Faculty of Social and Behavioural Sciences
Graduate School of Child Development and Education
Effect of a Funds of Identity Based Pedagogical Approach on
Cohesion in Multicultural Classrooms
Research Master Child Development and Education Research Master Thesis
Name of the student: Radha Vathsal Student Id: 12191906
Name(s) of the supervisor(s): Prof. Dr. Monique Volman and Dr. Bonne Zijlstra Date of submission: 24-Dec-2020
Abstract
In the present age of migration, schools are undoubtedly an ideal setting to promote social cohesion as they can facilitate respectful interaction between students from diverse
backgrounds, without bias and prejudice. A funds of identity (FoI) based pedagogy adopts a socio-cultural conceptualization of teaching and learning and advocates using what students already know and do outside of school as key resources to design the curriculum. In this study, we examined the effect of a FoI based pedagogy on cohesion amongst students in primary school classrooms. We also examined if the effect of such a pedagogy was different between and within majority and minority group students. The study followed a quasi-experimental design, and round-robin network data was collected at two time points from eight primary multicultural schools in Amsterdam (N = 301 students). We used the social relations model (SRM) to test our hypotheses and analyzed liking at the dyad level to
examine cohesion at the classroom level. We did not observe any effect of the intervention on cohesion amongst students. Using an alternative measure to examine cohesion is
recommended for future research. This study serves as an example for future quantitative studies in this emerging field of research. The study also presents network analysis as an appropriate method to examine the social context within classrooms.
Keywords: funds of knowledge, funds of identity, culturally responsive pedagogy,
cohesion, social relations model
Effect of a Funds of Identity Based Pedagogical Approach on Cohesion in Multicultural Classrooms
Our communities are increasingly becoming diverse and multicultural due to globalization, enhanced geographic mobility, and migration for economic, social, political and environmental reasons (Inglis, 2008). As a result, there has been a sharp rise in the population of students from different ethnic backgrounds in schools and classrooms (Forghani-Arani, Cerna, & Bannon, 2019; Inglis, 2008). In this growing reality of
multiculturalism, schools are increasingly being considered as places to promote a sense of social cohesion in society (Dijkstra, De la Motte, & Eilard, 2014; Radstake & Leeman, 2010). Schools are viewed as an ideal setting that facilitate interaction between children from
different backgrounds and develop interpersonal social skills in children that can yield social outcomes (Dijkstra et al., 2014; Radstake & Leeman, 2010). Therefore, the work in schools to develop a sense of cohesion within classrooms becomes critical in the current times than ever before (Baskerville, 2011; Forghani-Arani, Cerna, & Bannon, 2019).
Research suggests that embracing diversity in the classrooms while also promoting cohesion amongst students can be challenging for teachers as many teachers lack the
necessary skills and knowledge such as intercultural sensitivity, reflectivity about identities, and an ability to build a relationship of trust and proximity with their students (Forghani-Arani et al., 2019; Leeman, 2006; Radstake & Leeman, 2010). There is a need for teachers to broaden their thinking and adopt a culturally responsive pedagogy to better understand their pupils (Baskerville, 2011; Aronson & Laughter, 2016; Sleeter, 2012). In this study, we present a funds of identity based pedagogical approach (Esteban-Guitart & Moll, 2014a), as a promising means to use diversity in a positive way and promote cohesion amongst students in classrooms.
Funds of identity being a recent concept, empirical studies are limited. Existing studies suggest many positive outcomes of such a pedagogical approach such as improving student learning (Subero, Vujasinović, & Esteban-Guitart, 2017), student motivation (Subero, Llopart, Siqués, & Esteban-Guitart, 2018) and creating a positive classroom climate (Hviid & Villadsen, 2014). However, to our knowledge, there are not many studies that have
empirically tested the influence on cohesion amongst students in classrooms. Therefore, in the present study, we attempt to extend this area of research and investigate the effect of a funds of identity based pedagogical approach on cohesion amongst students in classrooms. Theoretical Framework
Funds of Identity: Definition
The term funds of identity was first introduced by Esteban-Guitart (2012) and later elaborated by Esteban-Guitart and Moll (2014a). It is based on the concept of funds of
knowledge which refers to a set of knowledge, skills and information unique to every
household due to the varied experiences of people within the household. These diverse experiences include their upbringing, social interactions and participation at home,
community and different types of work in the job market. Thereupon, funds of knowledge become one’s funds of identity when these resources are actively used as a box of tools to define oneself (Esteban-Guitart & Moll, 2014a). Funds of identity is therefore defined as, “historically accumulated, culturally developed and socially distributed resources which are essential for a person’s self-definition, self-expression, and self-understanding” (Esteban-Guitart & Moll, 2014a, p. 37). While funds of knowledge are shared collectively by all individuals in a household, funds of identity are unique to individuals as every individual interacts and internalizes the bodies of knowledge available to him or her differently in constructing one’s own identity (Esteban-Guitart, 2012).
Funds of identity is also closely associated with the concept of “perezhivanie”,
presented by Vygotsky which can be translated as lived experiences (Esteban-Guitart & Moll, 2014b). When viewed through the lens of “perezhivanie”, funds of identity refers to the life experiences that individuals accumulate and use to define themselves (Subero et al., 2017). To put it differently, every individual understands any given situation differently and behaves according to their lived experiences. While doing so, individuals create and recreate their identities in interactions which become the lived experiences of oneself (Esteban-Guitart & Moll, 2014a). Thus, funds of identity can be understood to be temporal and contextual, that is, there is a core identity and many sub-identities that are mobilized, adapted and changed according to the external circumstances and myriad interactions with the environment (Poole & Huang, 2018).
There can be many types of funds of identity for every individual. Esteban-Guitart and Moll (2014a) categorized five types of funds of identity namely geographical (e.g., one’s hometown, a specific mountain), practical (e.g., a particular sport, hobby activity), cultural (e.g., festivals, national flag, a specific type of music), social (e.g., important people such as parents, close friends) and institutional (e.g., office, places of worship). Additionally, there can be digital funds of identity derived from the digital technologies that surround us today (Esteban-Guitart & Moll, 2014b; Poole, 2017a). Finally, Poole and Huang (2018) put forward an additional category called the existential funds of identity, which refers to the positive and negative life experiences individuals use to define themselves.
Funds of Identity: Pedagogical Approach
Funds of identity when used as a pedagogical approach, adopts a socio-cultural conceptualization to teaching and learning (Esteban-Guitart & Moll, 2014b). Inherent to this approach is the fundamental understanding that students learn in interaction with their environment mediated by students’ lived experiences and therefore, different students
internalize their environments differently. With this understanding, students’ funds of identity can be used as key pedagogical resources to design the curriculum that facilitates links
between academic learning, and what students already know and do outside of school (Subero et al., 2017). By forging links between the different spheres of learning, learning is promoted, and student identities are expanded and transformed (Esteban-Guitart & Moll, 2014b).
There are many ways to detect and use students’ funds of identity in curricular activities. These include specific arts-based methods such as self-portrait and significant
circle, and a variety of identity artefacts (Esteban-Guitart & Moll, 2014a; Jovés, Siqués, &
Esteban-Guitart, 2015; Subero et al., 2017). A self-portrait is a drawing made by students representing their state or who they are at the time of the drawing (Esteban-Guitart & Moll, 2014a). It is also referred to as an identity drawing and encourages students to include significant others, activities and things that are most important at the time of the drawing (Subero et al., 2017). Along the same lines, is the technique called significant circle or a
meaningful circle where students are asked to place objects, activities, people and institutions
that are an important part of students’ life at the time of the drawing within a circle (Esteban-Guitart & Moll, 2014a). It is also referred as a relational map and depending on where the students place the objects in the circle, their relative importance is interpreted; for example, items near the center of the circle are perceived as most important or significant (Esteban-Guitart & Moll, 2014a).
Identity artefacts are also an interesting resource to uncover students’ funds of
identity. An identity artefact can be written, spoken, musical, visual form or a combination of these modes on a specific or a broad academic topic that is relevant and meaningful to the students (e.g., a text on family migration history, a photo collage) (Subero et al., 2018, 2017). As an illustration, Poole (2017b) asked students to create a graphical representation of who
they were at that point in time in the form of an avatar and a word cloud and asked students to explain their creations in a written identity text. The author, who was also their teacher, reported that he obtained valuable insights about his students’ inner worlds which helped break down deficit thinking about his students. Usually, these methods are followed by a discussion with the teacher where the students are asked to explain their drawing (Esteban-Guitart & Moll, 2014a; Jovés, Siqués, & Esteban-(Esteban-Guitart, 2015; Subero et al., 2017).
Students’ funds of identity that emerge from these methods and discussions are then used to design learning activities in the classroom. For example, Saubich and Esteban-Guitart (2011) identified students’ interest and experience with ‘henna’, a plant-based dye used on hair, body, hand, nails, wool and leather and incorporated this information into the arts curriculum. Likewise, in another school’s context, they used students’ collective interest in soccer to develop geometry and math learning units. Similarly, Esteban-Guitart & Moll (2014a) uncovered taking care of animals as a student’s funds of identity from the significant circle exercise and discussions with the student and designed a learning unit in science dealing with farm animals. Hughes and Pollard (2006) developed a ‘shoebox project’ where students were given an empty shoebox and asked to transform the box into an ‘about me’ box by decorating and filling it with items that are special to the children such as toys and photos. The teachers discussed the reasoning behind the contents in the box with the students. They used the contents to build vocabulary, grammar, design writing lessons and math activities in the classroom. Evidently, there can be many appealing ways to discover students’ funds of identity, which can be seamlessly integrated into teaching and learning activities (see also Hogg & Volman, 2020).
Funds of identity based pedagogy is closely associated with the educational
philosophy of contextualized learning (Tharp, 1997) and connected learning (Ito et al., 2013) which emphasize connecting school to students’ lives by using “spontaneous concepts” from
students’ everyday life and activities to teach “scientific concepts” (Esteban-Guitart & Moll, 2014b). When learning is designed around common interests, shared values and practices amongst peers in the classroom, inevitably it becomes meaningful and inclusive (Cummins, 2007). Besides, funds of identity based pedagogy also encourages students to create
knowledge and conceptual artefacts collaboratively by investing their identities and negotiating the meaning of the academic content presented at school (Subero et al., 2017).
Additionally, the conception of funds of identity based teaching is in line with the notion of culturally responsive pedagogy, also referred to as culturally-relevant teaching elaborated by Ladson-Billings (1995).She explained culturally responsive pedagogy as a way of teaching that allowed students to maintain their cultural integrity and develop cultural competence while succeeding academically. She also stated that such a pedagogy must develop “critical perspectives that challenge inequities that schools (and other institutions) perpetuate” (p. 469). She described teaching as a process that is close to the Freirean notion of “teaching as mining or pulling knowledge out” (p. 479); knowledge as “shared, recycled and constructed” (p. 481) and not statically transmitted in the classroom; and learning as a collaborative process which improves social relationships amongst peers and between students and the teacher. Such a pedagogy is especially significant in today’s diverse classrooms as it has the potential to break down deficit thinking, bring about a fundamental shift in the teaching-learning processes in the classroom and expand teacher identities (Ladson-Billings, 1995). Some other terms for such a pedagogy that connects with children by responding to their cultural identities are culturally compatible (Jacob & Jordan, 1987),
culturally congruent (Au & Kawakami, 1994) and most recently culturally-sustaining pedagogy (Ladson-Billings, 2014).
Funds of identity based pedagogy is a recent development and research in this field is largely theoretical or qualitative; It is mainly focused on identifying and incorporating
students’ funds of identity in pedagogical practice (Hogg & Volman, 2020). Empirical studies on the educational outcomes of funds of identity based pedagogical approach are scarce. However, there is considerable discussion on the implications of such an approach at an individual (student) level and a collective (classroom/school) level.
On an individual level, Subero et al. (2017) posit that by encouraging students to create knowledge, funds of identity based teaching promotes learning and academic outcomes. Additionally, Subero et al., (2018) point out that connecting students’ funds of identity with educational practice could favor “affective-cognitive readiness” (p.168) and motivate students to engage in learning at school. Finally, recognizing that students’
identities form their lens through which they acquire knowledge and newer identities, Hviid and Villadsen (2014) theorize that using students’ funds of identity provides opportunities to engage students in a collective and critical process of meaning-making that fosters students’ identity development. On a collective level, Subero et al. (2017) point out that such a pedagogy can create a stimulating classroom environment that is learner-centered.
Furthermore, reflecting on the potentials of such a pedagogy for school practice, Hviid and Villadsen (2014) put forward that classrooms can become a cultural community with a “shared culture background” (p. 67) amongst students making schools culturally fair for all.
Undoubtedly, funds of identity based pedagogical approach is theoretically well supported and empirically promising. Against this background of research, in this study, we specifically investigate the influence of such a pedagogy on a collective level (i.e., cohesion amongst students in the classroom). However, at this point, it is essential to acknowledge that peer relationships shape the group processes of a classroom and any effect that manifests itself at a collective level must stem from interactions and relationships at an individual level (Schmuck & Schmuck, 1975; Friedkin, 2004; Rubin, Bukowski, & Parker, 2007).
Students in a classroom relate to one another both formally and informally. Formal aspects include working together in a group for a classroom assignment or relating to one another with respect to learning and academic outcomes. Informal aspects refer to how classmates relate to one another in terms of liking or friendship. Cohesion in classrooms results from these formal and informal aspects between students which also mediate the teaching-learning processes in the classroom (Schmuck & Schmuck, 1975). There is no clear definition of the term cohesion used in the classroom context (see also Volman & ’t Gilde, 2021). Generally, cohesion is viewed as a sense of connectedness with(in) a group. Moody and White (2003) defined cohesion as a measure of how well a group is held together;
Dijkstra et al., (2014) presented cohesion as two sides of a coin: “keeping things together and allowing room for variation”; and Lott and Lott (1965) referred to cohesion as the positive interpersonal ties between members in a group. Drawing from these definitions, in this study, we define classroom cohesion as a measure of connectedness amongst students within a classroom characterized by positive interpersonal ties that students have with other members of their classroom group.
In today’s diverse and multicultural communities, building cohesion amongst students in classrooms is necessary to promote social competence in individual students, social
cohesion and social capital in the society (Dijkstra et al., 2014). Stark, Mäs, and Flache (2015) have found that feelings of pleasantness or unpleasantness towards out-group students positively or negatively influence students’ general outlook towards these groups. Besides, there is considerable research that substantiates that peer liking is important for students’ emotional well-being (Ladd, 2006), academic adjustment (Roseth, Johnson, & Johnson, 2008; Wentzel, Caldwell, Wentzel, & Caldwell, 1997), motivation and academic competence (Flook, Repetti, & Ullman, 2005). Furthermore, peer disliking is associated with internalizing and externalizing behavior (Ladd & Troop-Gordon, 2003), declining classroom participation
and increasing school avoidance (Buhs, Ladd, & Herald, 2006), school dropout (French & Conrad, 2001) and bullying (Roland & Galloway, 2002). Evidently, building cohesion amongst students in a classroom can be a fruitful pursuit.
Several factors influence inter-personal ties amongst students in a classroom.
Commonly stated factors include gender, race or ethnicity and class composition. Homophily is a widely recognized factor which is the tendency to form friendships or like students who are similar in gender, race or ethnicity (Graham, Munniksma, & Juvonen, 2014; Mcpherson, Smith-lovin, & Cook, 2001). Likewise, class composition or presence of students from different ethnic groups is also known to determine inter-ethnic friendships between students (Stark et al., 2015). However, very few studies acknowledge the role of teachers in
influencing peer perceptions and peer relations. Teachers are the “primary architect of the classroom context” (p. 9) who regulate the formal and the informal interactions amongst students and between teacher and students (Hendrickx, 2017). Teachers are known to influence student interactions through modeling and social referencing Hendrickx, 2017). The mechanism of modeling suggests that students emulate teacher’s behaviour towards other students in the classroom. For example, when a teacher addresses everyone in the classroom in a positive and encouraging way, students are likely to embody and display this behaviour in their peer interactions. In a similar vein, the mechanism of social referencing suggests that teachers shape the way students view other students in the classroom. For instance, if a teacher tends to ignore a specific group of students, the rest might infer that the teacher dislikes those students and therefore might choose to dislike or stay away from them. In this way, teachers being the central point of contact for all students for most of their time in the classroom, act as an “invisible hand” in influencing peer relations/feelings in the classroom (Farmer, McAuliffe Lines, & Hamm, 2011). Consequently, when teachers adopt a funds of identity based pedagogy, taking into account the literature discussed above, we can
expect to see a shift in the way teachers think about their students which could transform their behaviour and actions in the classroom. In which case, they can act as an example for students, thereby influencing liking among students and therefore cohesion within
classrooms.
The Present Study
This study is part of a collaborative action research project between thirteen teachers from twelve schools in Amsterdam and researchers from the University of Amsterdam (‘t Gilde & Volman, 2021 Volman & ’t Gilde, 2021). The main research questions that guide the current study are - 1) What effect does a funds of identity based pedagogical approach have
on cohesion amongst students in multicultural primary classrooms in the Netherlands?
Furthermore, since the schools that participated in the study are highly multicultural and diverse with students belonging to different communities such as Surinamese, Moroccan, Turkish, Antillean, African and other nationalities, we were interested in exploring - 2) What
effect does a funds of identity based pedagogical approach have on cohesion within and between majority and minority groups in multicultural primary classrooms in the Netherlands? In light of the existing research, firstly, we hypothesized that cohesion in
classrooms where the teachers implemented a funds of identity-based pedagogy would be higher than in classrooms where such a pedagogy was not followed. Secondly, in the intervention condition classrooms, we expected cohesion to improve significantly between majority and minority groups versus within majority and minority groups at the end of the intervention compared to the start of the intervention.
Method Research Design and Procedure
The current study builds on the research by ‘t Gilde & Volman (2021) and Volman and ’t Gilde (2021), which followed a pretest-posttest research design with a control group.
The schools that participated belonged to a school-university partnership program in
Amsterdam. The teachers who were interested in actively participating in the research, signed up to be part of the intervention group. Accordingly, schools that had similar populations as the intervention group were reached out, and teachers who were interested in the topic of the study signed up to be part of the control group. The control group teachers were offered a presentation about the topic at the end of the study.
The intervention took place between September 2018 and June 2019 with thirteen teachers from twelve primary schools in Amsterdam, The Netherlands in two main phases. The first phase had a twofold objective: first, to provide the theoretical underpinnings and some examples of a funds of identity based pedagogy; second, to encourage these teachers to gain a first-hand experience in identifying students’ funds of identity and think about possible ways to incorporate them in teaching and learning activities. The first phase lasted for
approximately eight weeks, with two planned teacher-researcher meetings. In this phase, student network data was collected as part of the student questionnaire from both intervention and control group students which served as a pre-test measure. The second phase consisted of two cycles of eight weeks each. At the beginning of each cycle, the intervention group
teachers prepared an action plan and throughout the cycle focussed on creatively implementing the method in their respective classrooms. In each of the cycles, planned teacher-researcher meetings were organized to support the intervention group teachers and to gain insights from their classrooms. At the end of the second phase, student network data was collected again as part of the student questionnaire from both intervention and control group students which served as a post-test measure.
Participation in the study was strictly voluntary for teachers and students. All
teachers, students and parents signed an informed consent form to participate in this study. In order to use the data from the student questionnaire, a passive consent of participation was
obtained from the parents of the students since the questionnaire did not include any personal or sensitive information. The study was approved by the ethics committee of the Faculty of Social and Behavioural Sciences of the University of Amsterdam (2018-CDE-9433). Participants
Network data was collected from students belonging to grades 4, 5 and 6 (N = 301) out of which 145 students belonged to the intervention group, and 156 students belonged to the control group. In the intervention group, there were 76 boys and 68 girls (Nmissing = 1).
These students were between 8 and 13 years old (Median age = 10 years, Nmissing = 5) and
belonged to six different classrooms from five schools. In the control group, there were 81 boys and 74 girls (Nmissing = 1). These students were between 8 and 12 years old (Median age =
10 years, Nmissing = 6) and belonged to nine different classrooms from four schools. A detailed
overview of the participants is provided in Table 1.
The schools catered to a multicultural student population belonging to different SES. A majority of students in the intervention group (N = 98) and in the control group (N = 99) had a non-native Dutch background. The linguistic diversity of the participants is reported in Table 2. The intervention and the control group did not differ in their gender distribution (2
(1) = 0.01, p = .928) and in their distribution of native-Dutch versus non-native Dutch population (2 (1) = 1.03, p = .310). However, they differed in the age distribution (2 (2) =
21.72, p < .001). There were more students in the younger age categories in the control group than in the intervention group.
Intervention
All the teachers in the intervention group devised various methods to identify students’ funds of identity. Most teachers had one on one discussions and used informal conversations with their students to know about their likes, interests and past time activities. Some teachers asked students to fill in a questionnaire about their talents. Few others used
students’ drawings, observed them during classroom activities and breaks to understand what they enjoyed doing, what they were good at and what motivated them. Additionally, some teachers reported having come across students’ funds of identity during a discussion in the context of a lesson and sometimes accidentally where the students voluntarily communicated certain information about themselves without being prompted to share.
Teachers used students’ funds of identity in many ways in the classroom. In some occasions, the teachers leveraged individual student’s funds of identity (e.g., drawing, cooking or exceptional knowledge on a particular topic such as farming) by providing an opportunity to share his/her experience or expertise with others. For example, one of the teachers used a student’s expertise in English language by asking the student to teach a lesson to the whole class. On some other occasions, the teachers leveraged funds of identity shared by a group (e.g., interest in rap songs, sports such as football or mobile games such as Fortnite, Battleship, Minecraft) by planning lessons on these themes, planning group
activities where students exchanged information and worked together (e.g., making a play or a film) and designing creative activities that appealed to all the students (e.g., making a drawing or a poster). Sometimes, these activities also extended beyond the classroom; for example, a teacher organized an end of the year musical on culture, and another teacher mobilized students’ knowledge on the World War II by organizing an exhibition of artefacts made by children. In many cases, teachers planned the activities or a teaching unit
incorporating students’ funds of identity, but there were also instances when teachers spontaneously used students’ funds of identity while teaching. Most importantly, though the teachers used students’ funds of identity differently in their classrooms, notably all teachers in the intervention group attempted to integrate this new method into their classroom practice throughout the academic year (‘t Gilde &Volman, 2021).
Cohesion. One network item in the student questionnaire asked students to report how nice they found each of their classmates on a scale of 1 (not at all nice) to 5 (very nice). We used this item as a measure of liking at the dyad level and cohesion at the collective level. In classrooms where on average students reported other students as nice or very nice, students in such classrooms were assumed to form a connected or a cohesive group compared to
classrooms where students reported others as not at all nice, not nice or doesn’t matter. This item was measured twice during the study: at the beginning (T1) and at the end of the study (T2). All students were provided with a list of the names of their classmates and a number assigned to each of them. In the student questionnaire, they were asked to use this list to mark their responses.
Majority and minority group membership. Students’ majority or minority group membership was determined by an item that measured the language spoken at home. Students were asked to choose between three options namely a) only Dutch b) Dutch and
another language (with an open-ended option to state the other language) c) another
language (with an open-ended option to state the other language). Students who reported that
they spoke only Dutch at home or stated that they spoke Dutch and English at home were coded as the majority group. Students who reported that they spoke Dutch and another language other than English at home, and students who reported that they spoke only another language at home were coded as the minority group. Within the minority group, a specific
group membership variable was created based on the specific language(s) stated in their
open-ended answer (e.g., Surinamese, Moroccan, Turkish). Specific groups that were distinguished are reported in Table 2.
Dyad level group membership variables. Students’ majority and minority group membership information was used to create dyad level dummy variables indicating the four possible tie types: ties within majority group, ties within specific minority group and ties
between majority and minority group and ties between minority group. The rating
provided/received by every student for every other classmate marks a tie. A student has as many ties as the number of ratings (s)he has provided/received. Ratings exchanged within majority group is referred to as ties within majority group, which was dummy coded as 1. Similarly, ratings exchanged within the specific minority group is referred to as ties within
specific minority group, which was dummy coded as 1 (e.g., within Moroccan students or
within Turkish students). Ratings exchanged between the majority and the minority group is referred to as ties between majority and minority group, which was dummy coded as 1. Finally, ratings exchanged between the minority group is referred to as ties between minority
group, which was dummy coded as 1 (e.g., between a Moroccan student and a Turkish
student).
Age. Students were asked to mark their age at the beginning (T1) and at the end of the study (T2) by choosing a discrete age between eight and thirteen
Control variables. Gender is known to influence peer liking amongst students, and there is substantial research in support of a tendency to like same-gender peers more than opposite-gender peers especially in young children (e.g., Hayden-Thomson, Rubin, & Hymel, 1987; Sippola, Bukowski, & Noll, 1997). Students reported their Gender in the questionnaire by choosing between a boy or a girl. We controlled for gender by creating a dummy variable at the dyad level, similarity in gender, by recoding same-gender ties as 1 and opposite-gender ties as 0.
Data Analysis
In this study, network data was collected at the classroom level at two time points. Since all students within a classroom rated each other in terms of liking, we have a
(longitudinal) round-robin data. Our fundamental unit of observation is this liking reported on a scale of 1 to 5 between students at the dyad level. A social relations model (SRM) is
appropriate to analyze such interpersonal dyadic phenomena especially when the variables are measured on an ordinal or a continuous scale (Back & Kenny, 2010; Knight &
Humphrey, 2018).
As explained by Back and Kenny (2010), each individual in a round-robin data set is both an actor/sender and a partner/receiver, and each dyad consisting of two directed ties represents a unique relationship between two individuals. As a result, additional sources of variance are accounted for in a dyadic framework using SRM: a sender effect, a receiver effect and a dyadic reciprocity effect. The sender effect accounts for an individual’s consistent tendencies in rating relations with others. The receiver effect accounts for how relations with specific individuals are generally rated by others. Dyadic reciprocity is the measure of the extent to which ties are reciprocated. Besides, generalized reciprocity reflects the extent to which the propensities of sending and receiving ties are related amongst
individuals. It is measured as the covariance between the actor/sender and the
partner/receiver effect. Finally, the relationship effect can be viewed as a kind of residual which remains after accounting for actor/sender, partner/receiver, dyadic and generalized reciprocity effects (Knight & Humphrey, 2019).
Most often, as in the present study, researchers collect multiple round-robin network data from different groups. In such scenarios, the most fitting analysis strategy is to use a multilevel parameterization of SRM (Snijders & Kenny, 1999). In such a framework, an additional variance at the group level can also be estimated to account for the differences between the round-robin networks. We used the functions created by Andrew Knight (Knight & Humphrey, 2019) in the nlme package (Pinheiro, Bates, DebRoy, Sarkar, R Core Team, 2020) in the software environment RStudio version 1.3.1073 (RStudio Team, 2020) to prepare the data and to perform our analysis. The functions, data preparation and analysis
strategy are detailed in the article by Knight and Humphrey (2019). The analysis was performed using the R version 4.0.2 (R Core Team, 2020).
Two-way interaction between time and condition was tested to examine the effect of the intervention on liking at the dyad level. Additionally, three-way interaction between time, condition and tie type was tested to examine the effect of the intervention on liking between and within majority and minority group students.
Results
Table 3 presents the descriptive statistics of all the variables in this study. The variables condition and time were measured at the class level and the network level respectively whereas the other variables were measured at the dyad level. There was no significant difference in mean liking between intervention condition and control condition networks at the start of the study (b = -0.17, t(13) = -1.07, p = .305). Mean liking at the dyad level at both time points were above the neutral mid-point of the liking scale for both
intervention condition and control condition networks. However, mean liking for the
intervention condition networks was found lower at the end of the study compared to the start of the study (from M = 3.41 to M = 3.31). Mean liking in ties between same-gender peers was higher than mean liking in ties between cross-gender peers; M = 3.83 versus M = 2.90
(averaged across time points) in the intervention condition networks and M = 3.83 versus M = 3.05 (averaged across time points) in the control condition networks.
Main Analysis
A two-level model that accounted for the variance at the network level and at the dyad level fit better than the model that ignored the multilevel (ML) network structure (2 (5) =
4120.09, p < .001).A two-level ML structure was appropriate as the variance at the
classroom level completely overlapped with the variance at the network level. In other words, when an SRM model with two additional levels for class and network was estimated, the
variance at the second level (i.e., network level) was estimated as zero, and the variance was reported only at the highest level (i.e., at the third level) (Three level: σ2
classroom = 0.06,
σ2network= 0, σ2dyad=1.02; Two level: σ2network= 0.06, σ2dyad=1.02). Therefore, a two-level ML
structure was specified for all models in the analysis.
Most of the variance in student liking was found at the dyad/relationship level (72.66%) and negligible variance was found between networks (4.02%) (Model 1, Table 4). Correlation between the actor and partner effect as indicated by generalized reciprocity was moderate (r = .36) which meant that actors who rated their peers high on the liking scale were also rated highly by their peers. Dyadic reciprocity indicated a large correlation (r = .53) between liking a specific partner and being liked by the specific partner in turn that is, there was a strong tendency to reciprocate liking between students.
We did not observe a significant main effect of time and condition, and a significant interaction effect between time and condition in the model that tested the effect of the intervention across time points and conditions (Model 2, Table 4). Between the timepoints, there was a decrease of 0.01 points in liking for the control group, b = -0.01, t(26) = -0.06, p = .952 whereas in the intervention group, there was a decrease of 0.07 points in liking rated by the students, b = 0.07, t(26) = 0.46, p = .652. At the end of the intervention, at time point 2, control condition classrooms were on average 0.23 points higher than the intervention condition classrooms, b = 0.23, t(26) = 1.55, p = .133. This finding indicated that mean liking between students in the intervention and the control condition classrooms were comparable at the start and the end of the study, and there was no significant effect of the intervention.
Furthermore, we did not observe a significant three-way interaction between tie type, time and condition. In other words, there were no significant differences in liking within majority group students, within specific minority group students, between majority and minority group students and within minority students between time points and between
conditions (Model 3, Table 4). The only significant difference was observed between liking within specific minority peers versus liking between minority at T1 for the control condition classrooms. At T1, mean liking within specific minority students in the control condition was 0.36 points higher than the mean liking between minority students, b = 0.36, t(12317) = 4.00,
p < .001. Moreover, we observed remarkably high mean liking in ties within specific
minority group students in the control condition classrooms at T1, M = 3.85, 95% CI = [3.49, 4.21] and at T2, M = 3.76, 95% CI = [3.40, 4.13]. On a cautionary note, we were unable to account for the effect of similarity in gender as a covariate in the analysis as models with the variable resulted in convergence error possibly due to a strong effect of the predictor
observed consistently across classrooms. Exploratory Analysis
We performed some exploratory analyses as a way of examining our results. In the first set of exploratory analyses, presented in Figure 1 and Figure 2, we examined the effect of the intervention across time in each of the different tie-types. In the control condition classrooms, we observed significantly higher liking within specific minority group students compared to liking within majority, between majority and minority and between minority group students, b = 0.32, t(12323) = 2.66, p = .039; b = 0.34, t(12323) = 4.02, p < .001; b = 0.31, t(12323) = 4.76, p < .001 respectively. In the intervention condition classrooms, the mean liking within specific minority group students was significantly higher only compared to liking between minority group students, b = 0.16, t(12323) = 3.13, p = .001. We also examined liking across condition and time points in each of the different tie types. The highest mean liking was observed within specific minority group students followed by within majority group students, M = 3.59, 95% CI = [3.41, 3.77], M = 3.49, 95% CI = [3.41, 3.77] respectively. Furthermore, the liking reported within specific minority students was
.001 and the liking reported between majority and minority students, b = 0.18, t(12326) = 3.11, p = .010. These findings indicate a preferential liking amongst students belonging to specific minority groups (e.g., Moroccan, Surinamese and African).
Second, we were interested to know if the results of our study would be different if we analyzed the like and the dislike networks separately. Null results observed in the intervention condition classrooms could also be a result of equal increase in liking and disliking amongst the students. To establish the effect of the intervention, we considered it worthwhile to analyze like and dislike networks separately as finding a reduction in dislike amongst peers was as valuable in terms of classroom cohesion as finding an increase in liking amongst students.
In order to do so, we recoded the variable that measured liking on the scale of 1 to 5 into two dummy coded variables like and dislike; values greater than 3 were coded as 1 in the like variable and values less than 3 were coded as 1 in the dislike variable. As presented in Table 5, we examined the count and the proportion of like and dislike ties at both time points for the intervention and the control condition classrooms. Additionally, we also examined the count and the proportion of like and dislike ties within majority students, within specific minority students, between majority and minority students and between minority students.
We found that less than 25% of the total ties were dislike ties in the intervention condition and less than 20% of the total ties were dislike ties in the control condition. In the intervention condition, we did not see an overall reduction in the number of dislike ties at T2 versus T1. On examining the count and proportion for each of the four tie-types, we noticed a reduction in dislike only amongst students within specific minority students. However, in the control condition, an overall decrease in the proportion of dislike ties was observed amongst students and in each of the four tie-types from time point 1 to time point 2. Concerning the like ties, a general decrease over time in the proportion of like ties was observed for both
intervention and control condition classrooms. These findings do not suggest an effect of the intervention on dislike amongst students in the intervention condition classrooms.
Thirdly, we were interested to know why the control condition classrooms on average had higher liking (in the composite scale) versus the intervention condition classrooms across time points. Since control condition classrooms were on average younger than the
intervention condition classrooms, we wanted to understand if age had a negative effect on mean liking. So, we added actor age and partner age as predictors to the final model (i.e., Model 3). However, we did not find a significant effect of actor age, b = 0.05, t(11678) =1.74, p = .083. and partner age, b = 0.02, t(11678) = 0.62, p = .535 on mean liking at the dyad level.
Finally, we were interested to understand the notably higher liking within specific minority students in control condition classrooms compared to the intervention condition classrooms. As presented in Figure 3, we looked at the distribution of students’ language background by gender and found no special pattern in the control condition. Furthermore, we also compared the proportion of similar gender ties within the specific minority group to know if control condition classrooms had a higher proportion of similar gender ties versus the intervention condition. However, we found that, in same-gender ties within specific minority students, only 38.5% (N = 308) ties were from the control condition classrooms versus 61.5% (N = 492) ties in the intervention condition. Clearly, the differences between the conditions could not be explained by age, gender and language distribution between them.
Discussion and Conclusion
The present study examined the effect of a funds of identity based pedagogy on cohesion amongst primary school students. The study also examined if the effect of such a pedagogy was different between and within majority and minority group students. First, it was hypothesized that there would be an increase in overall cohesion amongst students in the
form of increased interpersonal liking at the dyad level at the end of the study compared to the beginning of the study in the intervention versus the control condition classrooms. Furthermore, we hypothesized that, the cohesion between majority and minority group students would show greater increase over time than within majority and minority group students in the intervention condition compared to the control condition classrooms. This is among the first studies on funds of identity pedagogy that used a (quasi) experimental design and at the same time explored the network perspective of student relations.
Interestingly, we did not find any support for our hypotheses in the study. There was not any notable difference in the overall mean liking amongst students in the intervention condition classrooms at the end of the study compared to the start of the study versus the control condition classrooms. This was also true for liking between majority and minority group students. On the contrary, the qualitative findings from the intervention reported by Volman and ’t Gilde (2021) suggest promising and encouraging effects of the funds of identity based pedagogical intervention on student cohesion. These effects were reported from both the teachers’ and the students’ perspective. The authors reported that both the teachers and the students across the intervention condition classrooms indicated with
examples that the intervention resulted in improved interactions amongst the students that is, with less prejudice, with increased respect, acceptance and admiration for each other. They also reported that the intervention resulted in a positive, friendly and safe classroom atmosphere. Intriguingly, these qualitative findings did not translate into our quantitative analysis.
Nonetheless, having used SRM as the analysis strategy provided valuable insights into the results. Using SRM is apt to study social phenomenon such as cohesion as it allows an examination of a relationship at a micro-level (Knight & Humphrey, 2019). For cohesion to exist in a classroom, it must be evident in the relationships shared between two students
(dyad). SRM can account for the complexities in the interaction by separating or
decomposing the contribution from the sender, receiver and also the dyad (Back & Kenny, 2010). These sources of variances have been emphasized in studies on the nature of
interpersonal or bi-directional relationships (e.g., Dunn, 1997).
In this study, we observed that liking amongst students was strongly reciprocated. Specifically, we noted that, on average, students who rated others higher on liking were also rated higher on liking by others (Generalized reciprocity = .36). Additionally, liking within dyads was also strongly reciprocated (Dyadic reciprocity = .53), which meant that most often dyads shared similar feelings of liking towards each other. Notably, we observed most of the variance in student reported liking (73%) at the dyad level. Since interpersonal liking is fundamentally a dyadic phenomenon, we can expect to see such strong reciprocities and a large relationship variance (Back & Kenny, 2010). It is also evident in the study by Zimmer-Gembeck, Waters, and Kindermann (2010) who observed similar dyadic reciprocity in reports of peer liking within primary school students in Australia. In essence, the large variation in the relationship component in this study indicates that liking varied highly and idiosyncratically from one student to another.
In the following paragraphs, we reflect on the possible explanations for the findings from this study. Research on liking and disliking reported within classrooms suggests that liking and disliking are a function of various aspects of a sender/actor, receiver/partner, environmental and contextual circumstances. These factors include physical appearance of a classmate, first-hand contact experience, knowledge on how others perceive a particular classmate, a general attitude towards a specific ethnicity that is, the out-group attitude, the ethnic composition of the classroom, the ethnic composition of the neighborhood a student lives in, and students’ personality traits (e.g., self-concept)(Back & Kenny, 2010; Stark et al., 2015; Zimmer-Gembeck et al., 2010). Considering that student reported liking towards their
peers results from such a multitude of factors, it can be hard to influence and may need persistent efforts.
Another plausible explanation for not finding any effect of the intervention on student cohesion is the high level of mean liking reported at the start of the intervention in the
intervention and control condition classrooms (M = 3.41and M = 3.44 respectively). In classroom environments characterized by high interpersonal liking, the intervention conceivably did not result in significant differences over time possible since there is less scope for improvement. Furthermore, the feelings of dislike that were reported could be deeply rooted for which the duration of this intervention could be insufficient and might require targeted intervention with those specific students.
Furthermore, lack of a substantial difference in overall liking between the intervention and control condition classrooms at the end of the intervention corroborated with the small percentage of the variance observed at the network level (3.5%) indicating inconsequential variation in student reported liking between the networks. Moreover, the control condition classrooms on average reported (marginally) higher overall mean liking scores and between majority and minority group students compared to the intervention group classrooms. This discrepancy between the conditions was unexpected and could not be explained by the language background, age and gender distribution between the two conditions. However, a reason for this finding could be that both the intervention and the control group classrooms that participated in the study were selected based on teacher self-nomination. It is possible that the teachers were already interested in the intervention and were probably incorporating some aspects of a culturally responsive pedagogy as part of their school’s teaching ideology. Perhaps, therefore, we did not see any difference between intervention and control group classrooms. Finally, though the intervention occurred over one academic year, the teachers actively implemented the unit or the lesson plans in the classroom in the second phase of the
intervention for two cycles each lasting for eight weeks. This duration of the intervention could be too short to observe any effect.
With regard to the findings within and between majority and minority groups, in the intervention condition classrooms, we did not observe any notable increase in the mean liking between majority and minority groups at the end of the study compared to the beginning of the study. However, one finding that stood out was that the highest mean liking was observed within specific minority students across classrooms and across time points. We noticed that, on average, across conditions and time points, students within specific minority group reported liking each other (58%) more than what was reported within majority (47%),
between minority (48%) and between majority and minority group students (47%). Likewise, they reported lower dislike towards each other (15%) compared to what was reported within majority (20%), between minority (21%) and between majority and minority group students (20%). It appears that students belonging to a specific minority group displayed some sort of preferential liking or positive bias in their responses for each other.
This finding can be explained by the homophily phenomenon (Mcpherson et al., 2001) or the social identity theory (Tajfel, Turner, Austin, & Worchel, 1979) which suggests that people generally have a preference and share friendships with people who are like themselves. As specific minority group students share many things in common such as their cultural and traditional values, experiences and ways of living, it might facilitate satisfying interaction amongst them. A clear preference within students of the same minority ethnic background was also observed in other previous studies (e.g., Baerveldt, Van Duijn, Vermeij, & Van Hemert, 2004; Vermeij, van Duijn, & Baerveldt, 2009).
Drawing from the above theories, generally, in multicultural classrooms, one could expect low levels of liking between majority and minority group students. However, we found that, the mean liking between majority and minority group (across time points and
condition) was on the higher side and similar to the mean liking between minority students,
M = 3.41 and M = 3.36 respectively. In other words, liking shared between a Dutch and a
Moroccan student was comparable to liking shared between a Moroccan and a Turkish student. As presented earlier, a similarity was also observed in the proportion of like and dislike ties in these relations. We can use this finding as an indication that the positive bias observed within specific minority students was not because of a negative bias against students belonging to other ethnic groups but due to a natural liking shared amongst them. This finding is reassuring as it reduces the possibility of discrimination, segregation and prejudice in these classrooms.
The major limitation of this study is the measure that was used to examine cohesion amongst students. We noted earlier that interpersonal liking uniquely varies between individuals and also stems from a multitude of factors. Given that liking towards peers is based on several factors which are unique for every student, it indicates a general attitude that children have towards their peers. However, in order to measure the effect of the pedagogical intervention, we might need a more specific or a more directed measure. Another downside is the presence of a neutral item in the answer scale. A sizeable proportion (31%) of
relationships are marked neutral by their peers. Students who are rated by their peers as
neither like or neither dislike could also be neglected children in the classroom (Lightfoot,
Cole & Cole, 2013). Alternatively, it is also possible that some students who negatively perceive some of their peers might refrain from reporting dislike and instead choose a neutral option.
To avoid such ambiguities, an alternative to measure the effect of the intervention on student cohesion could be to assess the ease in interactions between students. Such a measure is reasonable as adopting a funds of identity based pedagogy in the classroom creates
establish a sense of connectedness or nearness amongst students. For instance, students could be asked – In the past week, how frequently did you have positive interactions with this
classmate? The answer categories could be a 5-point Likert scale - always, many times, sometimes, rarely and never. Another alternative could be - At this point, how comfortable do you feel interacting with this classmate? The answer categories could be a 5-point Likert
scale - totally comfortable, comfortable, neither comfortable nor uncomfortable,
uncomfortable and totally uncomfortable. Additionally, by specifying a reference period in
the question, children are, in some way compelled to think and report only about their time in the classroom when the intervention took place. This way, we might be able to capture the impact of the intervention in quantitative terms.
Additionally, as Volman and ’t Gilde (2021) point out, it is possible that teachers misinterpret the objective of funds of identity pedagogy as a means to focus only on unleashing students’ hidden talents or overemphasize the differences between students or limit students’ identity to students’ culture (i.e., food, dance and music). A suggestion for future iterations of the intervention would be to include classroom observations during the implementation phase. The observation notes or a recording could help teachers overcome these misconceptions and at the same time be a measure of the intervention implementation fidelity as the proof of an effective intervention lies in its implementation. Furthermore, classroom observation could also be used to assess cohesion amongst students.
Despite the null results, this study has implications for theory and practice. In the spirit of research, we consider our null results to be important as they guide future research efforts in a new field of research. Funds of identity based pedagogy is theoretically very well supported, and the qualitative findings of the intervention by Volman and ’t Gilde (2021) are encouraging and exciting. It remains a challenge to extend these findings with quantitative support. This study is an effort in the right direction as it followed a strong research design
(quasi-experimental longitudinal design) with an appropriate analysis strategy. The study also contributes to the longitudinal dyadic studies in classroom settings. Through this study, we would also like to encourage the use of dyadic analysis as it provides a framework to understand social phenomena at greater depths. Additionally, the above-average level of liking that was found between students in these multicultural classrooms consistently at both time points indicates that these schools and teachers are in some way taking up the goal of social inclusion as part of their efforts to provide quality education. Besides, the finding of the positive bias or a preferential liking within specific minority students without an apparent disliking between majority and minority students is noteworthy. It weakens the possibility of discrimination and segregation among students in these classrooms. Finally, through this study, we draw attention to the growing relevance of a culturally responsive pedagogy, a need to build connectedness and a social nearness between students within classrooms in the present age of multiculturalism where social cohesion is critical in societies.
References
Aronson, B., & Laughter, J. (2016). The Theory and Practice of Culturally Relevant Education: A Synthesis of Research Across Content Areas. Review of Educational
Research, 86(1), 163–206. https://doi.org/10.3102/0034654315582066
Au, K., & Kawakami, A. (1994). Cultural congruence in instruction. In E. Hollins, J. King, & W. Hayman (Eds.). Teaching diverse populations: Formulating knowledge base (pp. 5– 23). Albany, NY: State University of New York Press. Retrieved from
https://books.google.com/books?hl=en&lr=&id=HztfaERZPXMC&oi=fnd&pg=PA5&d q=Cultural+congruence+in+instruction.+&ots=GZWfYdWLny&sig=2ASxyCz1Su7nqB MB0zCvXFhnxrE
Back, M. D., & Kenny, D. A. (2010). The Social Relations Model: How to Understand Dyadic Processes. Social and Personality Psychology Compass, 4(10), 855–870. https://doi.org/10.1111/j.1751-9004.2010.00303.x
Baerveldt, C., Van Duijn, M. A. J., Vermeij, L., & Van Hemert, D. A. (2004). Ethnic boundaries and personal choice. Assessing the influence of individual inclinations to choose intra-ethnic relationships on pupils’ networks. Social Networks, 26(1), 55–74. https://doi.org/10.1016/j.socnet.2004.01.003
Baskerville, D. (2011). Developing cohesion and building positive relationships through storytelling in a culturally diverse New Zealand classroom. Teaching and Teacher
Education, 27(1), 107–115. https://doi.org/10.1016/j.tate.2010.07.007
Buhs, E. S., Ladd, G. W., & Herald, S. L. (2006). Peer exclusion and victimization: Processes that mediate the relation between peer group rejection and children’s classroom
engagement and achievement? Journal of Educational Psychology, 98(1), 1–13. https://doi.org/10.1037/0022-0663.98.1.1
classrooms. Canadian Journal of Applied Linguistics, 10(2), 221–240. https://doi.org/10.1117/12.661937
Dijkstra, A. B., De la Motte, P. I., & Eilard, A. (2014). Social outcomes of education.
Concept and measurement. In A. B. Dijkstra & P. I. De la Motte (Eds.), Social outcomes
of education: The assessment of social outcomes and school improvement through school inspections (pp. 29–49). Amsterdam University Press.
Dunn, J. (1997). Lessons from the Study of Bidirectional Effects. Journal of Social and
Personal Relationships, 14(4), 565–573. https://doi.org/10.1177/0265407597144010
Esteban-Guitart, M. (2012). Towards a multimethodological approach to identification of funds of identity, small stories and master narratives. Narrative Inquiry, 22(1), 173–180. https://doi.org/10.1075/ni.22.1.12est
Esteban-Guitart, M., & Moll, L. C. (2014a). Funds of Identity: A new concept based on the Funds of Knowledge approach. Culture and Psychology, 20(1), 31–48.
https://doi.org/10.1177/1354067X13515934
Esteban-Guitart, M., & Moll, L. C. (2014b). Lived experience, funds of identity and education. Culture and Psychology, 20(1), 70–81.
https://doi.org/10.1177/1354067X13515940
Farmer, T. W., McAuliffe Lines, M., & Hamm, J. V. (2011). Revealing the invisible hand: The role of teachers in children’s peer experiences. Journal of Applied Developmental
Psychology, 32(5), 247–256. https://doi.org/10.1016/j.appdev.2011.04.006
Flook, L., Repetti, R. L., & Ullman, J. B. (2005). Classroom social experiences as predictors of academic performance. Developmental Psychology, 41(2), 319–327.
https://doi.org/10.1037/0012-1649.41.2.319
Forghani-Arani, N., L. Cerna and M. Bannon (2019). The lives of teachers in diverse classrooms, OECD Education Working Papers, No. 198, OECD Publishing, Paris,
https://doi.org/10.1787/8c26fee5-en.
French, D. C., & Conrad, J. (2001). School Dropout as Predicted by Peer Rejection and Antisocial Behavior. Journal of Research on Adolescence, 11(3), 225–244.
https://doi.org/10.1111/1532-7795.00011
Friedkin, N. E. (2004). Social Cohesion. Annual Review of Sociology, 30, 409–425. https://doi.org/10.1146/annurev.soc.30.012703.110625
Graham, S., Munniksma, A., & Juvonen, J. (2014). Psychosocial Benefits of Cross-Ethnic Friendships in Urban Middle Schools. Child Development, 85(2), 469–483.
https://doi.org/10.1111/cdev.12159
Hayden-Thomson, L., Rubin, K. H., & Hymel, S. (1987). Sex preferences in sociometric choices. Developmental Psychology, 23(4), 558–562. https://doi.org/10.1037//0012-1649.23.4.558
Hendrickx, M. (2017). The role of the teacher in classroom peer relations. Utrecht University. Retrieved from
https://dspace.library.uu.nl/bitstream/handle/1874/347160/Hendrickx.pdf?sequence=1 Hogg, L., & Volman, M. (2020). A Synthesis of Funds of Identity Research: Purposes, Tools,
Pedagogical Approaches, and Outcomes. Review of Educational Research, 90(6), 862– 895. https://doi.org/10.3102/0034654320964205
Hughes, M., & Pollard, A. (2006). Home-school knowledge exchange in context.
Educational Review, 58(4), 385–395. https://doi.org/10.1080/00131910600971784
Hviid, P., & Villadsen, J. W. (2014). Cultural identities and their relevance to school practice.
Culture and Psychology, 20(1), 59–69. https://doi.org/10.1177/1354067X13515938
Inglis, C. (2008). Planning for cultural diversity (Vol. 87). Paris: UNESCO. Retrieved from https://inee.org/sites/default/files/Planning_for_Cultural_Diversity_EN.pdf
(2013). Connected learning: an agenda for research and design. Digital Media and
Learning Research Hub. https://doi.org/10.1016/B978-008044654-7/50201-6
Jacob, E., & Jordan, C. (1987). Moving to Dialogue. Anthropology & Education Quarterly,
18(4), 259–261. Retrieved from http://www.jstor.com/stable/3216655
Jovés, P., Siqués, C., & Esteban-Guitart, M. (2015). The incorporation of funds of knowledge and funds of identity of students and their families into educational practice: A case study from Catalonia, Spain. Teaching and Teacher Education, 49, 68–77.
https://doi.org/10.1016/j.tate.2015.03.001
Knight, A. P., & Humphrey, S. E. (2019). Dyadic data analysis. In The handbook of
multilevel theory, measurement, and analysis. (pp. 423–447).
https://doi.org/10.1037/0000115-019
Ladd, G. W. (2006). Peer rejection, aggressive or withdrawn behavior, and psychological maladjustment from ages 5 to 12: An examination of four predictive models. Child
Development, 77(4), 822–846. https://doi.org/10.1111/j.1467-8624.2006.00905.x
Ladd, G. W., & Troop-Gordon, W. (2003). The Role of Chronic Peer Difficulties in the Development of Children’s Psychological Adjustment Problems. Child Development,
74(5), 1344–1367. https://doi.org/10.1111/1467-8624.00611
Ladson-Billings, G. (1995). Toward a Theory of Culturally Relevant Pedagogy. American
Educational Research Journal, 32(3), 465–491.
https://doi.org/10.3102/00028312032003465
Ladson-Billings, G. (2014). Culturally relevant pedagogy 2.0: A.k.a. The remix. Harvard
Educational Review, 84(1), 74–84.
https://doi.org/10.17763/haer.84.1.p2rj131485484751
Leeman, Y. (2006). Teaching in ethnically diverse schools: Teachers’ professionalism.
https://doi.org/10.1080/02619760600795171
Mcpherson, M., Smith-lovin, L., & Cook, J. M. (2001). Birds of a Feather: Homophily in Social Networks. Annual Review of Sociology, 27, 415–444.
https://doi.org/https://doi.org/10.1146/annurev.soc.27.1.415
Moody, J., & White, D. R. (2003). Structural Cohesion and Embeddedness: A Hierarchical Concept of Social Groups. American Sociological Review, 68(1), 103–127. Retrieved from https://www.jstor.org/stable/3088904
Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team (2020). nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-151, Retrieved from https://CRAN.R-project.org/package=nlme.
Poole, A. (2017a). Funds of Knowledge 2.0: Towards digital Funds of Identity. Learning,
Culture and Social Interaction, 13, 50–59. https://doi.org/10.1016/j.lcsi.2017.02.002
Poole, A. (2017b). “I want to be a furious leopard with magical wings and super power”: Developing an ethico-interpretive framework for detecting Chinese students’ funds of identity. Cogent Education, 4(1). https://doi.org/10.1080/2331186X.2017.1316915 Poole, A., & Huang, J. (2018). Resituating Funds of Identity Within Contemporary
Interpretations of Perezhivanie. Mind, Culture, and Activity, 25(2), 125–137. https://doi.org/10.1080/10749039.2018.1434799
RStudio Team (2020). RStudio: Integrated Development for R. RStudio, PBC, Boston, MA Retrieved from http://www.rstudio.com/.
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Retrieved from https://www.R-project.org/. Radstake, H., & Leeman, Y. (2010). Guiding discussions in the class about ethnic diversity.
Intercultural Education, 21(5), 429–442. https://doi.org/10.1080/14675986.2010.521378
Research, 44(3), 299–312. https://doi.org/10.1080/001318802200003159
Roseth, C. J., Johnson, D. W., & Johnson, R. T. (2008). Promoting Early Adolescents’ Achievement and Peer Relationships: The Effects of Cooperative, Competitive, and Individualistic Goal Structures. Psychological Bulletin, 134(2), 223–246.
https://doi.org/10.1037/0033-2909.134.2.223
Rubin, K. H., Bukowski, W. M., & Parker, J. G. (2007). Peer Interactions, Relationships, and Groups. Handbook of Child Psychology, 3. Retrieved from
http://www.ncbi.nlm.nih.gov/pubmed/24654727
Schmuck, R. A. ., & Schmuck, P. A. (1975). Group Processes in the Classroom.
International Encyclopedia of the Social & Behavioral Sciences: Second Edition.
https://doi.org/10.1016/B978-0-08-097086-8.92078-6
Sippola, L. Κ., Bukowski, W. M., & Noll, R. Β. (1997). Dimensions of Liking and Disliking Underlying the Same-Sex Prefernce in Childhood and Early Adolescence.
Merrill-Palmer Quarterly, 43(4), 591–609. Retrieved from
https://www.jstor.org/stable/23093361
Sleeter, C. E. (2012). Confronting the Marginalization of Culturally Responsive Pedagogy.
Urban Education, 47(3), 562–584. https://doi.org/10.1177/0042085911431472
Snijders, T. A. B., & Kenny, D. A. (1999). The social relations model for family data: A multilevel approach. Personal Relationships, 6(4), 471–486.
https://doi.org/10.1111/j.1475-6811.1999.tb00204.x
Stark, T. H., Mäs, M., & Flache, A. (2015). Liking and disliking minority-group classmates: Explaining the mixed findings for the influence of ethnic classroom composition on interethnic attitudes. Social Science Research, 50, 164–176.
https://doi.org/10.1016/j.ssresearch.2014.11.008
