The negotiation of head start teachers’ beliefs in a transborder community

Early Childhood Research Quarterly 47 (2019) 134–144 Contents lists available at ScienceDirect Early Childhood Research Quarterly The negotiation of head start teachers’ beliefs in a transborder community Sarah Garrity ∗ , Alyson Shapiro, Sascha Longstreth, Jillian Bailey San Diego State University, 5550 Campanile Dr., San Diego, CA, 92182, USA a r t i c l e i n f o Article history: Received 17 November 2016 Received in revised form 28 September 2018 Accepted 31 October 2018 Keywords: Early childhood teacher beliefs Cultural communities Head Start Cultural conceptions of teaching and learning Latino culture Border pedagogy a b s t r a c t Using a cultural communities approach as our guide, the purpose of this exploratory study was to examine the beliefs of early childhood educators in a transborder community located along the U.S./Mexico border and known for its large Latino population. Participants were 77 Head Start associate teachers, teachers, site supervisors, and other education staff. The majority were Latina (83%) and bilingual in both Spanish and English (70%). The Teacher Beliefs Q-Sort was used to examine participants’ beliefs regarding discipline and behavior management, teaching practices, and children. Factor analyses suggested patterns of teacher beliefs that reflected both collectivist and individualistic orientations to teaching and learning. Latina educators scored significantly higher on factors that appeared consistent with a collectivist orientation and seemed to reflect the cultural value of raising children to be bien educado. Participants’ beliefs were also compared with those of faculty exemplars, and participants who prioritized items in way similar to the experts tended to be more educated and were more likely to be non-Latina. Findings point to the need to examine the intersection of education level, ethnicity, and recommended practice and how teachers negotiate these beliefs in transborder settings. Published by Elsevier Inc. It is widely recognized that the teacher is an essential element of a high quality early care and education (ECE) program. Teaching and caring for young children is demanding work, and teachers must prioritize competing demands and make decisions about a multitude of factors throughout the day. Pajares (1992) has proposed that teacher beliefs are the compass that guide the decisions teachers make on a daily basis, and many researchers interested in improving teaching and learning have cited the importance of teacher beliefs on practice (Calderhead, 1996; Fang, 1996; Nespor, 1987; Richardson, 1994). For the purposes of the current study, we define teacher beliefs as “an individual’s judgment of the truth or falsity of a proposition” (Pajares, 1992, p. 316), while acknowledging that these beliefs also encompass values, attitudes, opinions, judgments, conceptual systems, personal theories, and repertories of understanding (Pajares, 1992). Unlike knowledge, teacher beliefs are subjective, evaluative, and affective (Nespor, 1987), and reflect what teachers assume to be true about learning, learners, the learning environment, and the content being learned (Kagan, 1992). Given the subjective, personal, and experiential nature of teacher beliefs, we propose that ∗ Corresponding author. E-mail address: sgarrity@sdsu.edu (S. Garrity). https://doi.org/10.1016/j.ecresq.2018.10.014 0885-2006/Published by Elsevier Inc. they are deeply rooted in culture, and in order to explore the complexity of teacher beliefs, researchers must adopt a contextual view of development (Bronfenbrenner & Morris, 1998) that takes into consideration the various contexts (home, school, community) in which teaching and learning occur. Rogoff’s (2003) theory of culture and human development, particularly the construct of cultural communities, expands on the contextual model by proposing that shared experiences, practices, and goals lead to groups of individuals with common beliefs, values, and ideologies that extend over several generations. In this view, community held beliefs about the role of education, how education should be delivered, and the goals of education, are part of a belief system developed through participation in cultural communities. Using the cultural communities framework as our guide, the purpose of this exploratory study was to examine the beliefs about discipline, teaching, and children of Head Start educators (associate teachers, teachers, site supervisors, and education specialists) working in a transborder community located along the U.S./Mexico border and known for its large Latino population. Romo and Chavez (2006) note that the geopolitical border between the United States and Mexico represents a unique space in which languages, cultures, communities, and countries end, begin, and blend, and thus reflect the complexity, juxtaposition, and intersection of identities, economies, and educational and social issues. Reflecting our con- S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 textual paradigm and cultural communities lens, border pedagogy positions education as a political, social, and cultural enterprise (Giroux, 1988). While the study of border pedagogy has emerged as a robust field of study in the K-12 educational system (e.g., Cline & Necochea, 2003, 2006; Necochea & Cline, 2005), it has yet to be extended to the field of early childhood education. The current study aims to address this gap in the literature by exploring the beliefs of teachers working in a geographic area known for its sociocultural complexity and linguistic diversity, using a theoretical framework that foregrounds cultural practices and values. An additional goal of this study is to move beyond some of the limits of the survey methods often used to study teacher beliefs and obtain a larger sample size than is typically possible in qualitative research through the use of Q-sort methodology (Block, 1961; Stephenson, 1935). Our own positionality as White teacher educators working in a federally designated Hispanic serving institution located in a transborder region has also informed our interest in the cultural nature of teacher beliefs, and Q-sort methodology allowed for the examination of differences and/or similarities between our beliefs and those of practitioners working in our community. 1. Previous research on teacher beliefs Early studies on the beliefs of early childhood teachers centered on using survey methods to measure whether teacher beliefs reflected developmentally appropriate practice (DAP). Researchers have reported a relationship between education and the alignment of teacher beliefs with DAP, as early childhood majors consistently espouse practices that are more aligned with DAP than those without degrees or with a degree in another major (Buchanan, Burts, Bidner, White, & Charlesworth, 1998; File & Gullo, 2002; Smith, 1997; Spidell-Rusher, McGrevin, & Lambiotte, 1992; Vartuli 1999). McMullen and Alat (2002) found that a four-year degree, regardless of major, was associated with stronger endorsement of DAP than was a two- year degree in early childhood. Researchers have also examined the relationship between teacher beliefs and classroom practices, and although there is some evidence suggesting a positive relationship between teacher beliefs and practices (Bryant, Clifford, & Peisner, 1991; Charlesworth, Hart, Burts, & Hernandez, 1991; Charlesworth et al., 1993; McMullen et al., 2006; Oakes & Caruso, 1990; Vartuli, 1999), others have reported inconsistent or only weak correlations (File, 1994; Kontos & Dunn, 1993; Tiano & McNeil, 2006; Wilcox-Herzog, 2002). Wen, Elicker, and McMullen (2011), for example, found weak correlations between early childhood teachers’ self-reported beliefs and observations of their actual classroom practices and suggested that future research should explore “when, how, for whom, and under what conditions are teachers’ beliefs and practices consistent?” (p. 962), further supporting the need to examine beliefs using a more contextualized and nuanced approach. Several researchers interested in the relationship between beliefs and practices have examined how teacher beliefs are shaped by participation in cultural communities. Wishard, Shivers, Howes, and Ritchie (2003) examined teacher beliefs across 22 childcare programs and found that program and teacher articulated practices were deeply rooted in ethnicity. African American teachers, for example, articulated a strong belief in the importance of didactic learning practices with African American children, while emphasizing a focus on learning English with Latino children. McGill-Franzen, Lanford, and Adams (2002) examined the cultural nature of teachers’ language and literary-related beliefs and practices and concluded that because different communities used and valued language and literacy differently, children were socialized into the literacy practices of the preschool communities to which they belonged. Similarly, Sanders, Deihl, and Kyler (2007) found 135 that African American directors working in subsidized child care centers in a low-income area of Los Angeles altered DAP classroom practices to ensure that the children participating in their programs were prepared to enter into a society plagued by racial and social class stratification. Cross-cultural research also points to the role of culture and setting on teacher beliefs and reported practices. Wang, Elicker, McMullen, and Mao (2008) examined the similarities and differences between American and Chinese early childhood educators’ beliefs about DAP and personal and cultural influences on these beliefs. Results indicated that in both groups a higher education level was related to the endorsement of child-initiated learning. For Chinese teachers, contextual factors, including class size and setting, were related to beliefs. Teachers with very large class sizes endorsed a more teacher-directed curriculum, and teachers in urban settings valued child initiated learning significantly more than those in rural settings. Several studies have highlighted the need to attend to within group variability when considering racial and ethnic background and its relationship to teacher beliefs and practices. Garrity and Wishard Guerra (2015) used a cultural communities lens to explore beliefs about classroom language use of two Head Start teachers to gain insight into how teacher beliefs influence classroom practices, particularly language of instruction. Analysis of teacher interviews and assessments of classroom language use indicated that teachers had different beliefs about language use, despite their coming from a very similar ethnic and linguistic background, and that these beliefs were translated into classroom practice. Adair, Tobin, and Arzubiaga (2012) analyzed teacher interviews conducted as part of the Children Crossing Borders (CCB) study in the United States and compared the perspectives of immigrant and non-immigrant teachers, focusing specifically on the cultural expertise of immigrant teachers who work within their own immigrant community. Findings suggested that immigrant teachers were often unable to apply their full expertise to the care and education of children with whom they shared a similar cultural and linguistic background and felt a disconnect between their cultural knowledge and their pedagogical training. This disconnect was particularly apparent in the academic, teacher directed teaching they themselves had experienced as children (and was expected by the immigrant parents in their classroom) and the more play-based orientation promoted by DAP and in their college coursework and on the job training. This finding recalls Delpit’s (1986, 2006)Delpit, 1986Delpit’s (1986, 2006) work describing the struggle faced by African American teachers when trying to use their cultural knowledge to address the needs of African American students. 1.1. Studies using the Teacher Belief Questionnaire to measure teacher beliefs The Teacher Belief Questionnaire (TBQ) was developed to measure teacher beliefs about discipline, teaching, and children and to examine differences in beliefs and priorities among groups. The TBQ has been used to examine whether teachers enrolled in a professional development course held beliefs about discipline and teaching that were similar to those of exemplars (Rimm-Kaufman and Sawyer, 2004) and to identify differences in beliefs and priorities between groups of teachers with different experiences and training (Rimm-Kaufman, Storm, Sawyer, Pianta, & LaParo, 2006). The tool has also been used to describe beliefs of preservice teachers (Decker and Rimm-Kaufman, 2008) and to compare whether students at different points in their educational program held different beliefs and how these beliefs were related to those of faculty exemplars (LaParo, Siepak, & Scott-Little, 2009). Because Q-sort methodology allows for the investigation of groups of individuals who may have different beliefs and priori- 136 S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 ties, it is especially suited to our exploration of teacher beliefs in a border region. However, although the need to recognize within group differences and the fluid nature of beliefs is at the core of a cultural communities approach, it is critical to note that researchers have identified several social behaviors and goals for children’s development that are associated with the Latino culture. 2. Latino cultural values 2.1. A collective versus individual approach to child rearing Several important dimensions or patterns have been used to describe the different approaches to child rearing observed among cultural groups. One way in which these cultural variations has been conceptualized is as falling into two distinct categories: individualist and collectivist. Drawing on research and theory from developmental psychology (Greenfield & Bruner, 1966), anthropology (Whiting & Whiting, 1975), individualistic and collectivist approaches reflect different goals for teaching and learning. Individualism represents the dominant culture in the United States, in which children are socialized to be independent and strive for individual achievement, thinking, and self-expression. Personal choice is valued, as are egalitarian relationships and role flexibility. Collectivist cultures, on the other hand, are reflective of many immigrant cultures that value inderdependence and group success, adherence to norms and group consensus, and respect for authority and hierarchical roles (Rothstein-Fisch, Greenfield, Trumbull, Keller, & Quiroz, 2010). Parents transmit these cultural values, beliefs, traditions, and behavioral norms to their children via the process of ethnicracial socialization, in which children are socialized to become competent members of their own culture (Hughes et al., 2006). 2.2. Raising children to be bien educado Research indicates that Latino families often engage in practices that socialize children to become bien educado (Valdés, 1996). To be bien educado means to be “well educated”, “well brought up”, or “well bred”, and reflects the value placed on a strong moral character and highly valued social behaviors that is often found in Mexican American families (Harwood, Leyendecker, Carolson, Asencio, & Miller, 2002; Valdés, 1996; Valenzuela, 1999). The value of respect, or respeto, requires children to be highly considerate of adults, especially their elders (Halgunseth, Ispa, & Rudy, 2006). Respect includes not interrupting or arguing with adults (Delgado-Gaitan, 1994) and serves to reinforce patterns of authority originating in heritage cultures that are organized according to age, gender, occupational statues and familial roles (DelgadoGaitan, 1994; Valenzuela, 1999). This value extends to teachers, as the authority of the teacher is inculcated in children at a very young age (Goldenberg & Gallimore, 1995; Valdés, 1996). From a collectivist standpoint, respect also serves to maintain harmony within the extended family (Marin & Marin, 1991). Obedience is closely related to respect, and Valdés (1996) described that obedience is expected of Latino children in terms of both understanding who is giving the orders and following these instructions with respect. In this way, children are expected to be obedient, know what to do, and to comply without questioning. Similarly, cooperation reflects a goal of collectivist cultures and is closely related to familismo, the notion that children have a responsibility to contribute to the well-being of the family (Cauce & Domenech-Rodríguez, 2002). Cooperation involves an intentional focus on the interests of the group, and children are expected to engage in behaviors that result in sharing and harmonious social interactions with siblings and peers (García Coll & Magnuson, 2000; Valdés, 1996; Valenzuela, 1999). In sum, our examination of the extant literature on teacher beliefs highlights their complexity and suggests that they may have their roots in cultural contexts and develop via participation in cultural communities. Our research objectives are described below. 3. Research objectives The primary goal of this study was to examine the beliefs of early childhood educators working in a geographic area known for its sociocultural complexity and linguistic diversity, using a theoretical framework that foregrounds cultural practices and values. We aimed to address this goal via an exploratory study with three research objectives: 1) to describe early childhood teacher beliefs about discipline, teaching practices, and children, with a particular focus on elucidating the beliefs of Latina educators that may have their roots in cultural values; 2) to examine associations between the beliefs of early childhood faculty exemplars and practitioners; and 3) to explore associations between early childhood educator beliefs (results from objective one), the associations between early childhood educator beliefs and those of exemplars (results from objective 2), and demographic factors, including ethnicity, education level, age, and years working in the program. 4. Methods 4.1. Research context The current study was conducted at a large Head Start agency located minutes from the U.S./Mexico border and characterized by it large Latino population. The term Latino describes individuals living in the United States whose origins can be traced to the Spanish-speaking regions of Latin American, including the Caribbean, Mexico, Central American, and South America (Flores, 2000). The Latino population is expected to grow to 119 million by 2060 in the United States (U.S. Census Bureau, 2014), and Latinos comprise almost 20% of school-age children; it is estimated that this portion will grow to 40% by 2050 (U.S. Census Bureau, 2001). During the 2016 fiscal year, 37% of Head Start participants identified themselves as Hispanic/Latino and almost 24% of families spoke Spanish at home (Office of Head Start, 2016a, 2016b). U.S. census data (2014) indicated that 58% of residents living in the city where we conducted the current study are Hispanic/Latino, and 30.8% of the entire population is foreign born. Just over 57% of residents over the age of five speak a language other than English in the home. During the 2014–2015 school year, the time period in which data were collected, approximately 85% of the children/families served by the program were of Hispanic/Latino origin, and 49% identified their first language as Spanish. Hiring staff that represent the local community and the children and families served has long been a goal of the Head Start program, and this is reflected in the current study, as the majority of our participants were Latina (83%) and bilingual in Spanish and English (70%). The early childhood educators in our sample all worked in the Head Start preschool program, and classrooms were similar in terms of group size (no more than 20 children per classroom) and teacher to child ratio (1 teacher for every 10 children). 4.2. Participants Participants were 77 Head Start education staff (associate teachers, teachers, site supervisors, and education specialists) from a Head Start agency located close to the U.S./Mexico border. Participants were recruited during a regularly scheduled agency-wide staff development meeting for education staff. Eighty-seven staff were originally interested in the study and filled out the initial S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 137 Table 1 Demographics of participants in the whole sample and in the Latina and non-Latina subsamples. Demographic characteristics Gender Female Male Ethnicity Latina Caucasian African American Asian Multi-Racial Other Language English only Spanish only Spanish and English Education High school or GED Associates degree Bachelor’s degree Master’s degree Position Associate teacher Teacher Site Supervisor Education Specialist Other Age 21–30 31–40 41–50 51–70 Whole sample Latina subsample Non-Latina subsample N % N % N % 77 0 100 0 63 0 100 0 13 0 100 0 63 2 4 2 1 4 82.9 2.6 5.3 2.6 1.3 5.3 63 0 0 0 0 0 100 0 0 0 0 0 0 2 4 2 1 4 0 15.4 30.8 15.4 7.7 30.8 12 5 54 16.9 6.5 70.1 2.0 5.0 51.0 3.4 8.6 87.9 10 0 3 76.9 0 23.1 6 33 20 14 8.2 42.9 27.4 19.2 6 31 17 6 10.0 51.7 28.3 10 0 2 3 8 0 15.4 23.1 61.5 25 35 10 1 4 33.3 46.7 13.3 1.3 5.4 24 29 5 1 3 38.7 46.8 8.1 1.6 4.8 1 6 5 0 1 7.7 46.2 38.5 0 7.7 9 27 19 19 12.2 36.5 25.7 25.7 8 23 14 16 13.1 37.7 23.0 26.2 1 4 5 3 7.7 30.8 38.5 23.1 Note: These descriptive statistics are based on the original data (before multiple imputation), and the percentages reported are the valid percentages after adjusting for missing data. demographics questionnaire and informed consent form; however, only 77 completed the Q-sort. There were no significant differences in demographics between those who completed the Q-sort and those who did not. Four $50 gift cards were raffled off as an incentive for participation. Participants included associate teachers (n = 35, 33%), teachers (n = 35, 47%), site supervisors (n = 10, 13%), and education specialists (n = 1, 2%). Table 1 describes demographic characteristics of participants in the full sample, the Latina subsample, and early childhood educators from other ethnicities for the purpose of comparison. forcing the prioritization of some statements over others. In this way, the TBQ required that participants rank their beliefs in relation to other beliefs rather than stating agreement or disagreement with each belief. Both face and construct validity of the TBQ have been established by assessing associations between the Q-sort and interviews regarding teaching beliefs (construct validity) and via teacher questionnaires (face validity). Reliability of the measure has also been established, with Spearman correlation coefficients ranging from .50 to .95 with a mean of .71 (SD = .11) (Rimm-Kaufman et al., 2006). 4.3. Measures 4.4. Procedures 4.3.1. Demographic Information A survey was developed by the researchers to obtain information about participants’ demographics. The survey assessed age, race, linguistic and educational background, position at Head Start, and years worked both in the field and in their specific position. 4.3.2. The Teacher Beliefs Q-sort (TBQ) (Rimm-Kaufman et al., 2006) The TBQ is comprised of three 20-item Q-sort exercises developed to assess teacher’s priorities among their beliefs. Q-sort 1 examines priorities about discipline and behavior management, Qsort 2 focuses on teaching practice priorities, and Q-Sort 3 assesses beliefs about children. Each Q-sort has five anchor cards and 20 statement cards. Anchor cards describe the scale used to rate each statement. For example, Q-sort 1 contains five anchor cards ranging from least characteristic of my approach or beliefs about discipline and behavior management to most characteristic of my approach or beliefs about discipline and behavior management. Participants must place only four of the 20 statements under each anchor card, ensuring that statements are equally distributed across the five anchors and Approximately half (48%) of participants completed the Q-sort during the morning session of a pre-designated staff development day, and the remaining participants completed the Q-sort during the afternoon session. Once informed consent was obtained, participants completed the demographic questionnaire. A pen and paper version of the TBQ was used in this study, and participants were instructed to complete the task independently. The first author explained the Q-sort technique, and participants were given a bag containing all of the statement cards broken down by Q-sort. All participants completed the Q-sorts in order (beginning with Q-sort one), and were instructed not to move on to the next sort until everyone had finished and placed the completed sort in their bag. Because this technique can sometimes be challenging, members of the research team were available to answer questions related to the procedure. To avoid the possible contamination of scores, members of the team walked around the room to ensure that participants did not discuss their responses and were completing the task independently. Data collection took approximately one hour to complete. 138 S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 4.5. Analytic approach Descriptive statistics were first used to describe the demographic characteristics of participants. The methods employed to examine the Teacher Belief Q-sort data in relation to ethnicity and other demographics were a replication of those used by Rimm-Kaufman and associates in their research validating this measure (Rimm-Kaufman et al., 2006). Factor analysis was used to describe early childhood educator beliefs about discipline, teaching practices, and children. The criterion method was used to examine associations between the beliefs of early childhood educators working in a transborder community and those of ECE experts reflecting current best practice recommendations in the field, including DAP. Factor scores and criteria correlation scores were created, which are described in detail in the sections below. The factor scores (reflecting teacher beliefs) and criterion correlation scores (reflecting associations between teacher beliefs and those of faculty exemplars) were then used in subsequent analyses to explore associations between the ECE teacher beliefs, the associations between teacher beliefs and those of exemplars, and demographic factors. Differences across ethnic groups were examined using independent sample t-tests for continuous variables (e.g., time in position) and Mann–Whitney U tests for ordinal scale data (e.g., education level scaled from high school diploma (low) to Master’s degree (high)). Associations between the variables of interest (factors and criteria correlation scores) and demographics were examined using Pearson’s correlations for continuous variables and Spearman’s correlations for ordinal scale data. Multiple linear regressions were also employed to disentangle the associations between the predictor and outcome variables. Effect sizes have been reported for all analyses, with eta squared effect sizes being calculated from the data yielded from the Mann–Whitney U tests. Effects sizes are typically interpreted across fields as small (Cohen’s d = .2, r2 = .01, 2 = .02), medium (d = .5, r2 = .09, 2 = .13), and large (d = .8, r2 = .25, 2 = .26), with Cohen (1988) explaining that effect sizes can differ across fields, and suggesting that anything explaining over 13.8% can be considered a large effect size in the social science field (reflected by an R2 , r2 , rs2 above .14). All statistics were conducted using the SPSS software package. The alpha significance values reported are two tailed p values. 4.5.1. The factor analytic method Meaningful factors were created through a factor analytic method to allow for the possibility that the factors emerging from such analyses may be different for early childhood educators and/or Latina educators. An exploratory factor analysis was first used for each of the three Q-sorts to identify the number of factors to be created for each sort, specifying that the factors must have eigenvalues greater than one. The scree plot for each factor analysis was used to ascertain the number of factors for each sort to be extracted by determining where the variance flattened out as recommended for researchers using standard statistical packages (Costello & Osborne, 2005). A principal components analysis was then conducted indicating the number of factors to be extracted for each sort, and using varimax rotation to maximize orthogonality of factors (S. DiStefano, Zhu, & Mindrila, 2009). Only items with loadings of .4 of higher were allowed to load onto factors. Factors were then calculated using a standardized summation method in which the mean was calculated for the standardized z scores of the items loading onto each factor. Items that loaded positively were added and items that loaded negatively were subtracted. This is one of several methods described as acceptable by DiStefano et al. (2009), and is considered particularly useful in examining associations across factors with different numbers of items, and to preserve the variation of the original data (Tabachnick & Fidell, 2001). The ω coefficient estimates of reliability were calculated for the factors because this method has been recommended for congeneric measures from the factor analysis framework (McDonald, 1999; Trizano-Hermosilla & Alvarado, 2016). The ω reliability coefficients ranged from .65 to .76, reflecting acceptable internal consistency. Specific reliability coefficients are reported in the results section below. Once computed, these factors were used as the outcome variables in subsequent analyses to examine associations between the factors and demographic variables. For example, factors scores for the Mistrust of children factor (described in the results section) were compared for Latina and non-Latina educators using t-test analyses. 4.5.2. The criterion method A criteria sort method was employed in order to examine associations between the beliefs of Head Start education staff and those of ECE faculty exemplars reflecting best practice recommendations. This method creates an exemplar sort to which individuals in the early childhood professional sample can be compared. The first and third authors, both of whom are ECE faculty, served as experts and completed the TBQ with the intent of sorting in a way that reflected recommended best practices in the field regarding approaches to discipline and behavior management, teaching practices, and children. The exemplar Q-sort was completed both within one week of the time participants completed the Q-sort and again nine months later to determine reliability over time as well as across raters. A composite exemplar score for each item within each sort was calculated by taking the mean of these sorts for both experts and both time points (the mean of all four expert sorts − two expert sorts and two time points). An interclass correlation of .95 reflects excellent reliability across both the two experts and across the two time points of measurement. A Cronbach alpha of .96 for the composite mean exemplar score also reflects excellent internal consistency. A series of Spearman’s correlations were then conducted, correlating the scores of each teacher for each Q-sort (one variable) with the composite exemplar scores (second variable). Spearman’s correlation was used because this method is ideal for examining ranked data. In order to do this, the data file was arranged so that the rows reflected each Q-sort question, and the columns represented the participant’s responses. This resulted in 3 Spearman’s rs correlations for each educator (one correlation for each sort). The resulting sort-level correlations for each teacher were then used to create a criterion correlation score for each sort reflecting the degree of relatedness of each participant with the exemplars. Standardized z-scores of these criterion correlation scores were then created for each participant for each sort. Thus, the criterion correlation scores are the standardized sort-level correlations for each teacher. The criterion correlation z-scores were then used in further analyses to elucidate our understanding of how the beliefs’ of participants compared to those of the faculty exemplars. 4.5.3. Missing data Although members of the research team were available to provide direction on how to complete the Q-sort activity, some participants had difficulty completing the task and data were missing at random for Q-sort and demographic items. In these cases, multiple imputation was used to replace the missing data using the multiple imputation procedure in SPSS. The imputation values were constrained such they fell within the possible range for that item. Results of the pooled data were reported. 5. Results 5.1. Associations between ethnicity and demographics A series of t-test (for continuous data) and Mann–Whitney U (for ordinal scale data) analyses were conducted to compare Latina S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 139 Table 2 Factor loadings for principal components analysis with varimax rotation of Q-Sort 1 on beliefs about classroom discipline and behavior management. Items from Q-Sort 1 Respect for children/emphasis on self-regulation 1. The primary goal in dealing with students’ behavior is to establish and maintain control. 2. A noisy classroom is okay as long as all the students are being productive. 4. When students are engaged in interesting problems and challenging activities, they tend to have very few discipline problems. 7. Peer interactions are best left to recess and snack time. 10. Classroom rules should be discussed and posted. 11. Self-monitoring behaviors (or self-regulation) are important skills for students to develop. 13. Students should try to solve conflicts on their own before going to the teacher. 14. Rules for the students’ classroom behavior need to be reinforced consistently. 15. Praise from me is an effective way to change students’ behavior 16. Students learn best in primarily teacher-directed classrooms. 17. If I treat students with respect, kindness, and concern, there are less behavior problems. 18. Verbal punishment is an unacceptable means of controlling students’ behavior; I believe it is more important to use only positive management techniques. −.49 Rule focused/teacher directed −.49 −.44 −.58 .41 .51 .40 .55 .47 .76 .52 −.42 Note: Factor loadings < .40 were suppressed, and two factors were indicated in the principal components analysis. Only items that loaded onto factors are included in this table. educators and those from other ethnicities. Results indicated that Latina teachers were significantly less educated (M = 3.38, Mdn = 3, reflecting Associates Degrees) than teachers from other ethnicities (M = 4.46, Mdn = 5, reflecting Masters degrees), U (75) = 144.5, 2 = .2 (moderate to large effect size), p < .001. Latina teachers were also more often in lower level positions (M = 1.71, Mdn = 2, reflecting associate teacher and teacher positions) than participants from other ethnicities (M = 2.22, Mdn = 2, reflecting teacher and site supervisor positions), U (75) = 226.0, 2 = .26 (large effect size), p = .008. These significant differences in demographics suggest qualitatively different samples. Thus, subsequent analyses were done both for the full sample reflecting early childhood education providers in a border community and for the Latina subsample. Further analyses comparing Latina teachers to those from other ethnicities were also conducted to help elucidate the teachers’ beliefs associated with Latina culture. This approach allowed for the exploration of teacher beliefs with a particular focus on elucidating the beliefs of Latina early childhood educators that may have their roots in cultural values. Because of the association between education level and ethnicity, with Latina participants having significantly less education, a series of multiple linear regressions were conducted in an attempt to disentangle the associations between the outcome variables and ethnicity versus education. In these cases, one variable (e.g., ethnicity) was entered into the regression to account for variance explained by that variable, and then the second variable (e.g., education) was entered in the second step. This process was then repeated with the second variable entered into the regression first (e.g., education). The significance of the change in R squared was and reported where significant, reflect- ing any significant variance explained by the second variable above and beyond the first variable. 5.2. Factor extraction and creation Factor analysis was used to describe early childhood educator beliefs about discipline, teaching practices, and children. An exploratory factor analysis indicated that two factors should be extracted for each Q-sort based on both the scree plots and the percent of variance explained dropping off after two factors. A principal components analysis was then conducted to extract two factors for each Q-sort. The first two factors for Q-sort one accounted for 23% of the variance, the first two factors for Q-sort two accounted for 27% of the variance, and the first two factors for Q-sort three accounted for 24% of the variance. Factors were created using the sum method described above. The ω congeneric reliability estimates were calculated for each factor. The two factors extracted for Q-sort one (QS1) assessing teacher beliefs regarding discipline and behavior management were named: 1) the Respect for Children/Emphasis on Self-regulation factor (ω = .72), 2) and the Rule Focused/Teacher Directed factor (ω = .65). The two factors extracted for Q-sort two (QS2) assessing beliefs about teaching practices were named: 1) the Collaboration and Reflection Valued factor (ω = .71), and 2) the Teacher Directed/Routine Focus factor (ω = .70). The two factors extracted for Q-sort three (QS3) assessing teacher beliefs regarding children were named: 1) Mistrust of Children (ω = .76), and 2) Focus on Choice, Individual Strengths factor (ω = .71). Tables 2–4 detail the factor loadings for each item that loaded onto each factor. Table 3 Factor loadings for principal components analysis with varimax rotation of Q-Sort 2 on beliefs about teaching practices. Items from Q-Sort 2 1. Having a morning routine. 2. Talking about our plan or schedule for the day 3. Welcoming each student by name to class 4. Doing an activity to create a sense of community 7. Having students share experiences 9. Conducting the business of the classroom (e.g. attendance) following a set routine 10. Reflecting and talking about something, such as a social interaction, that “worked” or “didn’t work” in our class 11. Reflecting on the content of an academic lesson and talking about what we learned 12. Using drill and recitation for factual information (calendar, etc.) 13. Modeling behaviors for students. 14. Introducing new objects or new activities in the room through demonstration 18. Using whole group instruction 19. Using a theme or project based approach to instruction 20. Working on group projects Collaboration and reflection valued Teacher directed/ routine focus .48 .75 −.62 −.55 .43 −.53 −.50 .53 .49 .48 .57 −.76 −.50 −.40 Note:Factor loadings < .40 were suppressed, and two factors were indicated in the principal components analysis. Only items that loaded onto factors are included in this table. 140 S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 Table 4 Factor loadings for principal components analysis with verimax rotation of Q-Sort 3 on beliefs about children. Items from Q-Sort 3 Mistrust of children 2. Many of the students in my class try to get away with doing as little work as possible 3. Students should feel as though they are “known” and “recognized” in the classroom 5. Each one of my students teaches me something 6. Almost all students are equally likeable and enjoyable 7. Most students respect teachers and authority 8. Students seldom take care of their materials if they are not supervised 9. Students learn best when they have good role models for their behavior 10. Students need some choice of activities within the classroom. 13. Students meet challenges best when they feel that their teachers care about them 14. Students need to feel safe and secure in the classroom 16. Students need to have their strengths recognized to promote learning 19. Some students show little desire to learn 20. Students are more motivated by rewards than they are by the acquisition of competence .49 Focus on choice, individual strengths .61 −.54 −.48 −.54 .53 −.46 .63 −.61 −.70 .47 .52 −.42 Note:Factor loadings < .40 were suppressed, and two factors were indicated in the principal components analysis. Only items that loaded onto factors are included in this table. 5.3. Criterion correlation score creation Spearman’s correlations assessed the relatedness of each participant’s Q-sorts with the Q-sorts completed by the exemplars (77 correlations for each Q-sort). The Spearman’s correlations for Q-sort one addressing beliefs about discipline and behavior management ranged from −.27 to .81, with an average correlation of .34 (SD = .26). The Spearman’s correlations for Q-sort two reflecting beliefs about teaching practices ranged from −.60 to .82, with a mean of .42 (SD = .25). The Spearman’s correlations for Q-sort reflecting beliefs about children three ranged −.32 to .82, with a mean correlation of .47 (SD = .26). 5.4. Beliefs about discipline and behavior management Q-sort one measured educator beliefs about discipline and behavior management. The Respect for Children/Emphasis on SelfRegulation factor (QS1 factor 1) included items that reflected a respectful approach to supporting child self-regulation. Items that loaded highly onto this factor included “if I treat students with respect, kindness, and concern, there are less behavior problems” and “self-monitoring behaviors (or self-regulation) are important skills for students to develop.” This factor was significantly associated with the criterion sort z scores for sort one, meaning that it was correlated with the sort done by the exemplars, r (77) = .50, r2 = .25 p < .001, reflecting a large effect size. Latina educators scored significantly lower on this factor (M = −.12) compared to those from other ethnicities (M = .44), t (75) = 2.59, d = .64, p = .01, reflecting a moderate effect size. This factor was not significantly associated with education or position in either the full sample or the Latina subsample. Regression results revealed that ethnicity predicted this factor above and beyond education, but not vice versa (R = .26, R2 = .07, R2 = .07, p = .03), with Latina educators scoring lower on this factor (ˇ = −.29, p = .03). The Rule Focused/Teacher Directed factor (QS1 factor 2) reflected a focus on rules, a teacher-directed approach, and maintaining a quiet classroom. Items that loaded highly onto this factor included, “Rules for the students’ classroom behavior need to be reinforced consistently” and “Students learn best in primarily teacher-directed classrooms.” This factor was not significantly correlated with the criterion correlation z-scores for Q-sort one. Latina educators scored significantly higher on this factor (M = .18) compared to those from other ethnicities (M = −.49), t (75) = 2.67, d = .67, p = .008. Spearman’s correlations indicated that this factor was significantly negatively associated with position in the full sample (rs (77) = −.31, rs2 = .10, p = .019), but not the Latina subsample. Associations were not evident with education in either the full sample or in the Latina subsample. Regression analyses indicated that neither ethnicity nor education predicted this factor significantly above and beyond the other variable. The criteria correlation z-scores for Q-sort one were significantly lower for Latina teachers (M = −.16) compared to those from other ethnicities (M = .71), t (75) = 3.12, d = .96 (large effect size), p = .003, reflecting lower correlations between Latina educators and the exemplars. Spearman’s correlations indicated that the criteria correlation z-scores were significantly associated with education (rs (77) = .24, rs 2 = .06, p = .04), and position (rs (77) = .26, rs 2 = .07, p = .03) in the full sample. Associations were not evident between these criterion z-scores and education or position in the Latina subsample, in which there was less variability for both of these variables. Regression revealed that ethnicity (ˇ = −.29) predicted this variable above and beyond education, but not vice versa (R = .35, R2 = .12, R2 = .07, p = .02), with Latinas having lower criteria correlation scores. 5.5. Beliefs about teaching practices Q-sort two focused on assessing teacher beliefs about teaching practices. The Collaboration and Reflection valued factor (QS2 factor 1) was comprised of items emphasizing the use of reflection to promote learning, and recognition of individual strengths. Items that loaded highly onto this factor included, “having students share experiences” and “reflecting on the content of academic lessons and talking about what we learned.” This factor was significantly correlated with the criteria sort two z-scores, rs (77) = .38, rs2 = .14 (moderate to large effect size), p = .001. Scores for this factor were not significantly different across ethnic groups or the demographic variables examined. The Teacher Directed/Routine Focus factor (QS2 factor 2) reflected a focus on daily routines and teacher directed activities. Items that loaded positively onto this factor were “having a morning routine” and “using drill and recitation.” An item with a high negative loading for this factor was “welcoming each student by name to class.” This factor was significantly negatively correlated with criteria sort two z-scores (rs (77) = − .50, rs2 = .25 (large effect size), p < .001), such that high scores on this factor were associated with low criteria correlation scores. Scores for this factor were not significantly differences across ethnic groups or the demographic variables examined. Correlations for the criteria sort z-scores for Sort 2 were not significantly different for Latina educators compared to those from other ethnicities. Spearman’s correlations revealed that the criteria correlation z-scores were significantly associated with education, but not position, in both the full sample (rs (77) = .31, rs2 = .10, p = .008), and the Latina subsample (rs (63) = .26, rs2 = .07, p = .04). Regression analyses indicated that neither ethnicity nor education S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 predicted this criteria sort measure significantly above and beyond the other variable. 5.6. Beliefs about children Q-sort three focused on assessing teacher’s beliefs regarding children. The Mistrust of Children factor (Q31 factor 1) reflected a general mistrust regarding children’s behavior and their desire to learn. Items that loaded highly onto this factor included, “students in my class try to get away with doing as little work as possible” and “students show little desire to learn.” This factor was significantly negatively correlated with the criteria sort three z-scores (r (77) = −.77, rs2 = .59 (large effect size), p < .001), such that individuals with higher correlations with the exemplars’ sort expressed less mistrust of children. Latina educators scored significantly higher on this factor (M = .1) compared to those from other ethnicities (M = −.48), t (75) = 3.53, d = .64, p < .001. Spearman’s correlations indicated that this factor was significantly negatively correlated with position in both the full sample (rs (71) = −.27, rs2 = .07, p = .02), and the Latina subsample (rs (63) = −.27, rs2 = .07, p = .04). Age was significantly correlated with this factor in the Latina subsample (r (63) = .26, rs2 = 07, p = .04) according to Pearson’s correlations, such that older educators were more likely to report a general distrust of children. This factor was not associated with education in either the full sample or the Latina subsample. Regression analyses indicated that neither ethnicity nor education predicted this factor significantly above and beyond the other variable. The Focus on Choice/Individual Strengths factor (QS3, factor 2) was comprised of items emphasizing knowing individual students, recognizing their strengths, and giving them choices. Items that loaded highly on this factor were “students should feel as though they are known and recognized in the classroom” and “students need some choice of activities within the classroom”. This factor was significantly correlated with criteria sort two z-scores, r (77) = .59, rs2 = .35 (large effect size), p < .001. Latina educators scored significantly lower on this factor (M = −.14) than those from other ethnicities (M = .50), t (75) = 2.5, d = .59, p = .01. This factor was not significantly associated with education, position, or age in either the full sample, or the Latina subsample. Regression analyses indicated that neither ethnicity nor education predicted this factor significantly above and beyond the other variable. The criteria correlation z-scores for Q-sort three were significantly lower for Latina teachers (M = −.15) compared to those from other ethnicities (M = .71), t (75) = 2.96, d = 1.04 (large effect size), p = .004, reflecting lower correlation between Latina educators and the exemplars. Spearman’s correlations revealed that the criteria correlation z-scores were significantly associated with education (rs (77) = .32, rs2 = .10, p = .008), and position (rs (77) = .43, rs2 = .18, p < .001) in the full sample. The association between the criteria correlation scores and position was also evident in the Latina subsample (rs (63) = .31, rs2 = .10, p = .02). Regression analyses indicated that neither ethnicity nor education predicted this criteria sort measure significantly above and beyond the other variable. 6. Discussion The current study examined the teaching related beliefs of early childhood educators in a predominantly Latino, transborder community with a particular focus on elucidating the beliefs of Latina early childhood educators that may have their roots in cultural values. The use of a cultural communities lens required us to draw upon cultural theories of development and learning to explicate our findings as we sought to explore how the beliefs of early childhood educators may reflect community goals for children. Results 141 indicated disparities in education and position, such that Latina educators were less educated and in lower positions than were teachers from other ethnicities, despite making up the majority of the teaching staff. These findings reflect those found in Whitebook, Kipnis, and Bellm’s (2008) examination of the ECE workforce in our state, which indicated that the workforce is significantly stratified by both educational level and job title, with the higher the educational attainment of a group, the less ethnically and linguistically diverse it is. Low educational requirements, a lack of workforce diversity, and low levels of compensation are issues that plague the field of early childhood, and it is important to situate our findings within this larger context. Findings from the factor analytic method suggested patterns of beliefs that reflected both collectivist and individualistic orientations to teaching and learning. Latina educators scored significantly higher on factors that appeared consistent with a collectivist orientation and seemed to reflect the cultural value of raising children to be bien educado and prioritized statements that valued the behaviors of respect and obedience. Results also suggested greater disparities in beliefs that reflected best practice recommendations for early childhood educators who were less educated and those who were Latina. The individualistic orientation was reflected by statements prioritizing respect for children, the need to support children’s self-regulation, and a focus on recognizing children as individuals. Statements also valued encouraging children to have choices and the importance of collaboration and reflection to promote learning. These values are an integral part of schooling in the United States and are embodied by DAP and theories of constructivism that are central to notions of best practice in ECE (NAEYC, 2009) and these factors were highly and significantly associated with the beliefs of the ECE faculty exemplars. Latina educators scored significantly lower on two of the three factors representing the individualistic orientation than educators from other ethnicities, which is not surprising given that Latino culture tends to have values more characteristic of a collectivist cultural orientation. Factors that seemed to be consistent with the goals of a collectivist society included statements that positioned the teacher as the authority figure and highlighted the importance of rules and group norms. These factors also suggested beliefs consistent with the goal of raising children to be bien educado, as participants prioritized statements that reflected culturally valued behaviors of respect and obedience. The teacher directed, didactic approach highlighted suggests the values of obedience and respect, while a mistrust of children positions adults as being responsible for keeping children free from temptation and curbing their natural inclination to misbehave. This finding suggests that Latina teacher beliefs may reflect the collectivist orientation more often than educators from other ethnicities, and provides support for our contention that the beliefs of early childhood educators are often rooted in their own cultural community and mirror the larger goals and values of the community context. However, our cultural communities framework also impelled us to move beyond the simple binary of collectivist versus individualistic orientations toward teaching and learning. Indeed, Rogoff (2003) cautioned against the common practice of categorizing individuals or groups based on a single category such as race, ethnicity, language, or social class, and several of our findings point to the complex intersection of identities, beliefs, and practices that are endemic to transborder communities. For example, although Latina teachers held different beliefs about both discipline and children than teachers from other ethnicities and from the faculty exemplars, there were no significant differences in their beliefs about teaching practices. Findings from the CBC study (Adair et al., 2012; Tobin, Adair, & Arzubiaga, 2013) indicated that immigrant teachers reported difficulty applying their cultural knowledge to the care and education of immigrant children 142 S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 in their classrooms and felt ‘stuck’ between their cultural knowledge and pedagogical training. This conflict included negotiating the different expectations of parents and their fellow teachers, as well as their desire to be culturally responsive while still being perceived as professional (e.g., promoting accepted best practices in the field) by their peers and supervisors, sentiments also highlighted by Delpit (1986, 2006); Delpit, 1986. It is possible that the Latina educators in our study felt this same tension, as evidenced by their beliefs about discipline and children, which were very hierarchical and authoritative, and their beliefs about teaching practices, which tended to be more child centered. When interpreting our findings, it is important to consider the highly regulated nature of the Head Start program. The Head Start Performance Standards require that teaching practices must promote critical thinking and problem solving, provide supportive feedback for learning, support children’s engagement in learning experiences and activities, and be developmentally appropriate (Office of Head Start, 2016a, 2016b). In addition, The Creative Curriculum and the Classroom Assessment Scoring System (CLASS) (Pianta, La Paro, & Hamre, 2008), both of which are widely used in the Head Start program, also promote these practices. It is thus possible that the Latina teachers in our sample felt as if they had to renounce the beliefs about teaching practices they held prior to becoming professionals in order to support teaching practices promoted by the program in which they worked. This interpretation would explain the lack of significant associations found between ethnicity and beliefs regarding teaching practices. Our cultural communities lens also supports this tentative assertion. Rogoff (2003) proposed that, although community goals and cultural practices are passed down from generation to generation, they often evolve to meet changing circumstances, such as immigration or participation in higher education. It is thus possible that the beliefs about teaching practices of the Latina teachers in our study have changed because of participation in professional development opportunities and completing college courses. Interestingly, the factors reflecting beliefs regarding teaching practices were not significantly associated with education level in either the full sample or the Latina subsample, and although no definitive conclusions can be drawn from these findings, they do point to the complexity of teacher beliefs. Findings using the criterion method to investigate associations between the beliefs of ECE faculty exemplars and practitioners indicated a clear and consistent association between teacher beliefs associated with best practice recommendations and education level, with significant associations for all three criteria sorts. This association was strong and more consistent in the full sample than in the Latina subsample in which there was less variability in education. In the full sample, participants in a higher position prioritized beliefs significantly more aligned with best practice recommendations about discipline and children than those in lower positions, although it is important to remember that there were fewer Latinas in higher positions. Although the majority of these findings were for the full sample, the findings for the Latina subsample provide additional support to our assertion that as Latina teachers obtained more education and other training, they adopted beliefs that were more reflective of a the child-centered, constructivist approach to teaching and learning espoused by DAP, the Head Start Performance Standards, and the Creative Curriculum. However, the criterion method sorts examining the associations between the beliefs of ECE experts and ECE practitioners were also significantly associated with ethnicity for two of the three sorts (Beliefs about Discipline and Behavior Management and Beliefs about Children), with Latina educators being less likely to espouse beliefs prioritized by the exemplars. This suggests that although education may be helping all educators align their beliefs with best practice recommendations, ECE faculty need to carefully consider how beliefs developed via participation in various cultural communities may influence the willingness of early childhood educators to adopt best practice recommendations presented during coursework and professional development opportunities. 6.1. Limitations It is critical to restate the exploratory nature of this study and note that because factor analysis relies on a posteriori interpretation by the researchers, results cannot be considered definitive or to be actual measures of the constructs under investigation. The fact that TBQ was not designed to measure individualist or collectivist cultural orientations or beliefs related to socialization goals for children is a limitation of this study that must be considered when interpreting our results. In addition, because we did not collect data on actual classroom practices, we cannot determine if the beliefs espoused by our participants were actually translated into classroom practice. Although some researchers have found an association between teacher beliefs and practices (e.g., McMullen et al., 2005, 2006; Oakes & Caruso, 1990), other studies have found only weak associations or no associations (e.g., Tiano & McNeil, 2006; Wilcox-Herzog, 2002). Thus, further research is needed regarding the link between teacher beliefs and practices, and is particularly needed to examine associations between the factors identified in the current study and teacher practices. The use of qualitative, ethnographic interviewing techniques may also provide additional insight into how community goals for children influence the beliefs of early childhood educators living in distinct cultural communities such as the transborder region examined in the current study. Additionally, our study was exploratory in nature and included a large number of statistical analyses, and thus should be replicated in future research. Our study was also limited in statistical power for detecting significant findings. Although this sample reflected early childhood education staff across 20 Head Start centers, the sample consisted of only 77 participants, and the sub-sample of non-Latina participants was particularly small (15 participants). It is also noteworthy that the first two factors in each factor analysis explained only 23% to 27% of the variance from each Q-sort, and that the internal reliability for the factor scores were only moderate. Thus, although findings comparing Latina and non-Latina participants are provocative in terms of better understanding Latina early childhood educators, they should be treated with caution and replicated in future research. Research with larger sample sizes, across multiple Head Start agencies, and across several states with transborder communities would be ideal for replication and further examination of the beliefs of early childhood educators. Finally, there was a strong association between ethnicity and education in the current study, which limited our ability to fully disentangle the influence of these variables. The majority of our analyses indicated that neither ethnicity nor education predicated significantly above and beyond the other variable, which indicates shared predictive variance of these two variables. Further research is recommended with a more even distribution of education across ethnic groups (over-sampling highly educated Latino early childhood educators and less-educated early childhood educators from other ethnic groups) to better disentangle the unique predictive ability of ethnicity and education. 6.2. Implications for practice Our primary goal in conducting this exploratory study was to learn more about the beliefs about discipline, teaching, and children of Head Start educators living in a transborder community known for its sociocultural complexity and large Latino population. Using a theoretical framework that foregrounds the saliency of social, political, and economic contexts on the development of S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 beliefs and practices, our findings suggest a need for early childhood researchers and educators to consider the ways in which transborder communities represent unique contexts for teaching and learning in the early years. Although the exploratory nature of our study precludes the drawing of definitive conclusions, our findings do suggest the need to better reconcile recommendations for DAP and other best practice recommendations with the cultural values of early childhood educators and the communities in which early childhood programs are located. This may be easier said than done, however, as the care and education of young children occurs in diverse settings in which practitioners have different traditions and cultures (IOM & NRC, 2015). For the past 25 years, NAEYC has been both thoughtful and intentional in its message that programs should be both developmentally and culturally appropriate. DAP guidelines urge programs to consider “the social and cultural contexts in which children live —referring to the values, expectations, and behavioral and linguistic conventions that shape children’s lives at home and in their communities (NAEYC, 2009, p. 10). Tobin et al. (2013) have noted that although both NAEYC and Head Start have clear guidelines and policies that stress the importance of preparing teachers to understand and be responsive to the needs of all children, the challenge comes with implementation of these guidelines and policies. Our findings support this assertion, and point to the need for additional research and discourse that takes into consideration the influence of unique contexts on the beliefs and priorities of early childhood educators. As ECE faculty responsible for preparing the workforce, we must grapple with the fact that our beliefs about teaching and learning may oftentimes be at odds with those of our students and those of educators working in our local community. As noted earlier, findings indicating disparities in education and position related to ethnicity reflect larger issues in the field. In our sample, the majority of teachers and assistant teachers were Latina, reflecting the predominant ethnicity of the children served in the Head Start program. However, the early childhood educators who had the highest levels of education and who were in supervisory positions were predominantly from other ethnic groups. Thus, it is possible that there is incongruence in belief systems between the more educated supervisors and the less educated Latina practitioners who are actively teaching the children. Although the Head Start agency met policy recommendations that staff represent the local community and the children and families served, this appears to have been true for teachers and assistant teachers, and may not be the case for ECE staff in supervisory positions. It may not be enough to simply hire ECE staff who represent the community and expect them to expertly balance both their cultural values and the nuances of practices promoted by the field and concretized by DAP, CLASS, and the Head Start Performance Standards, especially in transborder communities characterized by not only tension and conflict, but also by possibilities. Acknowledgement The authors would like to acknowledge the San Diego State University University Grants Program for funding of this project. References Adair, J. K., Tobin, J., & Arzubiaga, A. E. (2012). The dilemma of cultural responsiveness and professionalization: Listening closer to immigrant teachers who teach children of recent immigrants. Teachers College Record, 114(12) http://dx.doi.org/10.1080/10901027.2010.547652 Block, J. (1961). The Q-Sort method in personality assessment and psychiatric research. Springfield, IL: Charles C. Thomas. Bryant, D. M., Clifford, R. M., & Peisner, E. S. (1991). Best practices for beginners: Developmental appropriateness in kindergarten. American Educational Research Journal, 28(4), 783–803. http://dx.doi.org/10.3102/ 00028312028004783 143 Bronfenbrenner, U., & Morris, P. A. (1998). The ecology of developmental processes. In W. Damon, & R. M. Lerner (Eds.), Handbook of child psychology: Volume I. Theoretical models of human development (pp. 993–1028). New York, NY: John Wiley & Sons. Buchanan, T. K., Burts, D. C., Bidner, J., White, F., & Charlesworth, R. (1998). Predictors of the developmentally appropriateness of the beliefs and practices of first, second, and third grade teachers. Early Childhood Research Quarterly, 13, 459–483. http://dx.doi.org/10.1016/s0885-2006(99)80052-0 Calderhead, J. (1996). Teachers, beliefs, and knowledge. In D. C. Berliner, & R. C. Calfee (Eds.), Handbook of educational psychology (pp. 709–725). New York: Simon and Schuster. Cauce, A., & Domenech-Rodríguez, M. (2002). Latino families: Myths and realities. In J. Contreras, K. Kerns, & A. Neal-Barnett (Eds.), Latino children and families in the United States (pp. 3–25). Westport, CT: Praeger. Charlesworth, R., Hart, C. H., Burts, D., & Hernandez, S. (1991). Kindergarten teachers beliefs and practices. Early Child Development and Care, 70, 17–35. http://dx.doi.org/10.1080/0300443910700103 Charlesworth, R., Hart, C. H., Burts, D., Thomasson, R. H., Mosley, J., & Fleege, P. O. (1993). Measuring the developmental appropriateness of kindergarten teachers’ beliefs and practices. Early Childhood Research Quarterly, 8, 255–276. http://dx.doi.org/10.1016/s0885-2006(05)80067-5 Cline, Z., & Necochea, J. (2003). Education in the borderlands: A border pedagogy conceptual model. El Bordo, 11, 43–57. http://dx.doi.org/10.1080/08865655. 2005.9695638 Cline, Z., & Necochea, J. (2006). Teacher dispositions for effective education in the Borderlands. The Educational Forum, 70(3), 268–282. http://dx.doi.org/10.1080/ 00131720608984902 Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10(7), 1–9. http://dx.doi.org/10.4135/ 9781412995627.d8 Decker, L. E., & Rimm-Kaufman, S. E. (2008). Personality characteristics and teacher beliefs among pre-service teachers. Teacher Education Quarterly, 35(2), 45–64. Delgado-Gaitan, C. (1994). Parenting in two generations of Mexican American families. International Journal of Behavioral Development, 16, 409–427. http:// dx.doi.org/10.1177/016502549301600303 Delpit, L. (1986, 2006). Other people’s children: Cultural conflict in the classroom. The New Press. DiStefano, C., Zhu, M., & Mindrila, D. (2009). Understanding and using factor scores: Considerations for the applied researcher. Practical Assessment, Research & Evaluation, 14(20), 1–11. Fang, Z. (1996). A review of research on teacher beliefs and practices. Educational Research, 38(1), 47–65. http://dx.doi.org/10.1080/0013188960380104 File, N. (1994). Children’s play, teacher-child interactions, and teacher beliefs in integrated early childhood programs. Early Childhood Research Quarterly, 9, 223–240. http://dx.doi.org/10.1016/0885-2006(94)90007-8 File, N., & Gullo, D. F. (2002). A comparison of early childhood and elementary education students’ beliefs about primary teaching practices. Early Childhood Research Quarterly, 17(1), 126–137. http://dx.doi.org/10.1016/s08852006(02)00130-8 Flores, Juan. (2000). From bomba to hip-hop: Puerto Rican culture and Latino identity. Columbia University Press. García Coll, C. T., & Magnuson, K. (2000). Cultural differences as sources of developmental vulnerabilities and resources: A view from developmental research. In S. J. Meisels, & J. P. Shonkoff (Eds.), Handbook of early childhood intervention (pp. 94–111). Cambridge, UK: Cambridge University Press. Garrity, S., & Wishard Guerra, A. (2015). A cultural communities approach to understanding Head Start teachers’ beliefs about language use with dual language learners: Implications for practice. Contemporary Issues in Early Childhood, 16(3), 1–16. http://dx.doi.org/10.1177/1463949115600027 Giroux, H. A. (1988). Schooling and the struggle for public life: Critical pedagogy in the modern age. Minneapolis: University of Minnesota Press. Goldenberg, C., & Gallimore, R. (1995). Immigrant Latino parents’ values and beliefs about their children’s education: Continuities and discontinuities across cultures and generations. In P. R. Pintrich, & M. Maehr (Eds.), Advances in motivation and achievement: Culture, ethnicity, and motivation (pp. 183–228). Greenwich, CT: JAI Press. Greenfield, P. M., & Bruner, J. S. (1966). Culture and cognitive growth. International Journal of Psychology, 1, 89–107. http://dx.doi.org/10.1080/ 00207596608247117 Halgunseth, L., Ispa, J., & Rudy, D. (2006). Parental control in Latino families: An integrated review of the literature. Child Development, 77, 1282–1297. http:// dx.doi.org/10.1111/j.1467-8624.2006.00934.x Harwood, R. L., Leyendecker, B., Carlson, V., Asencio, M., & Miller, A. (2002). Parenting among Latino families in the U.S. In M. H. Bornstein (Ed.), Handbook of parenting: Social conditions and applied parenting (vol. 4) (2nd ed., vol. 4, pp. 21–46). Mahwah, NJ: Erlbaum. Hughes, D., Rodriguez, J., Smith, E. P., Johnson, D. J., Stevenson, H. C., & Spicer, P. (2006). Parents’ ethnic-racial socialization practices: A review of research and directions for future study. Developmental Psychology, 42(5), 747. http://dx.doi. org/10.1037/0012-1649.42.5.747 Institute of Medicine, & National Research Council. (2015). Transforming the workforce for children birth through age 8: A unifying foundation. Washington, DC: National Academies Press. http://dx.doi.org/10.13184/eidon.39.2013.89-91 144 S. Garrity et al. / Early Childhood Research Quarterly 47 (2019) 134–144 Kagan, D. M. (1992). Implications of research on teacher belief. Educational Psychologist, 27(1), 65–90. http://dx.doi.org/10.1207/s15326985ep2701 6 Kontos, S., & Dunn, L. (1993). Caregiver practices and beliefs in child care varying in developmental appropriateness and quality. Advances in Early Education and Day Care, 5, 53–74. Marin, G., & Marin, B. V. (1991). Research with Hispanic populations. Los Angeles, CA: Sage Publications, Inc. McDonald, R. (1999). Test theory: A unified treatment. Mahwah, NJ: Lawrence Erlbaum Associates. McGill-Franzen, A., Lanford, C., & Adams, E. (2002). Learning to be literate: A comparison of five urban early childhood programs. Journal of Educational Psychology, 94(3), 443. McMullen, M. B., & Alat, K. (2002). Education matters in the nurturing of the beliefs of preschool caregivers and teachers. Early Childhood Research and Practice, 4(2), 1–16. McMullen, M. B., Elicker, J., Goetze, G., Huang, H., Lee, S., Mathers, C., . . . & Yang, H. (2006). Using collaborative assessment to examine the relationship between self-reported beliefs and the documentable practices of preschool teachers. Early Childhood Education Journal, 34, 81–91. http://dx.doi.org/10.1007/ s10643-006-0081-3 McMullen, M., Elicker, J., Wang, J., Erdiller, Z., Lee, S. M., Lin, C. H., & Sun, P. Y. (2005). Comparing beliefs about appropriate practice among early childhood education and care professionals from the US, China, Taiwan, Korea and Turkey. Early Childhood Research Quarterly, 20(4), 451–464. http://dx.doi.org/ 10.1016/j.ecresq.2005.10.005 National Association for the Education of Young Children. (2009). Developmentally appropriate practice in early childhood programs serving children from birth through age 8.. Retrieved on May 5, 2017 from https://www.naeyc.org/files/ naeyc/file/positions/position%20statement%20Web.pdf Necochea, J., & Cline, Z. (2005). Borderland education and teacher education reform in California: Unfulfilled promises. Journal of Borderlands Studies, 20(1), 129–141. http://dx.doi.org/10.1080/08865655.2005.9695638 Nespor, J. (1987). The role of beliefs in the practice of teaching. Journal of Curriculum Studies, 19(4), 317–328. http://dx.doi.org/10.1080/0022027870190403 Oakes, P. B., & Caruso, D. A. (1990). Kindergarten teachers’ use of developmentally appropriate practices and attitudes about authority. Early Education and Development, 1(6), 445–457. http://dx.doi.org/10.1207/s15566935eed0106 4 Office of Head Start. (2016a). Head Start program facts. https://eclkc.ohs.acf.hhs. gov/sites/default/files/pdf/hs-program-fact-sheet-2016.pdf Office of Head Start. (2016b). Head Start program performance standards and other regulations. Washington DC: Office of Head Start, Administration for Children and Families, U.S. Department of Health and Human Services. Retrieved on August 9, 2017 from https://eclkc.ohs.acf.hhs.gov/policy/45-cfr-chap-xiii LaParo, K. M., Siepak, K., & Scott-Little, C. (2009). Assessing beliefs of preserve early childhood education teachers using Q-Sort methodology. Journal of Early Childhood Teacher Education, 30(1), 22–36. http://dx.doi.org/10.1080/ 10901020802667805 Pajares, M. F. (1992). Teachers’ beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62(3), 307–332. http://dx.doi. org/10.2307/1170741 Pianta, R. C., La Paro, K. M., & Hamre, B. K. (2008). Classroom Assessment Scoring System (CLASS). Baltimore: Paul H. Brookes. Richardson, V. (1994). The consideration of teachers’ beliefs. In V. Richardson (Ed.), Teacher change and the staff development process. New York: Teachers College. Rimm-Kaufman, S. E., & Sawyer, B. E. (2004). Primary-grade teachers’ self-efficacy beliefs, attitudes toward teaching, and discipline and teaching practice priorities in relation to the responsive classroom approach. Elementary School Journal, 104, 321–341. http://dx.doi.org/10.1086/499756 Rimm-Kaufman, S. E., Storm, M. D., Sawyer, B. E., Pianta, R. C., & LaParo, K. M. (2006). The Teacher Belief Q-Sort: A measure of teachers’ priorities in relation to disciplinary practices, teaching practices, and beliefs about children. Journal of School Psychology, 44(2), 141–165. http://dx.doi.org/10.1016/j.jsp.2006.01. 003 Rogoff, B. (2003). The cultural nature of human development. Oxford University Press. Romo, J. J., & Chavez, C. (2006). Border pedagogy: A study of preservice teacher transformation. The Educational Forum, 70(June (2)), 142–153. http://dx.doi. org/10.1080/00131720608984885. Taylor & Francis Group Rothstein-Fisch, C., Greenfield, P. M., Trumbull, E., Keller, H., & Quiroz, B. (2010). Uncovering the role of culture in learning, development, and education. In D. D. Preiss, & R. J. Sternberg (Eds.), Innovations in educational psychology: Perspectives on learning, teaching, and human development (pp. 269–294). New York: Springer. Sanders, K. E., Deihl, A., & Kyler, A. (2007). DAP in the ’hood: Perceptions of child care practices by African American child care directors caring for children of color. Early Childhood Research Quarterly, 22(3), 394–406. http://dx.doi.org/10. 1016/j.ecresq.2007.03.002 Smith, K. E. (1997). Student teachers’ beliefs about developmentally appropriate practice: Pattern, stability, and the influence of locus of control. Early Childhood Research Quarterly, 12(2), 221–243. http://dx.doi.org/10.1016/s08852006(97)90015-6 Spidell-Rusher, A. S., McGrevin, C. Z., & Lambiotte, J. G. (1992). Belief systems of early childhood teachers and their principals regarding early childhood education. Early Childhood Research Quarterly, 7(2), 277–296. http://dx.doi.org/ 10.1016/0885-2006(92)90009-n Stephenson, W. (1935). Technique of factor analysis. Nature, 136, 297. http://dx. doi.org/10.1038/136297b0 Tabachnick, B. G., & Fidell, L. S. (2001). Principal components and factor analysis. Using Multivariate Statistics, 4, 582–633. Tiano, J. D., & McNeil, C. B. (2006). Training Head Start teachers in behavior management using Parent–Child Interaction Therapy: A preliminary investigation. Journal of Early and Intensive Behavior Intervention, 3(2), 220. http://dx.doi.org/10.1037/h0100334 Tobin, J., Adair, J. K., & Arzubiaga, A. (2013). Children crossing borders: Immigrant parent and teacher perspectives on preschool for children of immigrants. Russell Sage Foundation. Trizano-Hermosilla, I., & Alvarado, J. M. (2016). Best alternatives to Cronbach’s alpha reliability in realistic conditions: Congeneric and asymmetrical measurements. Frontiers in Psychology, 7, 769. http://dx.doi.org/10.3389/fpsyg. 2016.00769 U.S. Census Bureau. (2001). 2000 census of population: General population characteristics. Washington, DC: U.S. Department of Commerce, Economics and Statistics Administration, Bureau of the Census. U.S. Census Bureau. (2014). Quick Facts: Chula Vista, California.. Retrieved on July 14, 2017 from https://www.census.gov/quickfacts/fact/table/ chulavistacitycalifornia/AGE765210 Valdés, G. (1996). Con Respeto. New York: Teachers College Press. Valenzuela, A. (1999). Subtractive schooling: U.S.–Mexican youth and the politics of caring. In SUNY series, the social context of education. New York, NY: State University of New York Press. Vartuli, S. (1999). How early childhood teacher beliefs vary across grade level. Early Childhood Research Quarterly, 14, 489–514. http://dx.doi.org/10.1016/s08852006(99)00026-5 Wang, J., Elicker, J., McMullen, M., & Mao, S. (2008). Chinese and American preschool teachers’ beliefs about early childhood curriculum. Early Child Development and Care, 178(3), 227–249. http://dx.doi.org/10.1080/ 03004430600722671 Wen, X., Elicker, J. G., & McMullen, M. B. (2011). Early childhood teachers’ curriculum beliefs: Are they consistent with observed classroom practices? Early Education & Developmen, 22(6), 945–969. http://dx.doi.org/10.1080/ 10409289.2010.507495 Whitebook, M., Kipnis, F., & Bellm, D. (2008). Diversity and stratification in California’s early care and education workforce. Berkeley, CA: Center for the Study of Child Care Employment, University of California, Berkeley. Retrieved from http://www.irle.berkeley.edu/cscce/wp-content/uploads/2008/01/ diversity stratification08.pdf Whiting, B. B., & Whiting, J. W. M. (1975). Children of six cultures. Cambridge, MA: Harvard University Press. Wilcox-Herzog, A. (2002). Is there a link between teachers’ beliefs and behaviors? Early Education & Development, 13, 81–106. http://dx.doi.org/10.1207/ s15566935eed1301 5 Wishard, A., Shivers, E., Howes, C., & Ritchie, S. (2003). Child care program and teacher practices: Associations with quality and children’s experiences. Early Childhood Research Quarterly, 18, 65–103. http://dx.doi.org/10.1016/s08852006(03)00007-3