL training, self-reported utilization of family-centered practices, and parent-teacher ethnic match as predictor variables with findings suggesting null effects (see Model 1, Table 4).We expected teacher professional characteristics, self-reported practice and ethnic/race match to be positively associated with perceptions of family-centered practices.This suggests that teachers' professional background and ethnic/racial match is not significantly associated with parents' perceptions of these practices by their classroom teacher.To answer RQ2, we then explored the role of classrooms' ethnic composition in predicting perceptions of family-centered practice.Specifically, we aimed to explore the differential effects of student classroom composition and teachers' self-reported practices, while controlling for teacher personal and professional characteristics.Model 2 considers the ethnic/racial composition of the classroom.When controlling for teacher personal and professional characteristics, teachers' family-centered practices significantly predict parents' perceptions of family-centered practice (B = 2.73, p < .01).This relationship was clearer when examining the interaction between teachers' self-reported practices and the different classrooms that were modeled.There was a significant interaction between teachers' self-reported family-centered practice, and classroom composition when predicting parents' perceptions of family-centered practice (B = -2.47,p < .05).In other words, this interaction suggests agreement between parent-and teacher-reported family-centered practice, and this agreement consistently changes depending on the classroom context.Figure 1 depicts this relationship such that in classrooms with a low minority percentage, teachers and parents agree more; but in classrooms with a high minority percentage, there is less of a relationship.

FIGURE 1. Interaction effects between classroom composition and teacher familycentered practices on parent perceived practices
Note.Classroom composition was determined based on the average percent of students of ethnic minority backgrounds in Head Start classrooms (76%).Each slope represents three different classroom compositions, with low (58%), average (76%) and high (93%) proportion of students who identify as of mixed-race or ethnic minority backgrounds (African-American, Asian, American Indian, Hispanic, etc.).
graph reveals a significant main effect for teacher self-reported practice.When teachers rate themselves highly, parents similarly perceive these practices at similar levels and tend to agree.We can see trend-level significance for classroom composition such that as classrooms become less homogenous (larger proportion of students of diverse backgrounds), parents perceive less of these family-centered practices.The significant interaction tells us that perceptions of practice vary as a function of the classroom composition.In classrooms with a relatively small proportion of children from ethnically diverse backgrounds (53%), both parents and teachers perceive greater levels of family-centered practices being utilized.Conversely, as the proportion of ethnically diverse children increases (93%), both parents and teachers report fewer family-centered practice.

Examining Family-Specific Knowledge
To address RQ3, we explored how the teachers' professional characteristics (education, experience, DLL training) and ethnic/race match was associated with parents' perception of family-specific knowledge.Results from this HLM model suggest teacher professional and personal characteristics are not predictive of parents' perceptions of family-specific knowledge, nor does parent-teacher race/ethnicity significantly predict parents' perceptions of family-specific knowledge.
To examine RQ4, Model 4 leverages the same predictors from Model 2, as well as the classroom composition variable, to explain variance in parents' perception of family-specific knowledge.While controlling for teacher race, classroom composition was not predictive of parent comfort but the direction of the beta suggests that when the proportion of ethnically diverse children decreases, parents report greater levels of comfort.Overall, the final two models infer that parents' perception of family-specific knowledge is a distinct relational variable as opposed to perceptions of family-centered practice.Although perceptions of family-centered practice may be associated with the context of the classroom in combination with teachers' self-reported practice, this does not apply to perceptions of family-specific knowledge.

DISCUSSION
Effective parent-teacher relationships are critical for children's academic and behavioral success (Garbacz et al., 2015;Minke et al., 2014).This collaboration contributes to school readiness, later academic success, greater academic motivation, and social-emotional skills (Halgunseth et al., 2009).The current study examines components of the parent-teacher relationship, and conceptualizes this relationship as both relational and pedagogical (intentional teaching practices).Family-centered practices include practices that support parent engagement including communication, collaboration and responsiveness to families' needs (Spielberg, 2011).When teachers are actively utilizing such practices combined with an understanding of the family context, teachers are better able to tailor their instruction to support children's learning (Blue-Banning et al., 2004).The purpose of this study was to examine factors that are associated with elements of an effective parent-teacher relationship in early childhood settings.
In addressing the current study aims, notable findings included the following: 1) we found a lack of nesting at the classroom level related to parents' perceptions of family-centered practice and parents' perceptions of family-specific knowledge; 2) we did not find that teacher professional and ethnic background to be associated with components of effective parent-teacher relationships; 3) parents and teachers agree have convergence and agreement when reporting on familycenteredness they perceive in classrooms.

Lack of Shared Experiences at the Classroom Level
Before conducting our analysis, we expected parents to report similar experiences of perceptions of practice as well as establish that these experiences may be attributable to the classroom teacher and the teachers' background.Interestingly, we found a lack of nesting within the classroom when calculating the intraclass coefficients (ICCs) for both outcomes.These results highlight the variability of parents' responses, which indicate considerable individual-level variability.Moreover, only 4% of parents' responses could be attributed to the classroom suggesting two alternative hypotheses: 1) parents' experiences and perceptions may be attributed to centers and center support to engage families; and 2) considering the nature of parent-teacher relationships, we must also attend to parent characteristics that may be associated to perceptions of practice not attributable to the classroom.
One example of center characteristics influencing the parent-teacher relationship is by examining program types.When comparing home-based programs to center-based care, the familiarity and intimate home context positively influenced parents' perceptions of their caregiver and promoted flexibility rather than rigid policies to respond to parents' needs (Bromer & Henly, 2004).Head Start has long recognized the importance of involving and supporting families.A provision of resources supports the systematic integration of family engagement efforts within Head Start centers, but these efforts must also be sustained by center leadership, administration and familysupport staff (Mattera et al., 2013).Examining center characteristics beyond the classroom may inform centers' approach to engaging families by utilizing practices similar to those in home-based settings to promote a welcoming and family-centered approach to early care and education.
Parent characteristics that may also be related to differential experiences are related to the amount and quality spent communicating with their classroom teacher such as employment status and time constraints (Christenson, 2004).Additionally, parents' expectations of this relationship, sense of self-efficacy, and attitudes have been shown to influence parent ratings of parent-teacher partnerships (Dunst & Dempsey, 2007).In gathering information from both perspectives, it is critical to understand how external characteristics may be associated with parent-teacher relationships within Head Start centers.In summary, the results from the current study are consistent with the general understanding and conceptualization of the parent-teacher relationship as multidimensional, and professional development programs need to address the ever-changing demographics and families that Head Start programs serve.

Teacher Background & Effective Parent Teacher Relationships
Within our sample, teachers' professional background (including education, experience and DLL training) was not associated with parents' perception of family-centered practice or parents' perceptions of family-specific knowledge, two key elements of the parent-teacher relationship.This suggests that perceptions of family-centered practices and family-specific knowledge may vary as a function of several individual, interpersonal, or classroom dynamics beyond teacher professional characteristics.This also suggests that current pre-service and in-service training is not directly contributing to an increase in the utilization of these practices, as reflected in the literature.Previous research has highlighted a combined effect of lack of training and confidence as barriers to effectively engaging in positive relationships with families, specifically with families of diverse backgrounds (Christenson, 2004).Additionally, family-centered training and professional experience is associated with positive attitudes about families, but not necessarily a change in behavior towards engaging in family-centered practices (King et al., 2003).
It is important to further examine how teachers can foster greater levels of parent comfort.In fact, it is critical to distinguish the unique difference between family-centered practices and the process and relational variable of comfort as comfort develops over time.We theorize that years in the education systems, combined with immigration status (Carreón et al., 2005), and family structure can negatively impact the development of comfort with classroom teachers.
It is also important to note we did not control for the length a child was in the program or for siblings.Information was not collected on the capacity at which parents knew teachers, rather, at their level of comfort in disclosing personal information.Given that parents' perceptions are a unique contributor to the conceptualization of the effective parent-teacher relationship, we must attend to other factors that may influence this variable beyond teacher characteristics.In addition to teacher professional characteristics and ethnic/racial match, previous research has identified trust and respect as critical in the ability for parents to engage in productive communication and collaboration with their classroom teacher.Conversely, it is imperative for parents to be able to trust teachers' expertise but also for teachers to gain that trust through support, communication and acceptance (Blitch, 2017).

Congruent Perceptions & Classroom Context
As previously established, parents' and teachers' perceptions of parent involvement often differ (Wong & Hughes, 2006).This is especially true when parents and teachers' perceptions of relationship quality differ (Minke et al., 2014).In the current study, teacher's self-reported practices positively predicted parents' perception of family-centered practice suggesting that parents' and teachers' perceptions of family-centered practice align.These shared perceptions are elements of an effective parent-teacher relationship such that shared perceptions and congruent beliefs are associated with relational trust (Bryk & Schneider, 2002).Further, when families and teachers have mutual beliefs, these interactions serve as a foundation for collaborative partnerships which can further support later academic success (Halgunseth et al., 2009).When parents are able to partner and communicate with their classroom teacher, these positive interactions set the stage for effective collaboration and partnership to support the child.
When examining the effects of classroom composition and teachers' practice on parents' perceptions of these practices, we found a differential effect of classroom composition on their perceptions of family-centered practice.When examining specific classrooms profiles (lower proportion of ethnic minorities versus higher proportion), we found that parents' perceptions of family-centered practice were higher in classrooms with lower proportion of students from minority groups compared to a predominantly ethnically diverse classroom.When examining teachers' self-reported family-centered practices, we see similar effects on how teachers are perceiving their own classroom practice.This finding supports how classroom context and student ethnic background plays a role in perceptions of classroom practices (Steinberg & Garrett, 2016).This may also indicate that the context of the classroom may also influence teachers beyond perceptions of practice, but the instruction taking place and teachers' behavior.Within this Head Start sample, the average classroom contained 76% of students of ethnic minority backgrounds.With the ethnic composition of classrooms changing and the growing number of classrooms becoming ethnically diverse, these findings identify a need to prepare teachers with specialized training.
In classrooms with predominantly ethnic minority students, teachers reported utilizing less familycentered practices compared to teachers in less ethnically diverse classrooms.Teachers may not feel as prepared to serve the needs of these diverse families to identify their use of family-centered practices.These results are timely and highlight the need to support teachers with specialized training in fostering improved parent-teacher relationships.Lack of training and confidence to engage families have been identified previously as barriers in foster effective relationships (Christenson, 2004).It is possible that current training around family and parental engagement is not sufficient to foster greater use of family-centered practices or parent comfort in sharing personal information with their classroom teachers.Although the results of this study suggest families are able to have congruent perceptions of classroom practices that involve collaboration, communication, and responsiveness, we must also attend to classroom context and its effects on perceptions of practice.As classrooms across the nation are becoming more diverse, teachers' must be trained to meet the needs of families entering Head Start.Although Head Start's family engagement framework and tenants highly encourage family engagement, this study has highlighted the need to further support teachers in engaging with these families and supporting them in the development of family partnerships.

CONCLUSION
The purpose of this study was to examine teacher professional and classroom characteristics that may be associated with components of effective parent-teacher relationship quality.Specifically, we wanted to explore how education, training, experience, teachers' perceptions of familycentered practice and ethnic/racial match support parents' perceptions of family-centered practices being used.Additionally, we explored the effects of ethnic/racial classroom composition on perceptions of family-centered practices and the extent to which teacher's had and utilized familyspecific knowledge.Exploring how such factors may contribute to the enhancement of parentteacher relationships can inform future efforts to understand how to provide support for centers and teachers in providing family-centered practices for families they serve.A focus on the parentteacher relationship has highlighted the dimensionality of this construct and the importance of supporting effective partnerships between the home and school environments.We found that pedagogical family-centered practices are unique from relational practices (increasing parent comfort).Additionally, teachers' who are more experienced, have more education and training do not necessarily predict engaging in these practices.Finally, we found that classroom composition has differential effects on classroom practices.The results highlight the need to further examine individual, interpersonal, classroom and center dynamics beyond teacher professional characteristics in order to provide support in the development of effective partnerships.

Limitations
Given the complexity of the design and size of data for the FACES 2014, there are numerous anticipated limitations and obstacles for the present study.A limitation of this secondary analysis is that while all parents got to rate individual teachers, teachers rated their average relationship with all parents in the classroom, thus the data is not truly dyadic.Future investigations can gather direct dyadic data (i.e.parent-teacher pairs rating the same relationship) to further explore issues of congruence in relationship ratings as well as variability.Additionally, as relationship questions were only asked at one-time point, the analyses cannot speak to the developmental trajectory or growth in relationships over the course of a school year.

TABLE 3 .
Pearson's Product-Moment Correlations Among Key Variables Note.Teacher experience is the number of years teaching; FCPs = family-centered practices.Correlation matrix included only completed observations, excluding missing data via pairwise deletion.* p ≤ .05,** p ≤ .01,*** p ≤ .001