SOCIAL DOMINANCE IN PRESCHOOL: PREDICTING AFFILIATION AND COERCION
THROUGH INTERACTIONS AND RELATIONSHIPS
By
Megan C. Fedor

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Education Psychology and Educational Technology – Doctor of Philosophy
2013

ABSTRACT
SOCIAL DOMINANCE IN PRESCHOOL: PREDICTING AFFILIATION AND COERCION
THROUGH INTERACTIONS AND RELATIONSHIPS
By
Megan C. Fedor
Previous research has indicated a variety of factors impact affiliative and coercive
interactions and relationships between peers. This study continues this work by examining how
individual characteristics and characteristics nested within a relationship impact affiliation and
coercion. 104 preschool children were observed during freeplay. Teachers completed ratings for
each child and children completed sociometric nominations. Results indicated that children rated
high on social dominance and prosocial resource control were more likely to engage in social
play. Children played with children that they nominated as preferred peers, and did not have
conflict with those that they nominated as disliked peers. Children were more likely to play with
and have conflict with peers of similar age and same sex. Children who were high on social
dominance were more likely to engage in social play. Finally children who were rated as high on
coercive resource control and low on bistrategic resource control were more likely to be in
conflicts compared to their peers. Overall the study indicated certain individual and dyadic
characteristics are related to affiliation and conflict.

ACKNOWLEDGEMENTS

I would like to thank my committee members for their guidance and feedback on this
project. I am truly honored to have had Dr. Cary Roseth as my advisor. He was one of my
inspirations to earn my degree in educational psychology, and has provided me with endless
support and encouragement as I progressed through my graduate program.
Thank you to Zach for your love, support and encouragement during this process – if it
were not for your encouragement to be an educational psychologist I wouldn’t be earning this
degree. Thank you to Stacy for the endless laughter and support, as we became scholars. Thank
you to my daughter Molly, and my family for their continued advice and care during this
process.

iii

TABLE OF CONTENTS

LIST OF TABLES

vi

INTRODUCTION
Peer Interactions and Relationships
Interactions
Relationships
Affiliative relationships
Coercive relationships
Social Dominance Interactions and Dyadic Relationships
Interactions
Social dominance and play state interactions
Sociometric “like” nominations, observed social play,
and social dominance
Sociometric “dislike” nominations, observed conflict,
and social dominance
Dyadic relationships
Social dominance and observed social play
Social dominance status and observed conflict
Social dominance and conflict role
Measurement Related Questions
Teacher-rated social dominance and resource control
Sociometric nominations and observed play and conflict partners
Current Study

1
2
2
3
3
4
4
4
7
8

METHOD
Preschoolers
Observational Procedures
Video Coding
Teacher Ratings
Sociometric Nominations
Data Analysis
Descriptive statistics
Social network analysis

18
18
18
20
20
21
21
22
22

RESULTS
Social Dominance and Resource Control
Social Dominance and Play States
Sociometric “Like” Nominations
Sociometric “like” nominations and observed play partners
Sociometric “like” nominations, observed social play, and
social dominance status
Similarity in social dominance status and observed social play

24
24
24
24
25
25

iv

9
9
10
12
13
13
14
16
16

26

Sociometric “Dislike” Nominations and Conflict
Similarity in Social Dominance Status and Observed Conflict
Social Dominance Status and Conflict Role
Initiator
Target

28
29
30
30
30

DISCUSSION
Teacher-rated Social Dominance and Resource Control
Social Dominance and Play States
Sociometric “Like” Nominations
Similarity in Social Dominance States and Observed Social Play
Sociometric “Dislike” Nominations
Similarity in Social Dominance and Observed Conflict
Social Dominance Status and Conflict Role
Practical Implications
Limitations and Conclusions

32
32
33
34
34
37
38
40
41
42

REFERENCES

50

v

LIST OF TABLES
Table 1

Partial Correlations for Teacher-rated Dominance, Resource Control,
and Observed Play

46

Table 2

Relative Social Dominance, Resource Control, and Play Type

47

Table 3

Relative Social Dominance, Resource Control, and Conflict

48

Table 4

Conflict Role as Initiator or Target and Relationship History

49

vi

INTRODUCTION
The opportunities for children to interact with peers increase dramatically during the
preschool years, with children spending more time in social rather than solitary play (BlurtonJones, 1972) and upwards of 50% of their time with peers rather than adults (Ellis, Rogoff &
Cromer, 1981). The way in which preschoolers navigate this increased time in peer interactions
is important, as peer relationships promote the development of a variety of social and cognitive
skills that are predictive of later social and emotional well-being (Rubin, Bukowski & Parker,
2006).
Preschooler’s peer relations are unique in that they frequently represent children’s first
horizontal relationships, meaning that all members start off with a relatively equal balance of
power (cf. child/parent relationship). Because of this, one of the developmental tasks that arise in
peer relationships is how to negotiate power in terms of controlling limited resources (e.g., toys,
friends). Children use a variety of strategies to gain and maintain control of these limited
resources, including coercive (e.g., aggressive), prosocial (e.g., cooperative) and bistrategic
strategies (e.g., a combination of the former) (Charlesworth, 1996; Hawley, 1999, 2002, 2003;
Hawley & Little, 1999; LaFreniere & Charlesworth, 1983; Pellegrini, Roseth, Mliner, Bohn, Van
Ryzin et al., 2007; Roseth, Pellegrini, Bohn, Van Ryzin, & Vance, 2007; Roseth, Pellegrini,
Dupuis, Bohn, Hickey et al., 2011; Vaughn, Vollenweider, Bost, Azria-evans, & Snider, 2003).
Theory and research related to preschoolers’ social dominance, defined as naturally occurring
differences in resource control (Hawley, 1999; Pellegrini, 2008), suggest that children using
bistrategic strategies are more likely to control resources than children using only coercive or
prosocial strategies. The use of bistrategic strategies is also positively related to social
dominance status, increased levels of peer regard, attention, and affiliation (Hawley, 2003;

1

LaFreniere & Charlesworth, 1983; Pellegrini et al., 2007; Roseth et al., 2007; Roseth et al.,
2011).
Clarifying the behavioral antecedents (e.g., resource control strategies) and social
consequents (e.g., increased peer regard, attention, and affiliation) of social dominance has
increased understanding of the social context in which preschoolers’ peer relations occur. Yet a
gap in knowledge remains about the actual peers and peer groups with whom socially dominant
preschoolers interact. Accordingly, the purpose of this dissertation is to examine whether social
dominance is associated with increased affiliation with preferred peers and with peers
characterized by commensurate levels of social dominance.
This introduction is divided into three parts. Part One introduces two levels of analysis
for preschoolers peer relations: interactions and dyadic relationships. Part Two applies these
levels of analysis to preschoolers’ social dominance, explaining how one level of analysis (e.g.,
coercive and affiliative interactions) informs other levels (e.g., social dominance relationships).
Part Two also reviews theory and research on preschoolers’ social dominance, discussing the
role that social dominance plays in preschool children’s dyadic relationships. Finally, Part Three
summarizes the specific hypotheses tested in this dissertation.
Peer Interactions and Relationships
Preschoolers’ peer relations may be understood in terms of two levels of analysis:
interactions and dyadic relationships.
Interactions. An interaction is defined as a “social exchange between children in which
participants’ actions are interdependent, such that each actor’s behavior is both a response to, and
stimulus for, the other participant’s behavior” (Coplan & Arbeau, 2009). Thus, in its most basic
form, an interaction is a behavior of one individual towards another (e.g., Child A takes X from

2

Child B) (Hinde, 1979). Historically, children’s peer interactions have been characterized
dichotomously in terms of affiliative, or “moving toward others” (e.g., cooperation, prosocial
behaviors), and coercive, or “moving against” or “moving away from others” (e.g., aggression)
(Rubin et al., 2006, p. 576). Underlying this perspective is the assumption that affiliative
interactions index social competence while coercive interactions index social incompetence
(Bukowski, 2003; Hawley, 2007; for a review of antisocial correlates, see also Coie & Dodge,
1998). More recent work challenges this view, however, as a functional perspective reveals that
both affiliation and coercion may be used in the service of resource control. This perspective will
be discussed more fully in Part Two of this introduction.
Relationships. According to Hinde (1979), a history of interactions between two
individuals forms the basis of a dyadic relationship, which in turn influences the type and range
of subsequent interactions between those individuals. For preschoolers, this means that different
histories of dyadic interactions yield different kinds of peer relationships (Brownell & Brown,
1992; Gifford-Smith & Brownell, 2003). For example, some dyadic interactions may result in
affiliative peer relationships, coercive peer relationships, or peer relationships characterized by
both affiliative and coercive interactions.
Affiliative relationships. Affiliative relationships are characterized by a history of
affiliation (e.g., cooperation, prosocial behavior) and, presumably, promote social competence by
socializing affiliative behavior. Traditional views of friendship formation suggest that individual
differences such as sex, ethnicity, and socioeconomic status predict with whom peers affiliate
with and, over time, form friendships (McPherson, Smith-Lovin & Cook, 2001). Research on
preschoolers’ affiliative relationships also tends to emphasize the role of demographic
characteristics such as sex (e.g., Fabes, Martin, & Hanish, 2003) and race (e.g., Shrum, Cheek &

3

Hunter, 1988), as well as behavioral homophilies (e.g., Rubin, Lynch, Coplan, Rose-Krasnor &
Booth, 1994; for a review see Vaughn & Santos, 2009). This dissertation study contributes to this
literature by testing whether social dominance status is associated with preferred play partners.
Coercive relationships. Coercive relationships are characterized by a history of
conflictual bouts, defined as agonistic events in which one child shows resistance (e.g., facial
expression, posture, tone of voice), physical aggression, or verbal aggression in opposition to the
influence attempt, anger, aggression, or argument of another child (Verbeek & de Waal, 2001;
see also Shantz, 1987). Presumably, coercive relationships promote social incompetence by
socializing the use of aggression in conflictual bouts (see e.g., Patterson, DeBaryshe, & Ramsey,
1989). Among preschoolers, such conflicts often involve a struggle for objects or directing
themes of play (Chen, Fesin & Tam, 2001; Laursen & Hartup, 1997), both considered to be
desired resources (Hawley, 2002). This dissertation study contributes to this literature by testing
whether social dominance status predicts who initiates and who is the target of conflictual bouts.
Social Dominance Interactions and Dyadic Relationships
Evolutionary theory (Darwin, 1859/1884; Hawley, 1999; Pellegrini, 2008) suggests that
resources are needed to promote survival, and are conceptualized as material (e.g., nesting sites),
social (e.g., mates), and informational (e.g., the acquisition of the latter two resources). For
preschoolers, resources range from materials, such as toys, to relationships, such as friends and
preferred playmates (Hawley, 2007; Pellegrini, Roseth, Van Ryzin & Solberg, 2011; Pusey &
Packer, 1997; Roseth et al., 2011).
Interactions. It is important to recognize that social dominance is conceptualized as a
relational variable indexing relative differences in the ability to control resources (Hawley, 1999;
Pellegrini, 2008). Thus, rather than index dominant and non-dominant (subordinate) behaviors,

4

social dominance indexes resource control status relative to other members of the group
(Hawley, 1999; Pellegrini, 2008; Roseth et al., 2011). At the group-level, this relative
comparison yields a continuum with children on the high end of the continuum accessing more
resources than children at the low end. Importantly, theory and research suggest that once social
dominance status is established within a group, group-level conflictual bouts tend to decrease
(Pellegrini et al., 2007; Roseth et al., 2007; Roseth et al., 2011). Thus, over time, social
dominance is associated with decreasing competition (e.g., aggressive conflict) over resources,
presumably because individuals become increasingly aware of their rank and the relative costs
and benefits associated with continued competition.
Research suggests that, like non-human primates (e.g., de Waal, 1986) and older children
(e.g., Hawley, 2003; Pellegrini & Bartini, 2001), preschool children use both prosocial (e.g.,
sharing, cooperating, turn taking) and coercive resource control strategies (e.g., hostile and
aggressive resource control attempts) to gain access to desired resources (Charlesworth, 1996;
Hawley, 1999, 2002, 2003; Hawley & Little, 1999; LaFreniere & Charlesworth, 1983; Pellegrini
et al., 2007; Roseth et al., 2011; Roseth et al., 2007; Vaughn et al., 2003). This functional
perspective of preschoolers’ affiliative and coercive behaviors challenges traditional views
exclusively linking coercion to peer dislike and suggests a more complex relationship between
coercion and affiliation.
Focusing on social dominance and affiliation, Strayer (1989; 1992) examined the
correlation between affiliation and social dominance in a sample of 1- to 5-year-old children.
Social dominance was conceptualized solely in terms of coercive behavioral correlates, defined
as “initiated conflict: attack, threat and competition” (p.155) and affiliation was measured by
adding the total affiliative activity received (e.g., proximity, social orientation, physical contact,

5

gesture). Results indicated that social dominance was correlated with affiliation among 4- and 5year old children and thus, supported the view that high social dominance status is associated
with increased levels of affiliation.
LaFreniere and Charlesworth (1983) examined social dominance and affiliation in a
short-term longitudinal study of 18 preschool children. Observational data were collected for a 9month period and the authors measured social dominance rank (defined as attacks, physical
struggles or assertiveness) and affiliation (defined as the proportion of time a child spent in
interactive play). Results suggested that children who are socially dominant also affiliate at
greater rates with their peers. Similar to Strayer however, dominance was conceptualized solely
in terms of coercion, rather than other types of resource controlling strategies, such as prosocial
and bistrategic strategies.
Finally, Roseth et al. (2011) examined social dominance and resource control strategies
in a short-term longitudinal study of 88 preschool children. Observational data were again
collected for a 9-month period but the authors defined social dominance status in functional
terms (i.e., naturally occurring differences in resource control) rather than particular behaviors
(e.g., coercion). Results suggested that rates of both coercive and prosocial resource control
strategies varied as a function of social dominance status, and that children who engaged in postconflict reconciliation were most-liked by peers. These findings suggest that the social
consequences of resource control competition depend on more than single behavioral tendencies
such as using more or less coercion than other children, and that even preschoolers tend to prefer
peer interactions involving both coercive and affiliative behaviors over interactions only
involving only coercion.

6

In sum, research on preschoolers’ social dominance has examined the relationship
between coercive resource control strategies and peer affiliation (LaFreniere & Charlesworth,
1983; Strayer, 1980b; 1989; 1992) and the acceptability and effectiveness of both coercive and
affiliative resource controlling strategies (Hawley, 2002, 2003; Roseth et al., 2011). What is
missing from this work is a more precise view of how social dominance and bistrategic resource
control strategies relate to interactions with particular social partners. After all, different social
relationships are thought to afford different interactions patterns (Hinde, 1979), and social
dominance relationships in particular may be associated with different play states and different
social consequents. The next sections consider this issue more specifically by examining whether
different play states (social vs. parallel play) and sociometric correlates (like versus dislike
nominations) depend on social dominance.
Social dominance and play state correlations. Broadly, children’s affiliative play may be
divided into two categories: social play (e.g., interactions with peers) and parallel play (e.g.,
playing beside peers). Social play is affiliative in nature, and allows for resource control
opportunities in terms of proximity to peers, the opportunity to gain desired toys from playmates,
and the direction of roles and themes of games. Parallel play lacks an interactive component and
is characterized by close proximity and mutual awareness of playmates (Howes & Mathenson,
1992). Parallel play affords resource control opportunities in terms of proximity to peers and the
opportunity to gain desired resources. This dissertation examines whether socially dominant
preschoolers (i.e., bistrategic resource controllers) are more likely than non-dominant peers to
engage in social and parallel play. Specifically, this dissertation examines whether children who
engage in greater proportions of social and parallel play also (a) tend to use prosocial and

7

bistrategic resource control strategies more than peers and (b) are rated by teachers as being
higher on social dominance.
Sociometric “like” nominations, observed social play, and social dominance. Children
higher on the social dominance continuum within a group are afforded greater access to desired
resources and are considered to be socially competent (Hawley, 2002; 2003; LaFreniere &
Charlesworth, 1983; Pellegrini, 2008; Roseth et al., 2011; Strayer & Strayer, 1976). Theory and
research also suggest that preferential access to resources makes socially dominant preschoolers’
attractive to peers, at least as measured by sociometric ratings (Pellegrini, 2008; Roseth et al.,
2007; Roseth et al., 2008). What is missing from this research is observational evidence that
socially dominant children do in fact affiliate more frequently (e.g., actually interact with) with
peers with whom they prefer to affiliate (e.g., peer preference nominations).
Previous research on preschoolers’ social dominance has operationalized prosocial
resource control in terms of control over play (e.g., Hawley, 2002; 2003; LaFreniere &
Charlesworth, 1983), affiliative interactions (Pellegrini et al., 2007; Roseth et al., 2011; Strayer
& Strayer, 1976), and peer regard (Pellegrini et al., 2007; Roseth et al., 2011; Vaughn & Waters,
1983) but ignored who, exactly, affiliates with whom, and whether social dominance is
associated with actually spending time with desired play partners. To date, therefore, there exists
no observational evidence that socially dominant children do in fact have more affiliative
opportunities than other children. This dissertation addresses this void by testing whether
socially dominant children tend to affiliate (spend at least 10% of their freeplay time) with their
desired play partners (measured by peer preference nominations) at a greater proportion than less
dominant children. Specifically, this dissertation examines the correspondence of peer preference
nominations to actual time spent with peers (e.g., child A plays with 3 of 3 peers they nominated

8

as preferred playmates; child b plays with 1of 3 peers they nominated as preferred playmates).
Simple regression is then used to determine if social dominance and resource control strategies
predict correspondence between peer nominations and time spent with peers.
Sociometric “dislike” nominations, observed conflict, and social dominance. Because
socially dominant children are able to negotiate the balance between resource control and
affiliation, it may be the case that these children are aware of whom they acquire or attempt to
acquire resource from, and whether or not that outcome will result in a disliked outcome
(Pellegrini et al., 2007; Roseth et al., 2007; Roseth et al., 2011). Because of this, socially
dominant children may show a higher degree of consistency between children they dislike, and
children with whom they have actual coercive conflict. This dissertation attempts to answer this
question by examining the consistency between sociometric nominations of “dislike” and actual
conflict, as it relates to social dominance status. Cluster analysis was used to determine if social
dominance and resource control strategies predicted correspondence between negative
sociometric nominations and actual conflict.
Dyadic relationships. The current study is examining social dominance through
preschoolers interactions and relationships. One goal is to examine how the use of social
dominance and different social dominance strategies are related to affiliative and coercive
relationships in preschool years. It may be the case that over time, children use different
strategies at different developmental time periods due to changes in their ability to recognize
successful strategy use, or acceptable behavior according to teachers, parents and peers. While
the current study is not examining changes in strategy use over time, it is examining the
manifestation of multiple strategies in the preschool years, and how the use of those strategies
may relate to relationship formation. Examining this with a preschool sample is particularly

9

important due to several factors. First, preschool can be thought of as sensitive period for the
development of peer relationships (Hay, Caplan & Nash, 2011). It is at this point where peer
play increases due to opportunity (e.g., preschool) and the development of skills such as
perspective taking, theory of mind and empathy. Thus, the interactions between children are ever
increasing.
Secondly, and in accordance with the first reason, children at this age have had fewer
socialization opportunities compared to older children. For instance, children are just learning
classroom rules and norms, and likely these are not yet internalized. Given this, the interactions
children have with one another will likely be informed at least partially by instinctual resource
controlling bids, rather than socialized resource controlling bids. Thus, the manifestation of
social dominance and social dominance strategies may differ as a function of age due to the
increasing socialization opportunities throughout development. Together this makes
preschoolers particularly unique to study because it provides a window into first dyadic peer
relationships without the socialization of acceptable social dominance strategies.
This next section links social dominance to preschoolers’ affiliative and conflictual
dyadic relationships.
Social dominance and observed social play. Social dominance hierarchies are
characteristics of groups that may facilitate or restrict with whom children interact (Rubin et al.,
2006). Children with a higher dominance status within a group may have the opportunity to
socialize with peers with similar status, reinforcing beneficial and accepted behavioral correlates
of resource control. Conversely children who are on the low end of a social dominance
continuum may want to affiliate with children of a higher dominance status, yet their place on

10

the continuum is prohibitive of doing so. Thus, these children may continue to socialize with
submissive or coercive resource controllers, which in turn impacts their own socialization.
Even in preschool, children affiliate selectively (Fabes, Martin & Hanish, 2009), often
based on behavioral and status homophiles (Vaughn & Santos, 2009). However research has yet
to examine how preschool social dominance hierarchies impact dyadic affiliation. It may be the
case that children high on the social dominant continuum within a group spend time with other
highly dominant children (e.g., birds of a feather). However it may be the case that highly
socially dominant children recognize that in order to maintain resource control, they need to
affiliate with children of lower dominance status.
Past work has used child level analysis of social dominance status and resource control
within a classroom as predictors of affiliation (LaFreniere & Charlesworth, 1983; Hawley, 2002,
2003; Strayer, 1980b; 1989; 1992; Roseth et al., 2011). This dissertation adds to this literature by
using social network analysis to determine how the characteristics of the relationship, rather than
the individual may impact affiliation (Wasserman & Faust, 1994). Social network analysis is
relevant to the peer relationships literature as it considers the collection of individuals within a
relationship as predictors of interactions (e.g., affiliative or coercive) and outcomes (e.g.,
resource exchange) within that relationship. Thus, the individual within the context of the dyad is
taken into account, rather than simply one’s individual characteristics or behaviors (Wasserman
& Faust, 1994).
Social network analysis has been used in past research on peer relations to investigate the
nature of affiliative dyadic relationships. For instance, Mercer and DeRosier (2010) found that
pairs of children in middle school who were similar in feelings of isolation were more likely to
affiliate with one another. Additionally, research has found adolescent peer affiliation to be

11

based on externalizing and internalizing similarity within dyads (Snijders et al., 2007; Veenstra &
Steglich, 2011). Additionally, some research with social network analysis has begun to examine
negative peer structures. For example, Huitsing and Veenstra (2012) examined how classroom social
network structure influenced bullies and victims of bullying among elementary school children. We
know of no previous research using social network analysis to examine affiliative and coercive
relationships in preschool settings.

Another aim of this dissertation is to use social network analysis to examine how
similarity in social dominance status and resource control strategies relates to affiliative
relationships. Similarity across these variables would be consistent with the demographic
homophily hypothesis. In this study, affiliation is measured by observational data of actual time
spent affiliating with a peer, and similarity in dominance status within groups is determined by
finding the difference scores between partners.
Social dominance status and observed conflict. Traditional views of conflict suggest that
conflict results from individual characteristics, such as genetic predispositions, environmental
affordances or constraints (e.g., one’s status within a social dominance hierarchy), and the
combination of the two (de Waal, 2000). More recent conceptualizations of conflict emphasize
the way dyadic relationships contribute to conflictual outcomes. For example, De Waal’s (1996,
2000) Relational Model provides one account of the likelihood of conflict between a pair of
individuals. The Relational Model posits “that the tendency to initiate aggression increases with
the number of opportunities for competition, the resource value and the reparability of the
relationship, while it decreases with the risk of injury and the value of the relationship” (de Waal,
2000, p. 27). Thus, individuals who spend more time together, have highly valuable resources to
exchange and the ability to reconcile may be more likely to have conflictual bouts. Conversely,
partners who have few opportunities to interact, have less attractive resources, and lack the
12

ability or desire to reconcile may have fewer conflicts.
Consistent with the Relational Model, social network analysis may provide insight as to
the characteristics nested within the preschoolers’ dyadic relationships that predict conflict.
Accordingly, another purpose of this dissertation was to investigate the nature of dyadic
relationships, and how their characteristics may predict conflict between pairs of children, even
after social dominance hierarchies have formed.
Social dominance and conflict role. A gap in the social dominance literature lies in
understanding how social dominance relates to roles (e.g., initiator, target) within conflicts.
Thus, a goal of this dissertation was to investigate the role that relationship characteristics, such
as social dominance and resource control strategies, have on conflictual bouts during a spring
term, presumably after social dominance hierarchies are already well established. Based on the
notion that a dominance status is stable within our sample, and conflict is more likely to be
initiated when an attractive resource is at stake (de Waal, 2000), it was predicted that those
initiating conflict would hold a lower social dominance rank relative to their chosen partner.
Targets, on the other hand were predicted to have greater social dominance and resource control,
attracting conflict based on their greater access and acquisition of desired resources.
Measurement Related Questions
As a whole this study aims to add to the literature by examining affiliation and coercion
through direct observations. Previous work has used teacher nominations to determine affiliative
and coercive relationships between peers. Additionally previous work has also used sociometric
nominations to determine liked and disliked peers. However these measures do not capture who
children are actually spending time with, and who children are actually having conflict with.

13

This study may help provide support for previous methods used to measure affiliative and
coercive relationships, or it may set a new standard for the measurement of dyadic relationships.
The following section addresses three remaining questions surrounding the measurement
of social dominance, resource control, and sociometric nominations by preschool children.
Teacher-rated social dominance and resource control. Aggression has been defined as
physical or verbal behaviors with the intention to cause harm towards another (e.g., Shantz,
1987). However, recent work has shown that not all aggression may have the same antecedent.
Dodge and Coie (1987) discussed the emergence of proactive and reactive aggression. The
former, which is of interest for the purpose of this dissertation, is that which results from a desire
to dominate or gain resources, such as nesting sites or territory for nesting sites (Dodge & Coie,
1987). Engaging in proactive aggression results from the attempt to gain accesses to desired
resources, and thus harm is a means to an ends, rather than indiscriminate (i.e., reactive)
aggression. Some forms of proactive aggression have been conceptualized as social dominance
and even social competence, in that the outcome is the acquisition of necessary resources, while
still maintaining positive relationships with others (i.e., reconciliation) (Bukowski, 2003;
Pellegrini, 2008). Based on this type of aggression, Dodge and Coie (1987) developed a reliable
measure of social dominance, measuring proactive aggression, as rated by adults.
While Dodge and Coie’s measure of social dominance focuses on proactive aggression,
ethologists (de Waal, 1986; Hinde, 1982; Tinbergen, 1963) suggested that social dominance was
not necessarily always aggressive in nature. Early on, dominance was conceptualized as a means
to access resources (e.g., food, mating) in the presence of others. Hawley’s Resource Control
Theory (Hawley, 1994) suggested that aggression (coercion) is one means to access resources,
but that affiliative behaviors (prosocial) can also be a means to gaining access to resources. This

14

theory goes a step further in suggesting function (e.g., resource control) over form of behavior
(e.g., aggression) (Hawley, 2007). Hawley (2003) developed a measure to determine social
dominance through the use of both prosocial and coercive strategies. Because Hawley’s theory
relies on multiple strategies, and specifically, a non-aggressive strategy, it has been presupposed
that Resource Control, as conceptualized by Hawley (1994) may be a different construct from
that of proactive aggression, as conceptualized by Dodge and Coie (1987). Yet research has yet
to establish if these two measures are related, and if so, to what degree. The current dissertation
used partial correlations to examine the relationship between the two social dominance measures.
Like traditional views of aggression, Dodge and Coie’s (1987) conceptualization of social
dominance is one of an individual level variable. That is, social dominance is thought to be a
characteristic of the child, and is measured as such (Dodge & Coie, 1987; Dodge, Coie, Pettit &
Price, 1990). However also like aggression, social dominance has more recently been
conceptualized as a group level variable, such that ones’ social dominance rank can only be
considered in comparison to others’ social dominance rank (e.g., Strayer, 1976). For instance, in
one setting, child B may be less dominant than child A, and more dominant than child C, giving
them a social dominance rank between those two peers. However, in another setting (such as
another classroom), child B’s social dominance status is dependent on others within that setting.
Although this view of social dominance is dependent on one’s dyadic relationships with
peers, studies have yet to measure social dominance in a way that represents this dyadic nature.
No previous work has both conceptualized and measured social dominance relationships from a
dyadic perspective. The current study adds to this literature by using multiple methods,
including direct observations, to examine how the nature of relationships among peers and social
dominance status impact affiliation and coercion. This is important because it’s the first study to

15

provide consistency between a dyadic conceptualization of social dominance and the methods
used to study it.
Sociometric nominations and observed play and conflict partners. A secondary
methodological question is exactly what is measured through sociometric nominations made by
preschool children. Sociometric procedures were created by Jacob Moreno (1934) to determine
individual’s preferences towards others and more recently have been defined as “a large class of
methods that assess the positive and negative links between persons within a group” (Cillessen,
2009). These methods are used frequently in modern peer relation literature to assess people’s
preference for members in a peer group (e.g., Cillessen & Borch, 2006; Hawley, 2002; Rodkin,
Farmer, Pearl & Van Acker, 2000; Roseth et al., 2011), and have been shown to be consistent
over time (Crick, Casas & Mosher, 1997). Procedurally, in preschool samples children are asked
to nominate peers that they “like to play with” and “do not like to play with” by pointing at
classmates pictures. At the dyadic level, sociometry enables researchers to better understand
friendship and antipathies and, at the group level, to create a continuum of acceptance and
rejected group members. Although this yields an index of acceptance within a group (Cillessen,
2007), research has cast doubts on whether or not children actually play with their preferred
playmates (Foot, Chapman & Smith, 1980; Strayer, Strayer & Chapeskie, 1980). Thus, this
dissertation contributes to the literature by using observations of actual play and conflict to
determine if children’s sociometric nominations capture actual dyadic relationships, or just
preference within a group.
Current Study
To summarize, the broad goal of the dissertation is to examine social dominance and the
strategies used by preschoolers to access and maintain control over resources in the context of

16

preschoolers’ peer relationships. At the individual level, it was predicted that social dominance,
and specifically, prosocial and bistrategic resource controlling strategies would be indicative of
the amount of time children spend in social and parallel play states. It was also predicted that
social dominance and resource control strategies would be associated with affiliation, desired
playmates and conflict with specific partners, as indicated by sociometric nominations.
Additionally, it was hypothesized that homophily in terms of social dominance status and
resource control strategies would predict affiliation and coercive relationships, and social
dominance and resource control scores would predict roles in dyadic conflict. Finally,
measurement questions concerning social dominance, resource control, and sociometric
measures are also examined.

17

METHOD
Preschoolers
The data for this dissertation were collected in seven classrooms located at four Head
Start centers in a Midwestern city in the United States from February 2009 through April 2009.
Classrooms I, II, and III met five mornings per week, and Classrooms IV, V, VI and VII met five
afternoons per week. Procedures associated with this study were reviewed and approved by the
sponsoring university's Institutional Review Board, IRB# 08-823D, and also by the Head Start
District Office and the individual classroom teachers. Permission slips were sent home by each
classroom teacher and children with signed permission slips participated in the study.
In all, 98% of preschoolers participated in the study, totaling 104 children (55 girls)
ranging in age from 41 to 73 months (M = 55.71, SD = 7.17). Classrooms included multi-aged
students. Across the entire sample, 39% of the children were African-American, 19% were
Hispanic, 20% European-American, and 2% Asian-American, and 20% represented multiple
ethnic groups. Admission into Head Start is based on income; families must meet state income
(low) eligibility requirements. The sample included one child on the autism spectrum; the child
with autism was not included in analyses. Four children left the school during data collection, six
children joined the school during the year. Only children who were present from the start until
the end of the study were used for analysis purposes.
Observational Procedures
Video recordings of freeplay were conducted every weekday the preschool was in session
during the eight week observational period by ten graduate and undergraduate student
researchers. Prior to entering classrooms the researchers participated in a training regimen that
involved videotape viewing and discussions. Before data collection began, student researchers

18

entered the classrooms and mock-recorded the students so that the children would get acclimated
to a researchers presence. Data collection began after two weeks of practice taping. Across the
semester, on average, each student was recorded six times. Numbers varied due to absences from
school.
Children were observed according to focal and event sampling procedures and
continuous recording rules. Due to the rare nature of coercive competitive events, event sampling
took precedence over focal observations. Each day observers entered a classroom with a predetermined, randomized list of children to videotape. All observations were conducted during
free-play periods lasting approximately one hour, and children were observed in their
classrooms, gymnasiums, and playgrounds. In all locations, researchers moved around the room
to film the focal child in a manner unobtrusive to the children. Reactivity was likely minimal as
the children became accustomed to the researchers’ presence.
Focal sampling/continuous recording rules (Pellegrini, 2004) governed focal
observations. Working from randomized lists, observers located individual children and recorded
behaviors in a 10-minute period. A variety of behaviors were recorded but those relevant to this
dissertation were the child’s play states. Play states of interest for this dissertation study were
social play (dyadic social interaction, either associatively or cooperatively), parallel play (focal
child is playing beside, but not with others), and solitary play (focal child plays alone and is not
affected by peers). The proportion of time in play states were derived from play in a particular
state (e.g., social play) divided by total amount of play regardless of state.
Event sampling with continuous recording rules (Pellegrini, 2008) were followed when
an observer saw a conflict event in which one child shows resistance (e.g., facial expression,
posture, tone of voice), physical aggression, or verbal aggression in opposition to the influence

19

attempt, anger, aggression, or argument of another child (Verbeek & de Waal, 2001; see also
Shantz, 1987). Thus, this study focused on conflict events involving both a resource-control
attempt and resistance to the resource-control attempt. When an agonistic event was detected
involving children other than the focal child, focal sample/continuous recording stopped and
observers shifted to event sampling. Thus, event sampling always took precedence over focal
samples. For each conflict event, the identity of the initiating child and the identity of the target
child were noted.
Video Coding
After video recording, videos were transferred onto computers for behavioral coding
using Mangold Interact, a video coding software program. Video coding provides many
advantages compared to live coding, such as the ability to take multiple passes, discrepancy
discussions to avoid observer drift, and blind interobserver checks (Yoder & Symons, 2010). A
team of four graduate and undergraduate students coded coercive competition behaviors and play
states as defined previously. If multiple play states occurred within a ten-second interval, the
play state that the child was predominantly engaged in (i.e., > 5.01 seconds) was coded. During
focal and event sampling, peers and teachers with whom the focal child was interacting with,
playing near, or observing were coded. Interrater agreement was established by double-coding a
randomly selected 10% of the focal observations (Κappa = .85). Any discrepancies during
coding were resolved through discussion.
Teacher Ratings
Children’s classroom teachers completed an adapted form of the Teacher Checklist
(Dodge & Coie, 1987), a 1-7 Likert-like scale shown to be a valid measure of social competence
(e.g., Pellegrini & Long, 2002; Pellegrini et al., 2007; Roseth et al., 2007). Children were rated

20

on a variety of subscales, but for the purpose of the proposed dissertation, only the five item
subscale related to social dominance (e.g., “dominates classmates”; Cronbach’s alpha = .94) was
used. Children’s classroom teachers also completed a measure adapted from Hawley’s (2003)
Resource Control Inventory, a 1-7 Likert to assess children’s resource control strategies.
Teachers were asked to complete an 18 item scale designed to assess the children’s relative
frequency of prosocial resource control (e.g., “”has plans that are liked by others”, “gets what
he/she wants by being nice”; Cronbach’s alpha = .80), coercive resource control (e.g., “gets what
he/she wants by bullying”, “gets what he/she wants by tricking others”; Cronbach’s alpha = .92)
and general resource control (e.g., “seems to win over peers”, “gets the best role in games with
peers”; Cronbach’s alpha = .91). Results were standardized within classrooms. The variable of
bistrategic resource control was derived from combining coercive and prosocial resource control
scores (Hawley, 2003b).
Sociometric Nominations
Sociometric interviews were conducted with individual children, with interview questions
based on a peer-nomination measure used in prior research with preschool-age children (Crick,
Casas & Mosher, 1997). Children were interviewed by one of the research associates.
Procedurally, peer nominations began with the adult holding up a picture board containing
individual pictures of each of the child’s classmates, including opposite-sex peers. Following
standard procedures, children were asked to point to three peers in response to the following
questions: Someone you like to play with; someone you don’t like to play with; and a special
friend. Each child’s individual scores were determined by summing the number of nominations
received from their peers, to get an index of peer preference.
Data Analysis

21

Descriptive statistics. Descriptive analyses were used to examine the relationship
between teacher rated social dominance and resource control, as well as the relationship between
social dominance and play states. Partial correlations were used in order to control for age and
sex, and percentages were used to examine the match between children’s actual sociometric
nominations (“like” and “dislike”) and actual play and conflict. Match is defined as consistency
between one preschooler’s nomination of a peer and observed play, or observed conflict with
that peer.
Social network analysis. Social network analysis was used to answer the questions
regarding the accuracy of sociometric nominations in relation to actual play/conflict and social
dominance scores, of who plays with whom, who has conflict with whom, and conflictual roles
between dyads. In general, social network models allow us to determine the relational
components that influence individual behavior (e.g., influence models) and affiliatory (or
coercive) relationships (e.g., selection models) (Frank, 2011). This dissertation focuses on
selection models by attempting to determine how individual characteristics impact the likelihood
of affiliative relationships between peers and the likelihood of coercive relationships between
peers. Important in the use of social network models is the recognition of both the individual
characteristics that impact affiliation or coercion, and the characteristics nested within the group
that impact affiliation or coercion (Frank, 2011). Level 1 analysis is that which examines the
relationship between the dyad in order to determine the likelihood of affiliation or coercion. For
instance, research has indicated that children of similar age are more likely to spend time
together (Rubin, Lynch, Coplan, Rose-Krasnor & Booth, 1994; for a review see Vaughn &
Santos, 2009). Thus, a child’s age only predicts affiliation when considering its relation to the
age of another peer. For the purpose of this dissertation, analysis of this kind is described as

22

being at the level of the dyad. Level 2 analyses are those that examine the characteristics of the
individual that might predict affiliation or coercion. For instance, research has indicated that
older children within a class are more likely to have social interactions with peers (Fabes, Martin
& Hanish, 2009). Thus a child’s age in this case may predict the likelihood of affiliation when
comparing an older child to a younger child. For the purpose of this dissertation, analysis of this
kind is described as being at the level of the individual. Classroom effects were controlled for by
including classrooms as dummy variables in these models.

23

RESULTS
Social Dominance and Resource Control
The first goal was to determine if the teacher-rated social dominance measure was
correlated with the teacher-rated resource control measure and its subscales, while controlling for
age (Table 1). Results indicated that the teacher-rated social dominance measure was highly
positively correlated with overall teacher-rated resource control (r = .87, p < .01), coercive
resource control (r = .72, p < .01), prosocial resource control (r = .58, p < .01), and bistrategic
resource control (r = .85, p < .01). As expected, results suggest that the two measures were
highly positively related.
Social Dominance and Play State
The study’s second goal was to determine the relationship between observed play states
(e.g., social play, parallel play, and solitary play) and teacher-rated social dominance, controlling
for age. As indicated by Table 1, teacher-rated social dominance was positively correlated with
observed social play (r = .25, p < .01), as was overall teacher-rated resource control (r = .18, p <
.05), prosocial resource control (r = .31, p < .01), and bistrategic resource control (r = .25, p <
.01). Interestingly, teacher-rated coercive resource control was not significantly correlated with
observed social play, and prosocial resource control was negatively correlated with observed
solitary play (r = -.17, p < .05). Results suggest that social dominance and, in particular,
prosocial and bistrategic forms of resource control were positively related to observed social play
with peers, while prosocial resource control was negatively related to observed solitary play.
Sociometric “Like” Nominations
The third goal was to examine correlations between children’s sociometric “like”
nominations and observed social play and teacher-rated social dominance.

24

Sociometric “like” nominations and observed play partners. The first analyses
examined the consistency between sociometric nominations of “like-most” (i.e., children
nominated as “like to play with”) and the actual time spent playing with their nominated peers.
Overall, there were 201 (67%) matches between nominated peers and actual play partners. Boys’
nominations matched 69% of the time, and girls’ nominations matched 62% of the time. Fiveyear-old children had a higher match between nominations and actual social play (69%),
followed by 4- (67%) and 3-year-old children (57%).
Sociometric “like” nominations, observed social play, and social dominance status.
The next analyses examined if matches between sociometric like-most nominations and actual
social play varied as a function of teacher-rated social dominance. The following two level
model was run to determine whether sociometric nominations predicted actual play in relation to
individual social dominance status:

LEVEL1:
TieSocialii ' = θ 0 + θ1ii ' (absDom) + θ 2ii ' (Similar − gender) +

θ 3ii ' ( Abs.Age) + vii '
LEVEL2 :

θ 0i = γ 00 + γ i01 (Female) + γ i02 (absage) + u0i
θ1i = γ 10
θ 2i = γ 20
The variables of actual play and sociometric nomination were analyzed in level 2, and social
dominance status in level 1, controlling for age and sex. In contrast to the hypothesis, results
showed that children with higher social dominance scores were not significantly more likely to
match like-most nominations and actual play partners compared to other children (b = .28, p =

25

.27). Similar models were created to determine this relationship with prosocial resource control
and bistrategic resource control. Results did not prove to be different; that is, children with
higher social dominance scores were not significantly more likely to match like-most
nominations and actual play partners compared to other children (b = .21, p < .64; b = .17, p <
.45). Taken together and contrary to the hypothesized relations, results suggest that social
dominance, prosocial resource control, and bistrategic resource control were unrelated to the
likelihood of matching like-most sociometric nominations and actual social play.
Similarity in social dominance status and observed social play. The next set of
analyses examined whether similarity in social dominance scores predicted who played with
whom during social and parallel play. These analyses used the absolute difference of scores
between two children with regards to social dominance, prosocial resource control, coercive
resource control and bistrategic resource control to determine if there is a homophily effect
between scores and social and parallel play. Hierarchical Linear Modeling was used to estimate
the effects of the dyad (Level 1) and individual (Level 2) effects on observed social play and
parallel play (see Table 2). The model that was used is as follows:

26

LEVEL1:

p : socialplay wii '=1  
 log
 = θ + θ ( AbsDom) + θ (Similar − gender) +
2ii '

 0 1ii '


1− p wii ' 


θ 3ii ' ( Abs.Age) + θ 4ii ' ( AbsPRC) + θ5ii ' ( AbsCRC) + θ 6ii ' ( AbsBRC)
LEVEL2 :

θ 0i = γ 00 + γ i01 (Female) + γ i02 (cenage) + γ i03 (Dom) +
γ i04 (PRC ) + γ i05 (CRC ) + γ i06 (BRC ) + u0i
θ1i = γ 10
θ 2i = γ 20
θ 3i = γ 30
θ 4i = γ 40
θ5i = γ 50
θ 6i = γ 60
Results did not show evidence of a statistically significant relationship between social
dominance or resource control and likelihood of actual play. However, same sex (b = 2.7, p <
.01) and similar age (b = -.12, p < .01) predicted social play. That is, children of the same sex
and similar in age were more likely to play together. At the individual level, results showed that
children with higher social dominance scores within their class engaged in a higher rate of social
play between pairs (b = .6, p = .04).
There was no evidence that the likelihood of parallel play between pairs was associated
with social dominance or resource control. In fact, only same sex (b = 5.02, p < .01), and
similarity in age (b = -0.12, p = .04) were predictive of dyadic-level parallel play. At the

27

individual level, results showed that children with lower bistrategic resource control scores
within their class engaged in higher rates of parallel play (b = -6.44, p = .01).
Sociometric Dislike Nominations and Conflict
The first goal was to examine consistency between sociometric nominations of peer
dislike (i.e., children nominated as “I don’t like to play with”) and observed conflictual bouts.
Overall, there were only 62 matches (21%) between nomination and actual conflict. Boys had a
match between conflict and negative nominations 25% of the time, and girls had nomination and
conflict match 18% of the time. Three-year-old children had a higher match between
nominations and actual conflict (30%), followed by 4-year-old children (19%) and then 5-yearold children (19%).
The next set of analyses examined whether matches between sociometric nominations of
peer dislike and actual pair conflict varied as a function of social dominance status (Table 3). A
two level model was run to determine whether sociometric nominations predicted actual conflict
in relation to individual social dominance status. The variables of actual conflict and sociometric
nomination were analyzed in level 2, and social dominance status in level 1, controlling for age
and sex. However, results showed that children in with higher social dominance scores were not
more consistent in nominating children with whom they had previous conflicts as disliked
playmates. This model also considered coercive resource control to determine consistency
between sociometric nominations and actual conflict. Results indicated that there was no
difference between the children’s likelihood of matching like-least nominations and actual
conflict (F (1, 95) = .147, p = .70). These finding suggests that social dominance and coercive
resource control were unrelated to consistency between sociometric nominations of ‘like-least’
and observed conflict.

28

Similarity in Social Dominance Status and Observed Conflict
The fifth goal was to examine whether similarity in social dominance status was
associated with the frequency of observed conflictual bouts. The same procedures used for social
dominance status and observed social play were used in this analysis to determine the
relationship between dyadic variables. HLM analysis was used, utilizing the following model:

LEVEL1:

p : conflictualbout  wii '=1  
 log
 = θ + θ ( AbsDom) +
0
1ii '




1− p  wii ' 



θ 2ii ' (Similar − gender) + θ 3ii ' ( Abs.Age) + θ 4ii ' ( AbsPRC) +
θ5ii ' ( AbsCRC ) + θ 6ii ' ( AbsBRC) + θ 7ii ' (SocialPlay) + θ8ii ' (ParaPlay)
LEVEL2 :

θ 0i = γ 00 + γ i01 (Female) + γ i02 (cenage) + γ i03 (Dom) + γ i04 (PRC) +
γ i05 (CRC ) + γ i06 (BRC ) + u0i
θ1i = γ10
θ 2i = γ 20
θ 3i = γ 30
θ 4i = γ 40
θ5i = γ 50
θ 6i = γ 60
Dyadic level analysis only provided evidence for one relationship between conflict and
the predictors in the model. Specifically, children who spent more time playing together during
social play were more likely to have dyadic conflict (b = .01, p < .01). Individual level analysis
showed that children who were younger (compared to the mean – grand mean centered) (b = -

29

.01, p = .051) and male (b = -.11, p = .052) were marginally more likely to be involved in
conflicts. Children that had higher coercive resource control scores (within classrooms) (b = .18,
p < .05) were more likely to be involved in conflicts. Additionally, children with lower
bistrategic resource control scores were more likely to get involved in conflicts as well (b = -.26,
p = .02).
Social Dominance Status and Conflict Role
The final set of analyses examined whether social dominance status predicts the role of
the initiator and target in observed conflict. HLM modeling was used to predict these
relationships.
Initiator. Table 4 presents the results of the role of initiator and relationship history
between dyads. Dyadic analysis suggested that when a conflict occurs between with a peer with
whom a child frequently engages in social play, that child is likely to be the initiator (b = .01, p <
.01). Individual level analysis suggested that children who were younger (compared to the mean)
(b = -.01, p = .05), males (marginally, b = -.06, p = .10) and children who have higher social
dominance scores (within classrooms) (b = .01, p = .02) were more likely to be initiators of
conflict.
Target. Table 4 also presents the results of the role of target and relationship history
between dyads. Dyadic level results suggest that there is a significant effect for conflict role as
target and social play. That is, when a child conflicts with a partner with whom they frequently
engage in social play, the focal child is more likely to be the target (b = .01, p < .01). At the
individual level, results suggested that males (b = -.06, p = .05) and younger children
(marginally, b = -.00, p = .06) were more likely to be targets of conflict. Additionally children

30

with higher social dominance scores (within classrooms) were more likely to be targets of
conflict (b = .01, p < .01).

31

DISCUSSION
The purpose of this study was to examine how social dominance and resource control
strategies used by preschool children impact affiliative and coercive interactions and
relationships.
Teacher-rated Social Dominance and Resource Control
The first question of this study was to determine if two measures, both attempting to
measure social dominance, were related. Results indicated strong partial correlations between
Dodge & Coie’s (1987) measure of social dominance and Hawley’s (2003) measures of coercive
resource control, prosocial resource control and bistrategic resource control. This is the first
known work to attempt to correlate these two differing measures of social dominance and the
results support the view that the two measures are highly related. Dodge and Coie’s (1987)
conceptualization of social dominance is one of proactive aggression. That is, social dominance
is an aggressive strategy that is a means to an end to gain access to desired resources. Hawley’s
(1999) conceptualization of social dominance is one of multiple strategies (cf. one single,
aggressive strategy) that, when used tactically, can be predictive of social competence. This is of
particular interest, because while Dodge and Coie’s (1987) measure of social dominance
primarily relies on aggressive, albeit proactive, behaviors Hawley’s measure uses multiple
strategies, one of which is affiliative in nature (i.e., prosocial resource control). Past work has
relied on these differing conceptualizations of social dominance as well. For instance, studies
conducted by LaFreniere and Charlesworth (1983) and Strayer (1989; 1992) show the
relationship between affiliation and social dominance through a lens of aggressive strategies,
whereas multiple Hawley studies (e.g., 1999, 2002, 2007) suggest that social dominance needs to
be understood as not just aggressive strategy use, but as a blend of multiple strategies. The high

32

correlations between these two measures provide support for the notion that function of the social
dominance behaviors (e.g., to gain access to resources) rather than form (e.g., actual type of
behavior) is key in understanding social dominance as a whole.
Social Dominance and Play States
The second goal of this study was to determine the relationship between social
dominance and observed social play, parallel play and solitary play. Results indicated that social
dominance, prosocial resource control and bistrategic resource control were positively correlated
with social play, and that prosocial resource control was negatively correlated with solitary play.
This is of particular importance, because although past studies have illustrated that children who
are socially dominant are preferred by peers, and have access to more resources, no known work
has studied this relationship using video taped observational data and both the Dodge and Coie
and Hawley measures of social dominance. These findings provide support for previous results
indicating that social dominance is associated with higher rates of affiliation in preschool
populations (LaFreniere & Charlesworth, 1983; Strayer, 1989; 1992), and that children who use
multiple strategies (e.g., coercive and prosocial) are preferred playmates of peers (Roseth et al.,
2011). This provides further support for the affiliative nature of certain social dominance
strategies, as well as suggests that social dominance as measured by Dodge and Coie may
actually indicate children who are using multiple strategies, rather than only more aggressive
strategies. Taken together, this growing body of work supports the notion that peer affiliation is
associated with the use of multiple resource controlling strategies, perhaps using aggressive
strategies to gain resources and affiliative strategies to maintain and mend relationships
(Pellegrini, 2008; Roseth et al., 2011). Of course, it is not known whether bistrategic and
prosocial strategies lend themselves to greater social play opportunities, or if children who

33

engage in greater rates of social play develop these strategies as a result of social play
experiences. Future research should continue to examine how these processes interact
developmentally to inform affiliative relationships and behaviors within those relationships.
Sociometric “Like” Nominations
The next goal of this study was to examine correlations between sociometric “like”
nominations, observed social play, and social dominance. Results indicated that approximately
two-thirds of preschoolers nominated children with whom they play as preferred peers, and this
was consistent regardless of the social dominance status of the nominating children. Findings did
not support the hypothesis that socially dominant peers may spend more time with preferred
peers compared to less socially dominant children. However, the results suggest that most
children tend to nominate children with whom they actually play. Sociometric measures are used
throughout peer relations as an indicator or preference for other peers from preschool through
adolescence (Cillessen, 2009). And although past work has illustrated that sociometric
nominations are consistent over time (Crick, Casas & Moser, 1997), it had not yet been
confirmed that nominations represent actual affiliation. The results of the current study are
important to the field because they provide evidence that children as young as preschool age can
reliably use sociometric nominations to indicate with whom they play.
Similarity in Social Dominance States and Observed Social Play
Another goal of this study was to determine whether social dominance status impacted
who plays with whom during social and parallel play using social network analysis. These results
add to the literature by providing the first evidence as to how the nature of the dyadic
relationship influences social play. While social network analysis has been used to determine
affiliation in peer relationships in older children (Huitsign & Veenstra, 2012; Mercer &

34

DeRosier, 2010; Snijders et al., 2007; Veenstra & Stegich, 2001), this is the first known work to
use the analytic technique to describe the relationship between preschool children’s affiliative
relationships based on social dominance and resource controlling behaviors. More specifically,
this is the first known work to examine how social dominance and resource controlling strategies
relate to affiliation. Past work has shown that social dominance is associated with increased peer
regard (Roseth et al., 2011), and affiliation (LaFreniere & Charlesworth, 1983), however no
work has examined this relationship through the use of social network analysis and examining
multiple strategies (i.e., coercive, bistrategic, prosocial resource control). This is especially
important when understanding the trajectory of conceptualization of social dominance. Early
work assumed social dominance to be an individual level variable, reflecting characteristics of
the individual. More recent work, however has shown that social dominance is dependent on the
context, such that one’s dominance status is relative to other’s within a group. In spite of this
new understanding, work has continued to measure social dominance as an individual level
variable. However this study is the first study known to examine this group level variable with a
group level analysis.
Results of the study, not surprisingly, indicated that same sex and similar age children
were more likely to affiliate at the dyadic level during social play and parallel play. This is
consistent with research on homophilies in preschool, suggesting that affiliation in preschool
often results from similarity of surface features such as sex, race and age (Epstein, 1989; also see
Vaughn & Santos, 2009).
Results also indicated that several individual behavioral characteristics were predictive of
social and parallel play (rather than dyadic level variables). Children with higher social
dominance status within their class were more likely to engage in social play, while children with

35

lower bistrategic resource control scores were more likely to engage in parallel play. Children
who are high on social dominance within classrooms are typically well likely by peers (e.g.,
Hawley 2002; Hawley & Little, 1999; Roseth et al., 2011; Strayer & Noel, 1986) and social
dominance has also been shown to be positively related to social competence (Vaughn & Waters,
1981). Therefore these children may be engaging in greater rates of social play as a result of their
elevated levels of attraction to their peers in concert with the increased social competence it takes
to maintain social interactions (compared to parallel or solitary play). Conversely, the use of
coercion, especially after social structures such as social dominance hierarchies have already
been formed, tends to be disliked by peers (Roseth et al., 2011), so these children may have less
affiliative opportunities and thus find themselves in parallel play situations rather than social
play due to their resource control strategy of choice.
Results also indicated that children with low scores on bistrategic resource control were
more likely to play in parallel play compared to peers. Using bistrategic resource control as a
strategy is associated with high levels of liking by peers (e.g., Roseth et al., 2010), and requires
the use of multiple strategies to gain access to resources, possibly indicative of social
competence. If a child is not particularly socially skilled in terms of resource control, he/she may
not have as many opportunities to play in social play, and may find themselves on the outside,
looking in. This may be while children low on bistrategic resource control find themselves
around, but not playing with peers. Future research should examine this relationship in more
depth however. For instance, it can be hypothesized that the current study’s results are due to a
deficit in social skills and the potential unattractiveness of low bistrategic resource controlling
playmates, but next steps need to examine the relationship between the use of strategies and

36

social competence, as well as children’s like or dislike of low bistrategic resource controlling
peers.
Sociometric “Dislike” Nominations
In terms of conflict, this study measured the relationship between sociometric “dislike”
nominations and actual conflict between the nominator and the nominated. Results showed only
21% consistency between the nomination and actual conflict. This suggests that something other
than actual conflict between and individual child and a particular peer accounts for dislike. In
this study, conflict was conceptualized as a competition for limited resources (Chen, Fesin &
Tam, 2001; Hawley, 2002; Laursen & Hartup, 1997), and conflict does not always result in long
term negative outcomes (e.g., reconciliation). It may be the case that conflict is not always a
predictor of dislike then, because the nature of conflict may involve the use of multiple strategies
(i.e. bistrategic resource control), resulting in a struggle for resources, but also reconciliation
(Roseth et al., 2011). Additionally the current results indicated that a predictor of conflict was
more time spent together in social play. Consistent with the relational model, children are having
conflict with those whom they are around more frequently. Thus, conflict is not an indicator of
dislike, but possibly an indicator of affiliation, and perhaps even preference. One question that
remains is what do negative nominations represent, if they do not represent conflict. It may be
the case that negative nominations are more associated with coercive behaviors, or a lack of
frequent interactions. Next steps for research include determining what, if any, types of
interactions negatively nominated peers have with their nominators.
Conflict also occurs less frequently after social dominance hierarchies are formed
(Pellegrini et al., 2007; Roseth et al., 2007; Roseth et al., 2011). It may also be the case that
conflict is infrequent enough after social dominance hierarchies have been formed that

37

consistency between dislike nominations and conflict is harder to measure. Coercive resource
control however, is predictive of dislike especially after social dominance hierarchies have
formed, and more consistency may be seen between coercive behaviors and dislike nominations.
Consistency between dislike nominations and actual conflict did not increase when using social
dominance status as a predictor. Children with high social dominance scores were no better at
nominating a match between disliked and actual conflict. This may be more support for the
notion that disliked peers are not disliked due to dyadic conflict per se, but for another reason,
such as coercive behaviors, the absence of prosocial behaviors, or the lack of post-conflict
reconciliation (Roseth et al., 2011).
Similarity in Social Dominance and Observed Conflict
Another goal of this study was to determine whether social dominance status impacted
who conflicts with whom using social network analysis. Results suggested that conflict is more
frequent between children who engage in high amounts of social play together, and more
frequent between children of the same sex. This findings provide support for de Waal’s (2000)
Relational Model, which suggests that conflict occurs when there are greater opportunities to
have aggressive interactions and/or greater opportunities for reconciliation. Results of this study
provided evidence for a homophily effect in terms of affiliation. That is, children who are more
similar with respect to sex and age, are more likely to engage in social play with one another.
When children engage in social play, they have greater opportunities to compete for desired
resources (e.g., toys, themes of play), which may result in conflict. Additionally if peers engage
in greater rates of social play with one another, their dyadic relationship is primarily affiliative in
nature, and thus opportunities for reconciliation may be more likely. Future research should

38

investigate post-conflict reconciliation tendencies and use social network analysis to determine
how reconciliation impacts affiliation and conflict in dyadic relationships.
Consistent with previous research, results indicated that younger children and boys were
more likely to engage in dyadic conflict, compared to older children and girls (Clark-Lempers,
Lempers & Ho, 1991). Sex differences were not altogether clear, however, due to reasons such
as small sample sizes, making differences difficult to detect (Hay, Caplan & Nash, 2009). Some
research does suggest though that children’s socialization may impact the degree to which they
are involved in conflict. For instance, girls’ aggression tends to be ignored more than boys, and
girls tend to relinquish objects of dispute more quickly than boys (Fagot & Hagan, 1985; Ross,
Tesla, Kenyon & Lollis, 1990). This may result in fewer conflicts due to ignored aggression and
the lack of persistence in contests over resources.
At the individual level, results also indicated that children with high coercive resource
controlling scores were also more likely to engage in dyadic conflict. Coercive resource control
measures resource control attempts that are aggressive in nature. Using aggressive behaviors
alone (without the use of prosocial behaviors) is not associated with acceptance especially after
social dominance hierarchies are established (Roseth et al., 2011). Additionally, children who are
low on acceptance ratings tend to engage in increased rates of conflict (Putzllaz, Hellstern,
Sheppard, Grimes & Flodis, 1995). It may be the case then that children who are engaging in
coercive resource controlling strategies are engaging in behaviors that are disliked by peers (i.e.,
proactive aggression), are disliked by peers, and are engaging in greater conflict in order to
attempt to gain access to desired resources. This pattern may be cyclical, in that the aggressive
resource controlling attempts correlate with continue dislike by peers, and this dislike by peers
resulting in access to fewer resources, causing a need for continued frustration and aggressive

39

attempts at resource control. Future research is needed however to determine if this relationship
is causal, or if these variables are moderating the relationship between coercive resource control
and conflict.
Lastly, at the individual level, children with low bistrategic resource controlling scores
were more likely to engage in dyadic conflict. According to Resource Control Theory, the ability
to use multiple strategies in different contexts may be suggestive of social competence
(Bukowski, 2003; Hawley, 2002; Hawley, 2007; Underwood, 2003; Vaughn, 1999).
Additionally, numerous studies have indicated that conflict decreases after social dominance
hierarchies form within a group (Pellegrini et al., 2007; Roseth et al., 2007; Roseth et al., 2011).
Children engaging in low rates of bistrategic resource control may be low in measures of social
competence, which could potentially be evident by increased rates of dyadic conflict (compared
to peers) after social dominance hierarchies have formed, consistent with the data in the current
study, which was taken in the last months of a year round preschool program.
Social Dominance Status and Conflict Role
The final goal of this dissertation was to use social network analysis to examine how
social dominance may impact one’s role within conflict (e.g., initiator or target of conflict).
Results did not reveal any significant result suggesting that social dominance or resource control
strategies do not predict one’s role in dyadic conflict. This is surprising because research has
suggested that conflict late in the term (i.e., after social dominance hierarchies are stable) is
disliked, and thus it was hypothesized that those initiating conflict may be disliked, and not
particularly strategic. Additionally, individual who are considered to be socially dominant and
strategic resource controllers were thought to be attractive targets for conflict, given their control
over desired resources. However the results of the current study do not support these hypotheses.

40

It may be the case that the infrequent nature of conflict made these constructs difficult to capture.
Future research needs to investigate how roles in conflict are influenced by status.
At the individual levels, results indicated that younger children, boys and children with
high social dominance scores were more likely to initiate and be the targets of conflict. This may
suggest that the lack of results using social network analysis to determine conflict role may be
the result of the general composition of those engaging in conflict. That is, the profile for
initiators and targets of conflict are similar (i.e., boys, younger, and high on social dominance
scores), and because of this, dyadic analysis may not show any differences in role due to the
similarity in characteristics of the individuals engaging in conflict.
Practical Implications
There are several practical implications related to the results of this study. First, in a
general sense, this study provides a window into the development of dyadic peer relationships,
based on interactions, both affiliative and coercive in nature.. Understanding how and why
children form relationships with peers is important because it sheds light on the beginning stages
of peer relationships. At the individual level, this work helps make teachers and parents aware of
potential correlates and behavioral modifications (e.g., resource controlling strategies, amount of
social play with particular peers) associated with different types of play, the frequency of peer
conflict, and between whom peer conflict occurs. We know that play is important and develops a
variety of strategies and these may be the characteristics is related to increase in play
opportunities.
Secondly, this study’s results provide evidence that children do indeed have preferred
playmates and do actually play with those playmates. When facilitating interactions in
classrooms and in play situations, teachers and parents should be sensitive to the child’s intent in

41

picking playmates, and also provide opportunities for socialization with other peers, as children
may continue to play with a particular child because he/she likes them but may shut out other
opportunities because they do have a preferred playmate.
Third, this study suggests a need for parents and teachers to understand the reasons for
conflict among peers, recognizing that they are not necessarily negative or maladaptive. Given
that conflict occurs between playmates, and negative nominations do not relate to conflict, we
must consider, what are the benefits of conflict? Peer relationships are likely benefiting from
these interactions, such as staying together after conflict, and reconciling. Thus, allowing
children to have conflict, and resolve conflict together rather than with intervention is likely
aiding in the development of important social and emotional skills.
Finally, similar to understanding that conflict is not necessarily a negative type of
interaction, and may in fact be beneficial, it is also important to recognize the reason or
developmental purpose behind different strategies used in resource control. So, for example,
aggressive toy taking may be followed by reconciliation behaviors, which ultimately may
alleviate some of the negative association with the coercive interaction. As teachers and parents,
it is important to recognize the interaction (e.g., snatching a toy) nested within the relationship
(e.g., friendship with reconciled).
Limitations and Conclusions
There are several limitations that should be acknowledged when interpreting the results
of this study. First, the data for this study was collected in an eight-week period in the spring
term of a year round preschool program. The time frame resulted in each child being observed
approximately six times each, for a 10-minute period, or approximately 60 minutes over the

42

course of eight weeks. Increasing the amount of time each child is observed would likely provide
a richer picture of the nature of dyadic relationships between peers.
A second limitation is the generalizability of the results due to the time in which data
collection took place. The current study collected data during the end of the spring term of a year
round preschool program. Previous research has shown that resource controlling strategies vary
is use and acceptance during a school term (Roseth et al., 2011), and that conflict fluctuates in
frequency over time (Pellegrini et al., 2007; Roseth et al., 2007; Roseth et al., 2011). The results
may illustrate the nature of dyadic relationships and social dominance for a static period of time,
but might not be generalizable in terms social play partners, social dominance and resource
controlling strategies, and conflict roles, social dominance and resource controlling strategies
over the course of the school year. For instance, this study’s results found that the roles of
conflicting dyads (i.e., initiator and target) are similar in respect to social dominance and
resource controlling strategies in the spring. However data collecting including the fall term may
show unique relationship dynamics due to the unstable social dominance hierarchy early in
group formation. Additionally research has suggested that certain strategies, such as coercive
resource control has been associated with peer preference in the fall term but then associated
with peer dislike in the spring term (Roseth et al., 2011). Future research should continue to use
social network analysis to capture the unique relationship between coercive and affiliative dyads
in preschool, but do so over a longer period of time to determine how already established
behavioral changes over time impact the nature of the dyadic relationship.
Finally the sample used in the current study may limit the generalizability of the results.
Head Start centers are unique in their demographics because their purpose is to provide
preschool for low-income populations. The homogeneity of the sample then, is considered to be

43

a limiting factor. For instance, this sample had approximately 300 conflicts over the course of an
eight-week period. This number may be large in comparison to data from other samples (Roseth
et al., 2011). The relationship between resource controlling strategy, social dominance and
conflict may vary as a function of the potentially more normative nature of conflict in this
particular sample. Future research needs to extend this work to a more diverse sample to increase
the generalizability of its results.
Despite its limitations, the study contributes to literature in several important ways. First,
this study provides evidence that Hawley’s (2002) measure of resource control strategies and
Dodge and Coie’s (1987) measure of social dominance are similar in construct. Although social
dominance tends to emphasize proactive aggression while resource control strategies
differentiate between coercive and prosocial forms, the two constructs were highly correlated.
This suggests that the underlying characteristics of being socially dominant and/or controlling of
resources are interrelated, and require multiple, related strategies. This provides a more complete
picture for the operationalization of the social dominance as defined by the controlling of desired
resources.
Second, this study provides evidence through social network analysis that preschool
children affiliate based on social dominance homophiles. Social dominance represents a variety
of behaviors that result in social dominance hierarchies, and this study shows that children of
similar social dominance status’ affiliate together. This provides insight as to group formation
and affiliation in preschool peer relations. Additionally, some of the behaviors captured by social
dominance measures and resource control measures are similar to those captured by perceived
popularity measures that are used in the elementary school years. The current study provides
evidence for affiliation based on these behaviors, and may be the first evidence of the formation

44

of groups based on characteristics related to perceived popularity. Future research should
examine the affiliative nature of children over time to determine if affiliation based on social
dominance status in preschoolers is correlated with affiliation based on perceived popularity in
early and middle childhood.
A third contribution of this study is providing evidence that actual dyadic conflict does
not equate to dislike among preschool children. Traditional views of conflict have equated the
aggressive nature of resource competition as unskilled and maladaptive (Brook & Newcomb,
1995; Bukowski, 2003; Coie & Dodge, 1998; Coie, Dodge & Kupersmidt, 1990). However
recent literature has suggested that resource control attempts are adaptive (for a review, see
Hawley, 2007) and that using difference strategies, such as resource controlling strategies and
reconciliation are associated with peer preference in preschool samples (Hawley, 2002; Hawley
& Little, 1999; Roseth et al., 2011). The results of this study provide further evidence that
engaging in conflict with peers does not necessarily result in a disliked peer status. This also
suggests that future research should continue to unpack the relationship between conflict and
peer acceptance, as well as behavioral correlates related to low status peers.

45

Table 1
Partial Correlations for Teacher-rated Dominance, Resource Control, and Observed Play

1. Social dominance score

2

3

4

5

6

7

8

.87**

.72**

.58**

.85**

.25**

-.08

-.07

.70**

.56**

.83**

.18*

-.05

-.05

.15**

.83**

.11

-.04

.02

.62**

.31**

-.14

-.17*

.25**

-.05

-.09

2. Resource control
3. Coercive resource control
4. Prosocial resource control
5. Bistrategic resource control
6. Social play (%)
7. Parallel play (%)
8. Solitary play (%)
* p < .05. ** p < .01.

46

Table 2
Relative Social Dominance, Resource Control, and Play Type
Independent Variables

Social Play

Parallel Play

Level 1
Social dominance

-0.58 (0.40)

0.12 (1.16)

Prosocial resource control

-0.16 (0.32)

1.15 (0.92)

Coercive resource control

-0.21 (0.39)

0.78 (1.11)

Bistrategic resource control

-0.52 (0.42)

-0.58 (1.45)

Same sex

2.71** (0.55)

5.02** (1.60)

Absolute age

-0.12** (0.05)

-0.19* (0.09)

Social dominance

0.60* (0.29)

0.11 (1.34)

Prosocial resource control

-0.31 (0.48)

2.38 (1.47)

Coercive resource control

-0.20 (0.57)

2.81 (2.04)

Bistrategic resource control

0.15 (0.76)

-0.11 (0.05)*

Sex

-0.41 (0.33)

-6.44 (2.48)

Age

-0.01 (0.02)

-0.07 (0.08)

3.17** (0.44)

11.75** (2.64)

Level 2

Intercept
Note. Standard errors are in parentheses.
* p < .05. ** p < .01.

47

Table 3
Relative Social Dominance, Resource Control, and Conflict

Independent Variables

Conflict

Level 1
Social dominance

-.04 (.03)

Prosocial resource control

-.01 (.02)

Coercive resource control

.02 (.03)

Bistrategic resource control

-.00 (.05)

Same sex

.07 (.03)

Absolute age

.00 (.00)

Social play partners

.01** (.00)

Parallel play partners

-.00 (.00)

Level 2
Social dominance

.07 (.04)

Prosocial resource control

.11 (.06)

Coercive resource control

.18* (.08)

Bistrategic resource control

-.26* (.11)

Sex

-.00 (.00)

Age

-.11 (.05)

Intercept

.50** (.12)

Note. Standard errors are in parentheses.
*p < .05
**p < .01

48

Table 4
Conflict Role as Initiator or Target and Relationship History

Independent Variables

Initiator

Target

Social dominance

-0.00 (0.00)

-0.00 (0.00)

Prosocial resource control

-0.01 (0.02)

-0.01 (0.02)

Coercive resource control

0.02 (0.02)

0.01 (0.03)

Bistrategic resource control

-0.00 (0.03)

-0.01(0.03)

Same sex

0.04 (0.03)

0.04 (0.03)

Absolute age

0.00 (0.00)

0.00 (0.00)

Social play partners

0.01** (0.00)

0.01** (0.00)

Parallel play partners

-0.00 (0.00)

-0.00 (0.33)

Social dominance

0.01* (0.00)

0.01** (0.00)

Prosocial resource control

-0.00 (01)

-0.00 (0.03)

Coercive resource control

0.02 (0.02)

-0.00 (0.05)

Bistrategic resource control

-0.00 (0.02)

-0.03 (0.06)

Sex

-0.06 (0.04)

-0.06* (0.03)

Age

-0.01* (0.00)

-0.00 (0.00)

0.18** (0.04)

0.20** (0.04)

Level 1

Level 2

Intercept
Note. Standard errors are in parentheses.
*p < .05
**p < .01

49

REFERENCES

50

REFERENCES
Blurton-Jones (1972). Categories of child-child interaction. In N. G. Blurton-Jones (Ed.),
Ethological studies of child behavior (pp. 97-127). Oxford, UK: Cambridge University
Press.
Brownell, C., & Brown, E. (1992). Peer and play in infants and toddlers. In V. B. Van Hasselt
(Eds.), Handbook of social development: A lifespan perspective (pp.183-200). New York:
Plenum Press.
Bugental, D. B., & Grusec, J. E. (2006). Socialization processes. In N. Eisenberg (Ed.),
Handbook of child psychology: Vol 3. Social, emotional and personality development, 6th
ed. (366--428). New York: Wiley.
Bukowski, W. M. (2003). What does it mean to say that aggressive children are competent or
incompetent? Merrill-Palmer Quarterly, 49, 390-400.
Charlesworth, W. R. (1996). Co-operation and competition: Contributions to an evolutionary and
developmental model. International Journal of Behavioral Development, 19, 25-39.
Chen, D, W., Fein, G., & Tam, H. P. (2001). Peer conflicts of preschool children: Issues,
resolution, incidence, and age-related patterns. Early Education and Development, 12,
523-544.
Cillessen, A. H. N. (2009). Sociometric Methods. In K. H. Rubin, W. M Bukowski & B. Laursen
(Eds.), Handbook of peer interactions, relationships and groups (pp. 143-179). New
York: Guilford Press.
Cillessen, A. H. N. & Mayeux, L. (2007). Variations in the Association between Aggression and
Social Status: Theoretical and Empirical Perspectives. In. P. H. Hawley, T. D. Little & R.
C. Rodkin (Eds.), 2007. Aggression and Adaptation: The Bright Side to Bad Behavior
(pp. 1-29). New York: Lawrence Erlbaum Associates.
Clark-Lempers, D. S., Lempers, J. D., & Ho, C. (1991). Early middle and late adolescents’
perceptions of their relationships with significant others. Journal of Adolescent Research,
6, 296-315.
Coie, J. D., & Dodge, K. A. (1998). Aggression and antisocial behavior. In W. Damon (Series
Ed.) & N. Eisenberg (Vol Ed.), Handbook of child psychology: Vol. 3. Social, emotional
and personality development (5th ed. Pp. 779-862). New York: Wiley.
Coplan, R. J. & Arbeau, K. A. (2009). Peer interactions and play in early childhood. In K. H.
Rubin, W. M Bukowski & B. Laursen (Eds.), Handbook of peer interactions,
relationships and groups (pp. 143-179). New York: Guilford Press.

51

Crick, N. R., Casas, J. F., & Mosher, M. (1997). Relational and overt aggression in preschool.
Developmental Psychology, 33, 579–587.
Darwin, C. (1859/1884). On the orgin of species by means of natural selection (6th ed.) New
York: Appleton & Co.
Dodge, K. A., & Coie, J. D. (1987). Social information processing factors in reactive and
proactive aggression in children’s peer groups. Journal of Personality and Social
Psychology, 53, 1146-1158.
de Waal, F. B. M. (1986). The integration of dominance and social bonding in primates.
Quarterly Review of Biology, 61, 459-479.
de Waal, F. B. M. (1996). Conflict as negotiation. In W. C. McGrew, L. F. Marchant, & T.
Nishida,, (Eds.). Great ape societies, (pp. 159-172). New York: Cambridge University
Press.
de Waal, F. B. M. (2000). The first kiss: Foundations in conflict resolution research in
animals. In F. Aureli & F.B.M. de Waal, (Eds). Natural conflict resolution, p. 34-53.
Berkeley, CA: University of California Press.
Ellis, S., Rogoff, B., & Cromer, C. C. (1981). Age segregation in children’s social interactions.
Developmental Psychology, 17, 399-407.
Epstein, J. L. (1989). The selection of friends. In T.J. Berndt & G. W. Ladd (Eds.), Peer
relationships in child development (pp. 158-187). New York: Wiley.
Fabes, R. A., Martin, C. L. & Hanish, L. D. (2003). Young children’s play qualities in same-,
other- and mixed-sex peer groups. Child Development, 74, 921-932.
Fabes, R. A., Martin, C. L. & Hanish, L. D. (2009). Children’s behaviors and interactions with
peers. Handbook of peer interactions, relationships and groups (pp. 45-62). New York:
Guilford Press.
Fagot, B. T., & Hagan, R. (1985). Aggression in toddlers: Responses to the assertive acts of boys
and girls. Sex Roles, 12, 341-351.
Foot, H. C., Chapman, A. J., & Smith, J. R. (1980). Friendship and social relations in children.
New York: Wiley.
Frank, K.A. 2011. Social Network Models for Natural Resource Use and Extraction. In Social
Networks and Natural Resource Management: Uncovering the Social Fabric of
Environmental Governance. Örjan Bodin & Christina Prell editors. Cambridge:
Cambridge University Press.
Gifford-Smith, M. E., & Brownell, C. A. (2003). Childhood peer relationships: Social
acceptance, friendship and peer networks. Journal of School Psychology, 41, 235-284.

52

Gottman, J. (1994). Why marriages succeed or fail. New York: Simon and Schuster.
Hawley, P. H. (1994). On being an elephant: A quantitative intraindividual analysis of the
behavior of Elephas maximus. Unpublished doctoral dissertation, University of
California, Riverside.
Hawley, P. (1999). The ontogenesis of social dominance: A strategy-based evolutionary
perspective. Developmental Review, 19, 97-132.
Hawley, P. (2002). Social dominance and prosocial and coercive strategies of resource control in
preschoolers. International Journal of Behavioral Development, 26 (2), 167-176.
Hawley, P (2003). Prosocial and coercive configurations of resource control in early
adolescence. Merrill-Palmer Quarterly, 49, 279-309.
Hawley, P. (2007). Social dominance in childhood and adolescence: Why social competence and
aggression may go hand in hang. In. P. H. Hawley, T. D. Little & R. C. Rodkin (Eds.),
2007. Aggression and Adaptation: The Bright Side to Bad Behavior (pp. 1-29). New
York: Lawrence Erlbaum Associates.
Hawley, P. H., & Little, T. D. (1999). On winning some and losing some: A social relations
approach to social dominance in toddlers. Merrill-Palmer Quarterly, 43, 185-214.
Hawley, P., Little, T., Pasupathi, (2002). Winning friends and influencing peers: Strategies of
peer influence in late childhood. International Journal of Behavioral Development, 26,
466-474.
Hay, D. F., Caplan, M., & Nash, A. (2009). The beginnings of peer relations. In K. H. Rubin, W.
M Bukowski & B. Laursen (Eds.), 2009. Handbook of peer interactions, relationships
and groups (pp. 180-194). New York: Guilford Press.
Hinde, R. (1979). Towards understanding relationships. New York: Academic.
Hinde, R. A. (1982). Ethology: Its nature and relations with other sciences. New York: Oxford.
Howes, C. (2009). Friendship in early childhood. In K. H. Rubin, W. M Bukowski & B. Laursen
(Eds.), 2009. Handbook of peer interactions, relationships and groups (pp. 180-194).
New York: Guilford Press.
Howes, C. & Matheson, C. C. (1992). Sequences in the development of competent play with
peers: Social and social-pretend play. Developmental Psychology, 28, 961-974.
Huitsing, G., van Duijn, M. A. J., Snijders, T. A. B., Wang, P., Sainio, M., Salmivalli, C., &
Veenstra, R. (2012). Univariate and multivariate models of positive and negative
networks: Liking, disliking and bully-victim relationships. Social Networks, 34 (4), 645657.

53

Huitsing, G. & Veenstra, R. (2012). Bullying in classrooms: Participant roles from a social
network perspective. Aggressive Behavior, 38 (6), 494-509.
LaFreniere, P. J., & Charlesworth, W. R. (1983). Dominance, attention and affiliation in a
preschool group: A nine-month longitudinal study. Ethology and Sociobiology, 4, 55-67.
Laursen, B., & Hartup, W. W. (1997). The dynamics of preschool children’s conflicts. MerrillPalmer Quarterly, 35, 281-297.
Laursen, B., & Pursell, G. R. (2009). Conflict in peer relationships. In K. H. Rubin, W.
M. Bukowski, & B. Laursen (Eds.), Handbook of peer interactions, relationships, and
groups (pp. 267-286). New York: Guilford.
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social
networks. Annual Review of Sociology, 27, 415-444.
Mercer, S. H. & DeRoiser, M. (2010). Selection and socialization of internalizing problems in middle
childhood. Journal of Social and Clinical Psychology, 29, 1032-1059.

Pellegrini, A. D. (2004). Observing children in the natural worlds: A methodological primer,
2nd ed. Mahwah, NJ: Erlbaum.
Pellegrini, A. D. (2008). The roles of aggressive and affiliative behaviors in resource control: A
behavioral ecological perspective. Developmental Review, 28, 461–487.
Pellegrini, A. D., & Bartini, M. (2001). Dominance in early adolescent boys: Affiliative and
aggressive dimensions and possible functions. Merrill-Palmer Quarterly, 47, 142–163.
Pellegrini, A. D., & Long, J. D. (2002). A longitudinal study of bullying, dominance, and
victimization during the transition from primary to secondary school. British Journal of
Developmental Psychology, 20, 259–280.
Pellegrini, A. D., Roseth, C. J., Mliner, S., Bohn, C. M., Van Ryzin, M., Vance, N., et al. (2007).
Social dominance in preschool classrooms. Journal of Comparative Psychology, 121, 54–
64.
Pianta, R.C. (1998). Enhancing relationships between children and teachers. Washington DC:
American Psychological Association.
Pusey, A. E., & Packer, C. (1997). The ecology of relationships. In J. R. Krebs & N. B. Davies
(Eds.), Behavioral ecology: An evolutionary approach (4th ed., pp254-283). Malden, MA:
Blackwell.
Putalla, M., Hellstern, L., Sheppard, B. H., Grimes, C. L., & Glodis, K. A. (1995). Conflict,
social competence and gender: Maternal and peer conflicts. Early education and
Development, 6, 433-447.
54

Roseth, C. J., Pellegrini, A. D., Bohn, C. M., Van Ryzin, M., & Vance, N. (2007). An
observational, longitudinal study of preschool dominance and rates of social behavior.
Journal of School Psychology, 45, 479-497.
Roseth, C. J., Pellegrini, A. D., Dupuis, D. N., Bohn, C. M., Hickey, M. C., Hilk, C.L. &
Peshkam, A. (2011). Preschoolers’ bistrategic resource control, reconciliation and peer
regard. Social Development, 20, 185-211.
Ross, H., Tesla, C., Kenyon, B., & Lollis, S. (1990). Maternal intervention in toddler peer
conflict: The socialization of principles of justice. Developmental Psychology, 26, 9941003.
Rubin, K. H., Bukowski, W., & Parker, J. G. (2006). Peer interactions, relationships and groups.
In N. Eisenberg (Ed.), Handbook of child psychology: Vol 3. Social, emotional and
personality development, 6th ed. (571-645). New York: Wiley.
Rubin, K. H., Lynch, D., Coplan, R. J., Rose-Krasnor, L., & Booth, C. L. (1994). “Birds of a
feather…”: Behavioral concordances and preferential personal attraction in children.
Child Development, 65, 1778-1785.
Shantz, C. U. (1987). Conflicts between children. Child Development, 58, 283-305.
Shrum, W., Cheek, N. H., & Hunter, S. M. (1988). Friendship in school: Gender and racial
homophily. Sociology of Education, 61, 227-239.
Snijders, T., Steglich, C., & Schweinberger, M. (2007). Modeling the coevolution of networks
and behavior. In K. Van Monfort, J. Oud, & A. Satorra (Eds.), Longitudinal models in the
behavioral and related sciences (pp. 41-71). Mahwah, NJ: Erlbaum.
Strayer, F. F. (1989). Co-adaption within the early peer group: A psychobiological study of
social competence. In B. H. Schneiger, G. Attili, J. Nadel & R. P. Weisenberg (Eds.),
Social Competence in Developmental Perspectives, (145-174). The Netherlands: Kluwer
Academic Publishers.
Strayer, F. F. (1992). The development of agonistic and affiliative structures in preschool play
groups. In J. Silverberg & J. P. Gray (Eds.), Aggression and peacefulness in humans and
other primates (pp. 150-171). New York: Oxford.
Strayer, F. F., & Noel, J. M. (1986). The prosocial and antisocial functions of preschool
aggression: An ethological study of triadic conflict among young children. In C. ZahnWaxler, E. M. Cummings, & R. Iannotti (Eds.), Altruism and aggression: Biological and
social origins (pp. 107-134). New York: Cambridge University Press.
Strayer, F. F., & Strayer, J. (1976). An ethological analysis of social agonism and dominance
relations among preschool children. Child Development, 47, 980-989.

55

Strayer, FF., Strayer, J., & Chapeskie, T. R. (1980). The perception of social power relations
among preschool children. In D. R. Omark, F. F. Strayer & D. Freeman (Eds.),
Dominance Relations (pp. 191-209), New York: Garland.
Tinbergen, N. (1963). On aims and methods of ethology. Zeitschrift fur Tierpsychologie, 20,
410-433.
Underwood, M. K. (2003). The comity of modest manipulation, the importance of distinguishing
bad behaviors. Merrill-Palmer Quarterly, 49, 373-389.
Vaughn, B. E. (1999). Power is knowledge (and vice versa): A commentary on “On winning
some and losing some: A social relations approach to social dominance in toddlers.”
Merrill-Palmer Quarterly, 45, 215-225.
Vaughn, B. E., & Santos, A. J. (2009). Structural descriptions of social transactions among
young children: Affiliation and dominance in preschool groups. In K. H. Rubin, W. M
Bukowski & B. Laursen (Eds.), 2009. Handbook of peer interactions, relationships and
groups (pp. 195-216). New York: Guilford Press.
Vaughn , B. E., Vollenweider, M., Bost, K. K., Azria-Evans, & Snider, J. B. (2003). Negative
interactions and social competence for preschool children in two samples: Reconsidering
the interpretation of aggressive behavior for young children. Merrill-Palmer Quarterly,
49, 245-278.
Vaughn, B. E., & Waters, E. (1981). Attention structure, sociometric status and dominance:
Interrelations, behavior correlates and relationships to social competence. Developmental
Psychology, 17, 275-288.
Veenstra, R. & Steglich, C (2011). Actor-based model for network and behavior dynamics: A tool for
examining selection and influence processes. In B. Laursen T. D. Little, & N. A. Card (Eds.),
Handbook of developmental research methods. New York: Guilford.

Verbeek P., & de Waal, F. B. M. (2001). Peacemaking among preschool children. Journal of
Peach Psychology, 7, 5-28.
Wasserman, S. & Faust, K. (1994). Social network analysis: Methods and applications. New
York: Cambridge University Press.
Yoder, P., & Symons, F. (2010). Observational measurement of behavior. New York: Springer.

56