Journal of Adolescence 53 (2016) 75e90

Contents lists available at ScienceDirect

Journal of Adolescence
journal homepage: www.elsevier.com/locate/jado

Early adolescents' motivations to defend victims in school

bullying and their perceptions of studenteteacher
relationships: A self-determination theory approach
Tomas Jungert a, *, Barbara Piroddi b, Robert Thornberg c
a
Lund University, Sweden
b
University of Turin, Italy
c €ping University, Sweden
Linko

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this study was to investigate whether various dimensions of studenteteacher
Available online 18 September 2016 relationships were associated with different types of motivation to defend victims in
bullying and to determine the association between these types of motivations and various
Keywords: bystander behaviors in bullying situations among early adolescents in Italy. Data were
Bullying collected from 405 Italian adolescents who completed a survey in their classroom. Results
Bystander
showed that warm studenteteacher relationships were positively associated with
Defender
defending victims and with autonomous motivation to defend victims. In contrast,
Studenteteacher relationships
Motivation to defend
conflictual studenteteacher relationships were positively associated with passive
bystanding and with extrinsic motivation to defend victims. Different forms of motivation
to defend were found to be mediators between studenteteacher relationship qualities and
bystander behaviors in school bullying. Our findings suggest that teachers should build
warm and caring studenteteacher relationships to enhance students' autonomous moti-
vation to defend victims of bullying as well as their inclination to defend the victims in
practice.
© 2016 The Foundation for Professionals in Services for Adolescents. Published by Elsevier
Ltd. All rights reserved.

Bullying refers to repeated aggression directed at students who are disadvantaged or less powerful in their interactions
with those who bully (Jimerson, Swearer, & Espelage, 2010; Olweus, 1993), and it is a pervasive problem in schools
throughout the world (e.g., Harel-Fisch, Walsh, Fogel-Grinvald, De Matos, & Craig, 2011). Bullying is a social process
(Salmivalli, 2010) where many students who are not directly involved as bullies or victims are often present as bystanders
(e.g., Craig, Pepler, & Atlas, 2000). According to the participant role approach (Salmivalli, 1999), students who witness bullying
can undertake different roles, including assistants, who assist and join the bullies; reinforcers, who support the bullies by
cheering and laughing; outsiders, who remain passive and try to avoid participating in the bullying; and defenders, who try to
help the victims.
Defending victims might be motivated by more extrinsic factors such as teacher approval (Thornberg et al., 2012) or
gaining or maintaining friendship with the victim (Bellmore, Ma, You, & Hughes, 2012; Forsberg, Thornberg, & Samuelsson,
2014; Rigby & Johnson, 2006; Thornberg et al., 2012). Defending might also be more intrinsically motivated by internal factors

* Corresponding author. Lund University, Department of Psychology, Box 213, 221 00 Lund, Sweden.
E-mail address: tomas.jungert@psy.lu.se (T. Jungert).

http://dx.doi.org/10.1016/j.adolescence.2016.09.001
0140-1971/© 2016 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.
76 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

such as empathy (Forsberg et al., 2014; Thornberg, 2010; Thornberg et al., 2012), internalized moral standards (Bellmore et al.,
2012; Rigby & Johnson, 2006; Thornberg et al., 2012), or a perception that helping is simply consistent with the sort of person
they are (Rigby & Johnson, 2006). Passive bystanding (outsider role) in turn might be extrinsically motivated by enjoyment of
the event (Rigby & Johnson, 2006; Thornberg et al., 2012) and self-protection, including fear of retaliation, social disapproval,
social blunders, getting bullied, losing friends, or losing social status (Bellmore et al., 2012; Forsberg et al., 2014; Rigby &
Johnson, 2006; Thornberg, 2010; Thornberg et al., 2012). Pro-bullying (assistant and reinforcer roles) might be extrinsi-
cally motivated by gaining or maintaining friendship with the bullies (Forsberg et al., 2014; Thornberg et al., 2012) or a
perceived opportunity to seek higher social status and social approval among the popular bullies (Forsberg et al., 2014).
Research has shown that bullying is more frequent in school settings where bystanders reinforce bullying and is less
frequent when bystanders are more inclined to defend the victims (Nocentini, Menesini, & Salmivalli, 2013; Salmivalli,
Voeten, & Poskiparta, 2011). However, studies indicate that bystanders rarely support the victims (e.g., Craig et al., 2000).
Because a set of social and cognitive factors can inhibit bystanders from helping victims (Gini, Pozzoli, & Bussey, 2015; Pozzoli
& Gini, 2010; Thornberg & Jungert, 2013, 2014), it is crucial to examine students' motivation to defend victims in bullying
situations. Furthermore, even though warm and supportive studenteteacher relationships have been associated with de-
fender behavior in a previous study (Thornberg, Wa €nstro€m, Hong, & Espelage, 2016), it is currently unclear to what degree
this link might be mediated by intrinsic motivation (caring teachers as moral role models promoting identification and moral
internalization) or extrinsic motivation (caring teachers as reinforcers of moral behavior and punishers of immoral behavior)
to defend victims in bullying.

Self-determination theory

A general theory of human motivation that has gained more attention and has received growing empirical support in
recent decades is self-determination theory (SDT; Deci & Ryan, 1985; Ryan & Deci, 2000). This theory differentiates types of
motivation and proposes that there are two broad forms of motivation that represent two ends on a continuum of self-
volition (see Deci & Ryan, 2000). The motivation continuum extends from a motivation' to íntrinsic motivation'. The
former is the lack of motivation and the latter is the prototype of self-determined activity. Between a motivation' and íntrinsic
motivation' are four types of regulation. External regulation is the classic case of extrinsic motivation in which people's
behavior is controlled by specific external contingencies. People seek to attain a desired consequence such as tangible rewards
or to avoid a threatened punishment. Introjected regulation, which involves the person's ego and the emergence of feelings of
pride or guilt or shame when engaging (or not engaging) in a particular behavior, represents a partial internalization in which
regulations have not really become part of the integrated set of motivations, cognitions, and affects that constitute the self.
Identification is the process through which people recognize and accept the underlying value of a behavior. Identification
involves the acceptance and personal valuing of an acquired regulation. As compared with external and introjected regu-
lations, identified regulation is expected to be better maintained and to be associated with higher commitment and perfor-
mance. The fullest, most complete form of internalization of e xtrinsic motivation' is integrated regulation. This involves
identifying with the importance of behaviors but also integrating those identifications with other aspects of the self. An
example is when a person says, “I help people because helping is part of who I am”.
With reference to these types of regulations, external and introjected regulations are considered to be controlled moti-
vation, and identified and integrated regulations are considered to be autonomous motivation. However, although introjected
regulation is categorized as controlled motivation, it is still recognized as less controlled than external regulation. Autono-
mous motivation concerns actions that are experienced as emanating from or congruent with one's self. Hence, autono-
mously motivated behaviors reflect a person's volition, values, or interests and make one feel like they are the “origin” of the
behavior rather than a “pawn” acting out the behavior (Ryan & Connell, 1989). In contrast, controlled motivation is experi-
enced as emanating either from external contingencies and controls or from self-imposed pressures like feelings of shame,
guilt, or pride (Deci & Ryan, 2000). Thus, controlled behaviors are driven by a desire to maintain self-esteem, avoid self-
sanctions, please others, obey demands, and receive rewards and gains as well as to avoid punishments and costs
(Weinstein & Ryan, 2010).
Previous research has shown that autonomous motivation predicts stronger persistence than controlled motivation in
domains such as doing homework (Hagger, Sultan, Hardcastle, & Chatzisarantis, 2015), learning and academic performance
(Niemiec & Ryan, 2009; Taylor et al., 2014), physical activity (Teixeira, Carraca, Markland, Silva, & Ryan, 2012), health behavior
change and maintenance (Ng et al., 2012; Ryan, Patrick, Deci, & Williams, 2008), job performance (Moran, Diefendorff, Kim, &
Liu, 2012), creativity (Grant & Berry, 2011), and work satisfaction (Van den Broeck, Lens, De Witte & Van Coillie, 2013) to name
just a few. Moreover, as compared to controlled motivation, autonomous motivation to engage in prosocial behavior has been
found to be associated with actual prosocial behavior (Hardy, Dollahite, Johnson, & Christensen, 2015). In two studies by
Weinstein and Ryan (2010), it was found that autonomous helpers (university students) provided more help and were
perceived as more helpful than controlled helpers in situations where they reported their general prosocial acts in terms of
helping someone or doing something for a good cause on a given day and when they were asked to donate money to another
participant in the study.
Because peer influence is one of the strongest predictors of bullying in the school context (for a review, see Cook, Williams,
Guerra, Kim, & Sadek, 2010), students' self-determination to defend victims in bullying should be a crucial component. From a
theoretical point of view, Tsang, Hui, and Law (2011) argue, “If students are well equipped with self-determination, they are
 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90 77

less likely to acquiesce to negative peer pressure and school culture and are more able to take the positive side when they
witness bullying” (p. 2281). They should, in other words, be more determined to intervene in bullying situations. For example,
Knee and Neighbors (2002) found that among college students the relationship between peer pressure and alcohol drinking
was stronger for those who were oriented toward feeling controlled. The findings by Knee and Neighbors support previous
research on self-determination by showing that feeling less self-determined is associated with poorer health outcomes. This
is also relevant in relation to school bullying because a major reason why people drink alcohol is the desire to attain peer
acceptance and social approval, which also are some of the major reasons why people bully others. In other words, both
behaviors are grounded in social influence. Thus, feeling controlled and providing more extrinsic reasons for a particular
behavior have been linked to less adaptive outcomes in several domains, including educational settings, romantic relation-
ships, and medical contexts.
Furthermore, more positive attitudes toward bullying (Pozzoli & Gini, 2013; Pozzoli, Ang, & Gini, 2012) and social-
cognitive processes that distort moral reasoning and compassion (i.e., moral disengagement; Gini et al., 2015; Thornberg
& Jungert, 2013, 2014; Thornberg, Pozzoli, Gini, & Jungert, 2015) lower the probability that students will defend victims
when witnessing bullying. Therefore, we assume that adolescents with extrinsic prosocial motivation e those who have more
extrinsic reasons for helping a victim e would feel pressure to be passive bystanders or to reinforce the bullying rather than to
defend a victim if such behaviors would increase the likelihood of gaining peer acceptance and social approval, whereas a
more autonomous motivation to defend victims would make adolescents less susceptible to positive attitudes toward
bullying and moral disengagement and thus more determined to defend victims of bullying.
SDT has not yet been adopted to investigate bystander motivations and behaviors in school bullying. Thus, we still do
not know whether the differentiation between external, introjected, identified, and integrated regulations, or between
autonomous and controlled motivations to help victims of bullying, might contribute to explaining students' proneness to
pro-bully, to remain passive, or to defend the victim as bystanders in bullying situations. Based on SDT and research on
autonomous vs. controlled motivations in other domains, it is nevertheless plausible to hypothesize that autonomous
forms of motivation for prosocial behaviors are associated with prosocial and defender behavior in school bullying
situations.

Basic psychological needs, studenteteacher relationships, and motivation

In SDT, research has found that satisfaction of the three basic needs for autonomy, competence, and relatedness are
required for autonomous motivation (Ryan & Deci, 2000). The need for autonomy is defined as an individual's desire to be able
to act in line with one's own preferences, to make one's own choices, and to express feelings freely (Deci & Ryan, 2000).
Autonomy refers to volitional, harmonious, and integrated functioning in contrast to more pressured or alienated experiences
(Deci & Ryan, 2000). Autonomy should not be confused with independence, which deals with whether one is relying on
others or not (Ryan, 1995). The need for competence refers to an individual's sense of being efficient, of receiving positive
feedback, and of having opportunities to develop new skills (Deci & Ryan, 2000). The need for relatedness refers to the wish to
have caring bonds and positive alliances with others (Baumeister & Leary, 1995; Deci & Ryan, 2000). Relatedness also refers to
“the desire to feel connected to others e to love and care, and to be loved and cared for” (Deci & Ryan, 2000, p. 231). Enjoyable
interactions with peers and teachers will therefore attend to this need.
Autonomy-supportive teachers strive to meet the students' needs for autonomy and competence and have been shown to
be associated with autonomous motivation in academic engagement and achievement (for reviews, see Reeve, 2006; Ryan &
Deci, 2000). Autonomy support is defined as “the degree to which people use techniques which encourage choice and
participation toward school activities” (Guay & Vallerand, 1996, p. 215). Roth, Kanat-Maymon, and Bibi (2011) state that
“autonomy-supportive contexts involve acknowledgement of the child's feelings, taking the child's perspective, providing
rationale, allowing choice, and minimizing pressure” (p. 656), and this has been found to predict deeper internalization of
attitudes, values, and behaviors (for a review, see Roth et al., 2011). To what degree an autonomy-supportive teaching style
would have a positive transfer effect on prosocial and caring behavior like defending victims in bullying is, however, still
unclear. Nevertheless, autonomy support has been linked to prosocial behavior (Weinstein & Ryan, 2010), and Roth et al.
(2011) found that students' perceptions of autonomy-supportive teaching was positively related to their identified inter-
nalization of considerateness towards classmates, which in turn was negatively related to self-reported bullying.
Moreover, according to SDT (Deci & Ryan, 2000; Ryan & Deci, 2000), the satisfaction of the need for relatedness expressed
as caring bonds is one of the three basic needs required for internalization and autonomous motivation. Developing caring,
warm, and supportive relationships with students is part of being a caring role model and creating a prosocial classroom
(Jennings & Greenberg, 2009) to better meet students' need for relatedness. A positive or high-quality studenteteacher
relationship is related to closeness, which is about “the degree of warmth and positive affect between the teacher and the
child, as well as how comfortable the child is approaching the teacher” (Sabol & Pianta, 2012, p. 215). In warm stu-
denteteacher relationships, the student accepts the teacher as an important emotional and cognitive resource (Mashburn &
Pianta, 2006; Pianta, 1992; Pianta, Nimetz, & Bennett, 1997). A negative or low-quality studenteteacher relationship implies
constant conflict manifested as “the negativity or lack of rapport between the teacher and child” (Sabol & Pianta, 2012, p. 215).
A negative or conflictual studenteteacher relationship is characterized by disapproval and unpredictability. Students who
perceive their teachers as conflictual often find impairments in the relationship, such as attachment insecurity as well as
denial regarding the teacher (Milatz, Glüer, Harwardt-Heinecke, Kappler, & Lieselotte Ahnert, 2014).
78 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

Previous studies have found that positive, caring, warm, and supportive studenteteacher relationships are associated with
higher psychological well-being (Sarkova et al. 2014); prosocial behavior (Roorda, Vershueren, Vancraeyveldt, Van Craeyevelt,
& Colpin, 2014); social, emotional, and behavioral adjustment (Cadima, Leal, & Burchinal, 2010; Merritt, Wanless, Rimm-
Kaufman, Cameron, & Peugh, 2012); and lower levels of aggression and antisocial behavior (Hoglund, Klingle, & Hosan,
2015; Hughes, Cavell, & Willson, 2001; O’Connor, Dearing, & Collins, 2011; Merritt et al., 2012; Richard, Schneider, &
Mallet, 2011; Roorda et al., 2014; Runions et al., 2014), including peer victimization and bullying (Gregory et al., 2010;
Lucas-Molina, Williamson, Pulido, & Pe rez-Albe niz, 2015; Richard et al., 2011; Wang, Swearer, Lembeck, Collins, & Berry,
2015). Furthermore, and with relevance for the present study, a recent study has shown that the more caring, warm, sup-
portive, friendly, and respectful teacherestudent relationships were, the more students were likely to defend the victims
when witnessing bullying (Thornberg et al., 2016).
From a theoretical point of view, the link between positive, warm, caring, and supportive studenteteacher relationships
and defending might be mediated by autonomous motivation to defend. Wang et al. (2015) found that students' attitudes
toward bullying mediated the relationship between studenteteacher relationships and bullying. With reference to SDT
literature (Deci & Ryan, 2000), Wang et al. (2015) argue that students are more likely to internalize values and regulations
from the social group in which they feel secure and to which they can relate, such as having positive relationships with their
teachers. With reference to Wang et al. (2015), we would argue that when students have positive, warm, caring, and sup-
portive relationships with their teachers, they might be more likely to internalize their teachers' values. Such moral iden-
tification and internalization would in turn promote autonomous motivation rather than controlled motivation in prosocial
and caring behaviors such as defending victims in bullying. Currently, no studies have examined the role of students'
motivation to defend victims in bullying as a mediator between studenteteacher relationship quality and various bystander
behaviors in bullying.

The present study

In this study, we adopted SDT as a framework for understanding students' motivations to defend victims and their
bystander behaviors in school bullying. The current study was the first to examine whether autonomous motivation (based on
identified and integrated regulations), introjected motivation (based on introjected regulation), and extrinsic motivation (based
on external regulation) to defend were associated with different bystander behaviors in school bullying. The aim of this study
was to investigate whether three dimensions of studenteteacher relationships (warm, autonomy-supportive, or conflictual)
were associated with autonomous, introjected, and extrinsic motivations to defend bullying victims, as well as the association
between these types of motivations and different bystander behaviors (pro-bullying, passive bystanding, and defending) in
bullying situations among early adolescents in school. In the current study, we assumed that the quality of studenteteacher
relationships is based on the interplay between teacher characteristics and student characteristics. Previous studies support
our assumption and have, for instance, found that teachers' interactions with students are related to the characteristics of the
children themselves (for a review, see Pianta, 2006). At the same time, the quality of such relationships should be considered
as a contextual factor in which teachers and students influence each other, and this in turn influences students' academic
motivation and various social behaviors (Pianta, 2006; Wubbels et al., 2015). Thus, it is meaningful to use students' estimated
mean values of their studenteteacher relationship quality with each teacher even though they switch classes and teachers
throughout the day.
This study includes a sample of students in Italy where the prevalence rate of bullying has been found to be about 12% for
physical and about 50% for verbal and relational bullying (Vieno, Gini, & Santinello, 2011). Moreover, Italy is a country in
which an Observatory on School Bullying was created in each of its 20 regions in 2007 under the direction of the Regional
School Office. This has resulted in promotion, coordination, and supervision of anti-bullying initiatives at the school and the
community level in each region in the country. Additionally, national institutions have contributed to increased awareness
about bullying. Thus, there is reason to believe that Italian schools have policies and anti-bullying programs and that the
school management and the teachers have had training and experiences with how to respond to school bullying. With this
background, the present research investigated three primary hypotheses.
First, we hypothesized that autonomy-supportive studenteteacher relationships would be positively associated with
defending victims and would be mediated by autonomous motivation to defend. This hypothesis is based on the SDT tenet
that autonomy-support is related to autonomous motivation (Deci & Ryan, 2000) and positive outcomes such as prosocial
behavior (Weinstein & Ryan, 2010). We also draw from Roth et al. (2011) who showed that autonomy-supportive teaching
was positively related to students' identified internalization of considerateness towards classmates, which in turn was
negatively related to self-reported bullying.
Second, we hypothesized that warm studenteteacher relationships would also be positively associated with defending
victims and would be mediated by both autonomous motivation and introjected motivation to defend. This hypothesis is
based on the theory that warm studenteteacher relationships help to encourage and assist students emotionally and
cognitively (Pianta, 1992) so that they will be more likely to help victims. This association is hypothesized to work via
autonomous motivation, because warm relationships are believed to satisfy the need for relatedness that in turn supports
autonomous motivation, and via introjected motivation, because warm relationships might encourage students to reflect on
their responsibilities that in turn might increase expectations and feelings of guilt. Defending will be associated via auton-
omous motivation to a stronger degree and via introjected motivation to a weaker degree.
 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90 79

Finally, we hypothesized that conflictual studenteteacher relationships (a pattern of harshness, criticism, and destructive
conflicts) would be positively associated with passive bystanding and pro-bullying mediated by extrinsic motivation.
Conflictual relationships between teachers and students have often been associated with antisocial behavior and withdrawal
(Hamre & Pianta, 2001; Rudasill & Rimm-Kaufman, 2009), which is why we believe they will be associated with passive
bystanding and pro-bullying. We hypothesized that this association is mediated by extrinsic motivation to defend because
defending puts the defender at risk, and bullies tend to be powerful and to have high social status (Pellegrini et al., 2010;
Thornberg, 2011). The peer pressure to take the bullies' side or at least not go against them would therefore be strong (cf.,
Hamarus & Kaikkonen, 2008). Thus, extrinsic motivation to defend would be too weak to motivate students to defend victims
when taking into account these external contingencies that would motivate bystanders to remain passive or to take the
bullies' side. These hypothesized links culminated in the model shown in Fig. 1.

Method

Participants and procedure

Participants were recruited from 24 Italian public school classes (5th to 9th grade) from middle schools located in urban
areas about 5 months after the beginning of the school year, once in 2015 and once in 2016.
School principals and teachers authorized the participation of each class in the study. Prior to data collection, individual
consent for participation as well as active parental consent was obtained. Participants were informed that participation in the
study was voluntary and that they were allowed to refuse to participate and withdraw from the study at any time.
From the initial total sample of students who were invited to participate in the study (N ¼ 472), a total of 450 participants
were present in the classrooms at the time of the data collection, out of which 436 students completed all the measures of
interest. All students in the selected classes were included in the sample. The first data collection was in March of 2015 and
resulted in complete responses from 252 students. The second data collection was in March of 2016 and resulted in complete
responses from 184 students. However, 31 multivariate outliers (i.e., cases with a Mahalanobis distance exceeding the critical
value) were identified and removed prior to the analyses. The final sample consisted of 405 early adolescent students (53.33%
boys, 46.67% girls; M ¼ 12.28 years, SD ¼ 1.15 years, range: 11e15 years). This represents an 85% response rate. Of the par-
ticipants, 84 were in 5th grade, 91 were in 6th grade, 118 were in 7th grade, 92 were in 8th grade, and 20 were in 9th grade.
Sampling was stratified within the two specific Italian areas of the Italian Islands and Piedmont to be representative of the
schools' student populations in these areas. About 5% of the participants had an immigrant background, which is less than in
Italy as a whole, where they make up 8% of the population. Even though socio-economic status was not directly measured,
this stratified sampling strategy ensured that the sample in this study represented youths from a wide range of socio-
economic statuses, from lower-class and middle-class families, as in all public schools in Italy. The two selected regions in
Italy where chosen mainly in order to ensure a sampled population that accurately reflects the referent population. Another
reason for selecting the two regions was that we had good connection with schools in those regions, which made it possible to
have smooth data collection procedures. The participants completed an anonymous questionnaire during class time. Data
were collected in a paper/pencil format, where a researcher visited each classroom, distributed questionnaires to all present

Fig. 1. The hypothesized model.

80 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

students in each class, and stayed in the classroom until the students had filled in the questionnaires. The present researcher
could answer questions if students had any queries regarding items on the questionnaires.

Measures

The children's appraisals of teacher support

To explore the students' perception of their teachers, we used a short version of The Children's Appraisals of Teacher
Support (CATS items used by Spilt, Koomen, & Mantzicopoulos, 2010). This scale consists of 18 items and is composed of three
subscales that are based on the three hypothesized dimensions of warmth, conflict, and autonomy support that largely
correspond with the theoretical constructs that underlie teacher support (see Mantzicopoulos & Neuharth-Pritchett, 2003).
Warmth reflects whether the child feels cared for and valued. Conflict refers to perceived harshness and criticism from the
teacher. Autonomy-Support represents appraisals of teacher behaviors that stimulate children to pursue their own choices
and interests. The Warmth subscale measures students' perceptions of teacher support, encouragement, and acceptance (7
items: e.g., “My teacher likes me”), the Conflict subscale assesses the students' perceptions of negativity in interactions with
teachers (7 items: e.g., “My teacher easily gets angry with me”), and the Autonomy-Support subscale measures students'
perceptions of teacher practices that promote autonomy and self-directed activities (4 items: e.g., “My teacher lets me do
activities I want to do”).
In Italy, the teacher in Italian, History, and Geography is always the main teacher whom the students see every day (on
average 10 h a week out of a total of 25e30 weekly hours) in all grades through elementary and middle school. Thus, when the
students were asked to report on their perceptions of their relationships with their teacher, we asked them to think of their
teacher in Italian, History, and Geography. Participants were asked to select an answer from a continuum from 1 (“Completely
disagree”) to 5 (“Completely agree”) in response to each item in this scale. The scale was translated from English to Italian by
native Italian speakers, and the translation was checked and edited by native Italian speakers.

The motivation to defend scale

A 15-item scale was developed for the purpose of this study. These items were adapted from motivation scales in the SDT
literature, such as Prosocial Motivation in Children (Ryan & Connell, 1989), the Motivation to Help Scale (Weinstein & Ryan,
2010), and other surveys assessing self-determined motivation proposed by SDT (Deci & Ryan, 2000) such as the Academic
Motivation Scale (Vallerand et al., 1992). To maintain the resemblance of the present scale with the theoretical construct it is
meant to measure, key words in our scale were similar to the ones used in these previously developed scales. To adapt the
measure for pre-adolescence in a school context, the items were formulated to measure the participants' motivation to
defend bullying victims during the previous 6 months in order to harmonize with other questions in the questionnaire.
Participants were asked to think of situations when they had witnessed another student being bullied and to report “why they
would engage in helping a bully victim.” The scale was composed of four subscales measuring the four types of motivation to
help (extrinsic, introjected, identified, and intrinsic). The list included items such as “Because I like to help other people”
(intrinsic), “Because I think it is important to help people who are treated badly” (identified), “Because I would feel like a bad
person if I did not try to help” (introjected), and “To become popular” (extrinsic). As in the first part of the questionnaire,
participants were asked to select an answer from a continuum from 1 (“Completely disagree”) to 5 (“Completely agree”).

The participant role scale

We used a 20-item self-report scale (adapted from both Gini et al., 2015; and Thornberg & Jungert, 2013) aimed to tap
participants' proneness to act as a bully, as a pro-bully (assisting the bullies or reinforcing bullying by laughing and cheering
on the bullies), as an outsider/passive bystander (staying outside bullying and remaining passive), as a defender (defending
and helping the victim), or as a victim. Twelve of the items were originally in Italian, developed and validated by Gini, Pozzoli,
and Bussy (2014), and eight items were originally in Swedish, developed and validated by Thornberg and Jungert (2013), and
these were translated from the English version to Italian by native Italian speakers. The translation was checked and edited by
native Italian speakers. The items included behaviors such as “I tease some classmates, calling them nasty nicknames,
threatening or offending them” (bullying), “I laugh or cheer on the kids who tease or call a classmate nasty nicknames” (pro-
bullying), “When a classmate is hit or pushed, I stand by and I mind my own business” (passive bystanding), “I defend
classmates who are targeted by gossip or false rumors that are said behind their back” (defending), and “I am bullied through
the internet, messages or calls on the phone” (victimization). Participants were asked to rate how often (with reference to the
current school year) they had enacted the behavior described in each item on a 5-point scale (1 ¼ “It has never happened in
the last month”, 2 ¼ “It happened just once”, 3 ¼ “Two/three times a month”, 4 ¼ “About once a week”, 5 ¼ “More times a
week”). Due to the aim of the current study, the subsequent analyses were focused on the three possible bystander roles or
behaviors tapped in the scale: (a) pro-bullying, (b) passive bystanding, and (c) defending.

Statistical analyses

Validation of the scales constituted the first stage of analysis. Principal component analysis (PCA) and confirmatory factor
analysis (CFA) were used to assess the internal consistencies of the three scales measuring children's appraisals of teacher
support, their motivation to defend, and their participant roles. The reason for the decision to conduct both PCA and CFA was
 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90 81

to validate the three scales for the first time with an Italian sample. PCA tests the number of common factors that influence
measures, and it tests the strength and relationship between each common factor of the corresponding measure (DeCoster,
1998). The PCA allows us to identify the nature of the constructs that underlie responses given in the questionnaire, to
determine sets of items that interconnect, to demonstrate the depth and breadth of measurement scales, and to generate
factor scores that represent the underlying constructs (DeCoster, 1998). We then proceeded with CFA in order to determine
the extent to which the given set of items that were generated in the PCA and represent underlying constructs would measure
the hypothesized factors deduced from that theory (Ruscio & Roche, 2012) and to establish the validity of the factor model.
The PCA was carried out with Varimax rotation. Goodness of fit and scree tests were used to select parsimonious models.
CFA was carried out using the EQS program (Bentler, 1995) to confirm the selected models.
In the second stage of analysis, structural equation modeling based on the analysis of the covariance matrix was utilized to
examine the hypothetical model, which postulated associations between the constructs (latent factors) of children's ap-
praisals of teacher support and the three outcomes of defending role, passive bystander role, and pro-bullying role via
motivation to defend victims (all three of which are also latent factors). We then carried out CFA to examine the structure of
instruments involving indicators as observed variables and to test for invariance of factor loadings. The Lagrangian Multiplier
test (LM test) for adding parameters that might improve model fit, and the Wald test (W test) for identifying parameters that
are not significant, were employed.
In the next step, we tested the hypothetical model to evaluate its ability to explain adolescents' motivation to defend
victims and the roles they take when witnessing bullying in their schools. The hypothetical model was also tested separately
for both genders. To evaluate model fit, we first used the chi-square test. An acceptable model should have a nonsignificant
chi-square value. However, given that this test is known to be overly sensitive to sample size and small deviations from
multivariate normality (West, Taylor, & Wu, 2012), three additional criteria were used to evaluate model fit: the Comparative
Fit Index (CFI), Satorra Bentler Scaled Statistics (S-BSS) divided by the number of degrees of freedom (S-BSS/df), and the root
mean square error of approximation (RMSEA). S-BSS incorporates a scaling correction for the chi-square statistic when
distributional assumptions are violated. Its computation takes into account the model, the estimation method, and the
sample kurtosis values. The S-BSS has been shown to more closely approximate chi-squared than the uncorrected test sta-
tistic, to have robust standard errors, and to perform well (Satorra & Bentler, 2001). RMSEA is a summary statistic for the
residuals (i.e., the lower the number the better). As proposed in a recent review of appropriate reporting of SEM results, we
used CFI values of .90 or more to guide decisions regarding acceptable model fit (Hoyle & Panter, 1995), while a c2/df ratio
larger than 2 indicated an inadequate fit (Byrne, 1995). For the RMSEA, we followed Kline's (2011) recommendation to use
values of .06 or less as indicative of good fit. In anticipation of the possibility that variables might not be normally distributed,
maximum likelihood and robust methods were used simultaneously during each iteration. Where an elevated Mardia co-
efficient indicated high multivariate kurtosis, results of the robust method are reported.
We tested the mediating roles of motivation to defend victims with SPSS using the macro of Preacher and Hayes (2008) for
testing and comparing indirect effects in multiple mediator models. Separate analyses were conducted for defending, passive
bystanding, and pro-bullying as the dependent variables. Warm, autonomous, and conflictual studenteteacher relationships
were used as the independent variables, while autonomous motivation and extrinsic motivation to defend were modeled as
mediators. We performed separate analyses for females and males. Bootstrapping with the number of bootstrap samples set at
5000 was used to calculate 95% confidence intervals for the specific indirect effects for each motivation type. Preacher and
Hayes (2008) recommend bootstrapping, especially for testing mediation, because it does not require normality of the sam-
pling distribution. Bootstrapping furthermore provides distributions for each statistic from which confidence intervals can be
derived (Preacher & Hayes, 2004). As with simple and multiple regressions, the parameters can be read as regression weights.

Results

Validating the scales

Appraisals of teacher support

PCA was carried out for three-factor models of the appraisals of teacher support scale in order to validate it because this
was the first time this particular scale was used with an Italian sample. Due to a number of cross-loadings, a second analysis
was performed. This model also had one cross-loading, and a third analysis resulted in a model with three factors accounting
for 57.78% of the variance. Both Bartlett's test of sphericity [c2 (36) ¼ 560.01, p < .0.001] and Kaiser's measure of sampling
adequacy (.71) indicated acceptable matrix factorability. The first factor had an Eigen value of 2.51, the second had a value of
1.68, and the third had a value of 1.02. This was subsequently confirmed in a CFA. The model fit statistics for the model
(CFI ¼ .96; S-BSS c2/df ¼ 43.45/24 ¼ 1.81, p ¼ .010; RMSEA ¼ .045, 90% CI: [.02, .07]) indicated a good fit. Table 1 shows the
factor loadings for all items. The internal reliability for each of the subscales was as follows: Warm studenteteacher re-
lationships (Spearman-Brown ¼ .70), Autonomous studenteteacher relationships (Cronbach's alpha ¼ .74), and Conflictual
studenteteacher relationships (Cronbach's alpha ¼ .64).

Motivation to defend
This scale was first subjected to a PCA with Varimax rotation for a three-factor model in order to validate it because this
was the first time this particular scale was used in research on school bullying and with an Italian sample. In the first model,
82 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

Table 1
Confirmatory factor analysis of the children's appraisals of teacher support.

Item F1: Warm F2: Autonomous F3: Conflictual

1 “Likes my family” .50
6 “Likes me” .75
8 “Lets me choose work that I want to do” .40
10 “Lets me do different activities” .69
13 “Lets me do activities that I want to do” .81
12 “Gets (easily) angry at me” .73
15 “Tells me that I don't try hard enough” .87
16 “Has too many rules for the class” .91
17 “Tells me I don't listen” .56

Note: Correlations between indicators are r(F1, F2) ¼ .28**, r(F1, F3) ¼ .26**, r(F2, F3) ¼ .01.

items measuring identified motivation and intrinsic motivation loaded on the same factor. The three-factor model also had
double loadings, where Introjected 3 (I should) and Identified 3 (Important to fight violence) loaded both on the first factor
(Autonomous motivation) and the second factor (Introjected motivation). Thus, we decided to proceed with a two-factor
model and excluded Introjected motivation. The two factors accounted for 56.00% of the variance. Both Bartlett's test of
sphericity [c2 (21) ¼ 462.81, p < .001] and Kaiser's measure of sampling adequacy (.72) indicated acceptable matrix
factorability.
The first factor (autonomous motivation to defend) had an Eigen value of 2.31, and the second factor (extrinsic motivation
to defend) had a value of 1.62. Next, this model was confirmed in a CFA. The fit statistics for the model (CFI ¼ .98; S-BSS c2/
df ¼ 18.61/13 ¼ 1.03, p ¼ .136; RMSEA ¼ .033, 90% CI: [.00, .06]) indicated a very good fit. Table 2 shows the factor loadings for
all items. The internal reliability for both of the subscales was as follows: Autonomous motivation (Cronbach's alpha ¼ .73)
and Extrinsic motivation (Cronbach's alpha ¼ .60).

Participant roles
Because the purpose of this study was to investigate bystander behaviors, all items pertaining to bullying and victimi-
zation were excluded before carrying out the PCA of the three-factor model. The remaining items were subjected to a PCA
with Varimax rotation for a three-factor model in order to validate the three-factor model. In the first model, one item
measuring the defending role had a low loading. The final model had acceptable loading and accounted for 57.59% of the
variance. Both Bartlett's test of sphericity [c2 (45) ¼ 772.87, p < .0.001] and Kaiser's measure of sampling adequacy (.72)
indicated acceptable matrix factorability. Further analysis was carried out by means of a CFA. This model with the remaining
10 items had good fit (CFI ¼ .96; S-BSS c2/df ¼ 47.30/32 ¼ 1.48, p ¼ .05; RMSEA ¼ .03, 90% CI: [.01, .05]). Table 3 shows the
factor loadings for all items. The internal reliability for each of the subscales was calculated with Cronbach's alphas as follows:
Defending (.70), Passive Bystanding (.58), and Pro-bullying (.73).

Testing the model

Because introjected motivation had to be excluded from further analyses based on the results of the validation of the
Motivation to Defend Scale, we adjusted the hypothetical model. We kept all parameters except the ones related to intro-
jected motivation to defend, see Fig. 2. The tested model did not fit the data well (CFI ¼ .87, S-BSS c2 ¼ 501.42, df ¼ 286,
p < .001, RMSEA ¼ .043, 90% CI: [.04, .05]). The LM test for adding parameters guided us to add the association between
autonomous motivation and passive bystanding, as well the association between passive bystanding and pro-bullying.
Adding these parameters improved the fit of the model to the data (CFI ¼ .92, S-BSS c2 ¼ 379.43, df ¼ 260, p ¼ .001,
RMSEA ¼ .034, 90% CI: [.03, .04]). R2 values indicated that the model explained 14% of the variance in autonomous motivation
and 11% of the variance in extrinsic motivation. In addition, this model explained variance in defending behavior (13%),
passive bystanding (31%), and pro-bullying (40%).
As hypothesized, a warm relationship with teachers was positively related to autonomous motivation to defend (b ¼ .37),
and a conflictual relationship was positively related to extrinsic motivation to defend (b ¼ .27). An autonomous relationship
with teachers was not significantly related to motivation to defend victims (b ¼ .01) or the defending role (b ¼ .17). In

Table 2
Confirmatory factor analysis of the motivation to defend scale.

Item F1: Autonomous motivation F2: Extrinsic motivation

20 “Because I think is important to help people who are treated badly” .61
22 “Because I am the kind of kid who cares about others” .58
27 “Because I think it's important to fight violence and injustice” .64
33 “Because I like to help other people” .73
19 “To be rewarded by a teacher” .59
21 “To become popular” .68
23 “Because I would get into trouble if I didn't help” .48

Note: Correlations between indicators are r(F1, F2) ¼ .11*.

 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90 83

Table 3
Confirmatory factor analysis of the bystander behavior scale.

Item F1: Defending F2: Passive F3: Pro-bullying

bystanding
35 “I help or comfort classmates who are excluded from the group and isolated” .70
41 “I defend classmates who are targeted by gossip or false rumors that are said behind their back” .81
54 “I go and tell a teacher if I see a kid being called nasty names, threatened, hit or pushed by other .41
kids”
37 “When a classmate is hit or pushed, I stand by and I mind my own business” .53
40 “When I hear nasty rumors or mean things said about other students behind their back I mind my .50
own business”
49 “If I know that someone is excluded or isolated from the group I act as if nothing has happened” .67
47 “I also start to bully the student who is being bullied” .51
52 “I laugh or cheer on the kids who tease or call a classmate nasty nicknames” .65
53 “I watch the bullying because I think it is fun to see what happens” .69
55 “I join in with the kids who call another kid nasty nicknames, threaten, hit or push him/her” .72

Note: Correlations between indicators are r(F1, F2) ¼ .13*, r(F1, F3) ¼ .09, r(F2, F3) ¼ .28**.

addition, conflictual teacher-student relationships were significantly related to passive bystanding (b ¼ .30). Furthermore,
autonomous motivation was positively related to defending (b ¼ .34), while extrinsic motivation was significantly and
positively related to pro-bullying (b ¼ .24). In addition, contrary to the hypotheses, conflictual relationships with teachers
were not positively related to pro-bullying (.01), whereas autonomous motivation to defend was significantly and negatively
related to passive bystanding (b ¼ .35). Fig. 3 shows all path coefficients for the total sample.
For females (n ¼ 189), the model had an acceptable fit to the data (CFI ¼ .91, S-BSS c2/df ¼ 376.70/260 ¼ 1.45, p < .001,
RMSEA ¼ .05, 90% CI: [.04, .06]). For males (n ¼ 216), the model also had an acceptable fit to the data (CFI ¼ .91, S-BSS c2/
df ¼ 329.68/260 ¼ 1.27, p ¼ .003, RMSEA ¼ .04, 90% CI: [.02, .05]). Fig. 4 shows all path coefficients for females and males in
brackets. R2 values indicated that the model explained 14% of the variance in the defending role for both females and males. In
addition, this model explained variance in autonomous motivation to defend (39% for females (10% for males)), extrinsic
motivation to defend (17% (3%)), passive bystanding (25% (42%)), and pro-bullying (45% (39%)). As in the model of the whole
sample, autonomous motivation was positively associated with the defender role for both females (b ¼ .36) and males
(b ¼ .33) and negatively associated with passive bystanding (females: b ¼ .35, males: b ¼ .38). Extrinsic motivation to
defend was significantly associated with pro-bullying for females (b ¼ .30) and males (b ¼ .22), as in the model of the total
sample. Furthermore, conflictual studenteteacher relationships were positively associated with extrinsic motivation to
defend for females (b ¼ .41) but not for males (b ¼ .17). On the other hand, the direct path between conflictual relationships
with teachers was significant for males (b ¼ .42), as in the model of the total sample, but not for females (b ¼ .12). The as-
sociation between extrinsic motivation and pro-bullying was not significant for females and males in the separate models, nor
were the other paths for females or males.

Fig. 2. Second hypothesized model without introjected motivation.

84 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

Fig. 3. The best-fitting model for the whole sample. All path coefficients marked with an * are significant at p < .05.

Fig. 4. The best-fitting model for females and males. Path coefficients for males are in brackets. All path coefficients are significant at p < .05. * indicates pathways
where gender invariance was not obtained.
 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90 85

Mediation of motivation to defend

An initial multiple mediation analysis (Preacher & Hayes, 2004, 2008) with a bootstrap procedure (5000 iterations, bias-
corrected 95% confidence intervals) was run with warm teacher-student relationships as an independent variable, defending
behavior as the dependent variable, and autonomous motivation to defend as a mediator. The results confirmed the medi-
ating role of autonomous motivation in the relationship between warm studenteteacher relationships and the defending role
because zero was not contained in the confidence interval (95% CI: [.040, .133]; effect size ¼ .08).
A second analysis with autonomous studenteteacher relationships as an independent variable, defending behavior as the
dependent variable, and autonomous motivation to defend as a mediator was performed. The results confirmed the medi-
ating role of autonomous motivation in the relationship between autonomous studenteteacher relationships and the
defending role because zero was not contained in the confidence interval (95% CI: [.005, .063]; effect size ¼ .03).
We also ran two analyses to test the role of conflictual studenteteacher relationships. In the first one, we tested conflictual
studenteteacher relationships as an independent variable and extrinsic motivation to defend as a mediator and passive
bystanding as the outcome variable. The confidence interval (95% CI: [.001, .051]; effect size ¼ .02) suggests that extrinsic
motivation mediates the relationship between conflictual relationships and the bystanding role. Finally, we ran an analysis to
test conflictual studenteteacher relationships as an independent variable and extrinsic motivation to defend as a mediator
and pro-bullying as the outcome variable. Zero was not contained in the confidence interval (95% CI: [.005, .037]; effect
size ¼ .02), suggesting that extrinsic motivation mediates the relationship between conflictual relationships and the pro-
bullying role (Preacher & Hayes, 2004, 2008).
Mediating analyses for females and males confirmed the mediating role of autonomous motivation in the association
between warm studenteteacher relationships and the defending role because zero was not contained in the confidence
intervals (for females, 95% CI: [.071, .262], effect size ¼ .15; and for males, 95% CI: [.004, .095], effect size ¼ .03). The mediating
role of autonomous motivation in the relationship between autonomous studenteteacher relationships and the defending
role was confirmed for males (95% CI: [.011, .105]; effect size ¼ .05), but not for females (95% CI: [.059, .043]; effect
size ¼ .00).
Extrinsic motivation mediated the relationship between conflictual relationships and the bystanding role for females (95%
CI: [.005, .091]; effect size ¼ .04), but not for males (95% CI: [.006, .058]; effect size ¼ .00). Finally, testing the mediating role
of extrinsic motivation between conflictual studenteteacher relationships as an independent variable and pro-bullying as the
outcome variable indicated that extrinsic motivation only mediated the relationship between conflictual relationships and
the pro-bullying role for females (95% CI: [.011, .093]; effect size ¼ .04), and not for males (95% CI: [.002, .029]; effect
size ¼ .01).

Discussion

The present study is the first to examine whether autonomous and extrinsic motivations to defend were associated with
pro-bullying, passive bystanding, or defending in bullying situations among early adolescent students, as well as how these
motivations to defend could be associated with warm, autonomy-supportive, and conflictual studenteteacher relation-
ships. Our first hypothesis was not confirmed because defending victims was not directly associated with autonomy-
supportive studenteteacher relationships and because defending was not related to autonomy-supportive stu-
denteteacher relationships through autonomous motivation. There are at least a couple of possible reasons for the un-
expected finding that autonomy-supportive studenteteacher relationships were not linked to students' autonomous
motivation to defend victims in bullying situations. An autonomy-supportive studenteteacher relationship refers to an
interaction pattern that is more focused on teachers making room for students to experience agency and make choices in
schoolwork and classroom activities. Autonomy-supportive teachers in classrooms support students' needs for autonomy
and competence (Deci & Ryan, 2000; Ryan & Deci, 2000) and increase their autonomous motivation towards academic
learning and achievement (Guay, Ratelle, Roy, & Litalien, 2010; Niemiec & Ryan, 2009; Taylor et al., 2014). Nevertheless, our
findings suggest that early adolescents do not transfer such classroom experiences to bystander situations in bullying.
Further studies should examine whether teachers who display higher levels of autonomy support in terms of making room
for students to experience agency in prosocial, helping, and defender behavior would increase students' autonomous
motivation to defend victims in bullying. Thus, scale development to measure autonomy-supportive teaching in prosocial
and defender behavior is needed. In other words, a possible explanation for our findings is that the effect of autonomy-
supportive teachers might be more domain-specific and situated rather than generalized across various domains (cf.,
Bandura, 1999).
Furthermore, Thornberg (2010) found that teachers seem to inhibit rather than promote helping and defending behaviors
by communicating to students that they should not intervene because intervention is the responsibility and domain of the
teachers and other school staff. Thus, although teachers might be autonomy-supportive in the classroom in terms of
schoolwork and different activities, it is plausible to assume that they are more controlling and “autonomy-inhibiting” when
addressing students as bystanders in bullying and other forms of aggression and violence in school. Further research is
needed to examine how the degree of an autonomy-supportive studenteteacher relationship and its link with autonomous
motivation might vary across different contexts and activities. The current findings might also be confounded by the Italian
culture, and research with a cross-cultural approach should investigate these variables further.
86 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

As expected, and in accordance with SDT (Deci & Ryan, 2000), we found that autonomous motivation was positively and
significantly associated with defending. This indicates that early adolescents who are more autonomously motivated to
conduct a particular activity are also more prone to actually carry out the activity. The association between autonomous
motivation and behavior has been found across different activities and practices (e.g., Moran et al., 2012; Niemiec & Ryan,
2009), including prosocial behavior (Hardy et al., 2015). The current study contributes to the literature by being the first to
examine and confirm this association in the context of witnessing bullying. Our findings suggest that early adolescents who
have higher levels of autonomous motivation to defend also defend victims more often in practice. Such motivation seems to
make them more determined to resist peer pressure and other social inhibition effects such as diffusion of responsibility,
pluralistic ignorance, and audience inhibition (Bierhoff, 2002). To be autonomously motivated is to have agency and to be self-
determined in relation to a particular action. The action is experienced as emanating from and congruent with a person's self
and is a perception of an internal locus of causality (Deci & Ryan, 2000).
Our second hypothesis was confirmed because warm studenteteacher relationships were positively associated with
defending victims and were mediated by autonomous motivation to defend. Thus, the current findings contribute to the
literature on studenteteacher relationships (e.g., Hamre & Pianta, 2001; O'Connor et al., 2011; Roorda et al., 2014) by
demonstrating that positive, warm, and supportive studenteteacher relationships are not only associated with academic
achievement (Roorda, Koomen, Spilt, & Oort, 2011), prosocial behavior in general (Roorda et al., 2014), and lower levels of
externalizing behavior (Hughes et al., 2001; O'Connor et al., 2011), including bullying (Lucas-Molina et al., 2015), but are also
positively associated with students who are more self-determined to intervene in bullying situations to help a victimized
peer.
Introjected motivation could not be part of the tested models because the validation of the Motivation to Defend Scale
indicated that introjected motivation was not a factor. Introjected motivation includes contingent self-esteem that pressures
people to behave so that they will feel worthy and ego involvement that pressures people to strengthen their brittle egos
(Ryan, 1982). This type of motivation is both within the person and relatively controlled (Deci & Ryan, 2000), which might
explain why it was not possible to separate it from autonomous motivation, and it was also not part of externally regulated
prosocial motivation in the current study. Another reason for this might be due to the low number of items that were used to
measure introjected motivation. Thus, the hypothesized positive association between warm relationships and introjected
motivation to defend could not be confirmed. Future research needs to develop a better scale to measure introjected prosocial
motivation that includes more items that tap into both contingent self-esteem and ego involvement. Such an instrument
could be used to investigate whether introjected motivation is associated with defending.
Self-sanctions and basic emotions such as shame and guilt have been suggested to motivate and regulate people in
refraining from immoral behavior and behaving morally instead (e.g., Bandura, 1999; Tangney, Stuewig, & Mashek, 2007), and
it would be interesting to explore if moral emotions always have positive interpersonal consequences or not, in line with de
Hooge (2013) who demonstrated that shame motivates performance and approach behaviors to reinstate a damaged self and
withdrawal or avoidance behaviors when it is too risky or dangerous to restore the damaged self, as may be the case when
witnessing bullying. Guilt can also have many negative interpersonal consequences, such as withdrawal behavior, which
implies that for the moral emotions shame and guilt, behaviors can vary from antisocial to prosocial (de Hooge, 2013). Thus,
more research is needed in order to increase our knowledge on how adolescents' interjected motivation is related to their
participant roles when witnessing bullying.
Our third hypothesis was confirmed because conflictual studenteteacher relationships were positively associated with
passive bystanding and pro-bullying via extrinsic motivation. Thus, a negative and conflictual studenteteacher relationship is
not only associated with poor academic achievement (Roorda et al., 2011), externalizing and aggressive behaviors (Birch &
Ladd, 1998; Hughes et al., 2001; O'Connor et al., 2011; Pianta, Steinberg, & Rollins, 1995; Roorda et al., 2014), and a decline
in prosocial behavior (Birch & Ladd, 1998), but our findings also add its relationship with early adolescents who are less self-
determined to intervene to help bullied students. Whereas a positive and warm studenteteacher relationship supports
students' need for relatedness, which is one of the three basic needs required for internalization and autonomous motivation
(Baumeister & Leary, 1995; Deci & Ryan, 2000), a negative and conflictual studenteteacher relationship fails to meet this basic
need.
The costereward analysis explanation (REF) could be relevant to understanding the findings showing that extrinsic
motivation was positively associated with pro-bullying. Even though early adolescents might be motivated to defend a victim
in order to gain social rewards or avoid social costs or punishments, this extrinsic motivation could easily be overruled if
students perceive pro-bullying as more rewarding and the anticipated social or physical costs of defending as more severe.
Hence, extrinsic motivation to defend makes early adolescents more focused on the external contingencies, and thus to
behave to attain the most desired consequences. Because bullies tend to be powerful, popular, and prone to use aggression to
attain and maintain social dominance (Pellegrini et al., 2010; Thornberg, 2011), counteracting them would probably be
associated with extrinsically motivated students and with much costlier consequences than the consequences of not helping.
Finally, the structural models to evaluate adolescents' motivation to defend bully victims among females and males
separately had some interesting differences. The association between girls' warm studenteteacher relationships and
autonomous motivation was highly significant, while it was not significant for boys. In addition, the association between girls'
conflictual studenteteacher relationships and extrinsic motivation was highly significant, while it was not significant for boys.
On the other hand, conflictual studenteteacher relationships were directly and significantly related to passive bystanding for
boys, but not for girls. For girls, the association between warm relationships with the teacher and defending behavior was
 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90 87

mediated by autonomous motivation, and the association between conflictual relationships and both passive bystanding and
pro-bullying was mediated by extrinsic motivations, while for boys the associations seemed to be more direct between
motivation and specific roles and between conflictual relationships and passive bystanding. Motivation to defend plays an
important role for the chosen behavior when witnessing school bullying for both boys and girls, but the mediating role of
motivation is only of importance for girls, which implies that the studenteteacher relationships are more important for girls
than for boys. No previous study has examined gender differences in the association between studenteteacher relationships
and prosocial motivation. However, earlier findings on gender differences in the effects of teachers' support have been mixed
(see e.g. Colarossi & Eccles, 2003; Koomen, Verschueren, van Schooten, Jak, & Pianta, 2012; Rueger, Malecki, & Demaray,
2008; 2010). The significance of the impact of gender differences in studenteteacher relationships and the association
with outcomes related to bullying will be important to clarify in future research.

Limitations and implications

Some limitations of this study should be noted. All data were collected with self-report measures, which are vulnerable to
social desirability and shared method variance effects. Furthermore, we adopted a cross-sectional design, and we are
therefore not able to pinpoint the direction of effects between the variables. For example, it is not clear whether positive
studenteteacher relationships predict autonomous motivation to defend and actual defender behavior, or if teachers tend to
develop positive relationships with students who display prosocial and caring behavior. In addition, it is possible that the
relations found in the study are reciprocal. Korthagen, Attema-Noordewier, and Zwart (2014), for instance, concluded that
there is a two-way interactive process between studenteteacher relationship quality and student characteristics. In fact, we
assumed that the quality of studenteteacher relationships were produced by reciprocal interactions between teacher and
student characteristics, and that this relationship quality in turn was associated with students' motivation and behavior (cf.,
Pianta, 2006; Wubbels et al., 2015). In light of this, the identified associations and any causal conclusions based on the as-
sociations should be considered with great caution. Moreover, it is possible that studenteteacher relationship patterns and
motivation to defend change over time. Longitudinal studies are needed to better understand this. Furthermore, we have
delimited our focus in the current study on the quality of studenteteacher relationships as a contextual factor in school. Other
contextual factors that might be associated with students' sense of relatedness and autonomous motivation to defend victims
in bullying are, of course, childeparent relationships and peer relationships. Further research should examine these factors as
well as possible interaction effects between them.
Another limitation is that some of the scales had low Cronbach's alpha values, and some caution is warranted when
analyzing results that pertain to passive bystanding. However, the main results of the model relate to autonomous and
extrinsic motivation and the defending role, which had relatively high alpha values. Furthermore, this is the first time that
the Motivation to Defend Scale has been adopted in a study on bullying, and the scale might be improved by examining the
subscales. Further research on children's and adolescents' own voices on what they identify as motivating bullying (cf.,
Bosacki, Marini, & Dane, 2006) and various bystander behaviors in bullying (cf., Forsberg et al., 2014; Thornberg et al., 2012)
would also be helpful to develop more ecologically valid motivation scales. Because all constructs were measured with self-
reporting, the Motivation to Defend Scale and the bystander behavior scale might be measuring very similar constructs.
However, whereas the first scale was designed to measure types of motivation to defend (participants were asked to think
of situations when they had witnessed another student being bullied and reported why they would engage in helping a
victim followed by a list of items), the latter scale was designed to measure various behaviors as bystanders in bullying
situations (participants were asked to rate how often they had enacted the behaviors described in the scale). Despite these
limitations, the Motivation to Defend Scale might be used in research as well as in bullying prevention and intervention
programs.
Finally, a note of caution needs to be sounded regarding the generalization of the findings. This sample of early adolescents
from Italy might or might not be similar to the population of adolescents with whom the readers primarily work or are
interested in. Relating the current results to Hofstede's Model of Cultural Dimensions (Hofstede, 2001), it could be assumed
that in Italy, which is a masculine society with high power distance, higher occurrence of pro-bullying should be reported
because conflicts are more likely to be solved through aggressive behavior rather than negotiation. On the other hand, the
prevalence of school bullying has decreased from among the highest levels in Europe (Genta, Menesini, Fonzi, Costabile, &
Smith, 1996) to lower levels (Vieno et al., 2011), which might be due to the promotion, coordination, and supervision of
anti-bullying initiatives at the school and the community level in each Italian region. There is reason to believe that Italian
schools have policies related to bullying and that the teachers have had training in how to respond to school bullying. Thus,
future studies should try to replicate the our findings with other samples of students of different age levels and in various
cultural settings.
Despite these limitations, the current findings suggest that anti-bullying programs can counteract pro-bullying and
passive bystander behavior and encourage defender behavior among early adolescents by establishing warm and caring
studenteteacher relationships and promoting autonomous motivation to defend. Because previous research has revealed
that between-class variability in bullying can be partly explained by the prevalence of different bystander behaviors
(Nocentini et al., 2013; Salmivalli et al., 2011), efforts to influence bystander behavior are an important component of bullying
prevention. An example of this is the KiVa program, which is designed to reduce negative bystander behavior and increase
defender behavior by increasing students' empathy, self-efficacy, and anti-bullying attitudes (K€ arn€a et al., 2011, 2014). The
88 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

current study highlights the importance of also including a focus on studenteteacher relationship quality and autonomous
motivation to defend victims in the KiVa program. Our findings suggest that teachers should consciously and efficiently build
warm and caring relationships with their students in order to enhance students' autonomous motivation to defend, which in
turn appears to make them more prone to defend victims in actual bullying situations. In contrast, teachers should avoid co-
constructing conflictual relationships with their students because our findings suggest that such negative studenteteacher
relationships increase the risk of extrinsic motivation to defend, which in turn was linked to pro-bullying and passive
bystander behavior in actual bullying situations.

References

Bandura, A. (1999). Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 3, 193e209.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin,
117(3), 497e529.
Bellmore, A., Ma, T., You, J., & Hughes, M. (2012). A two-method investigation of early adolescents' responses upon witnessing peer victimization in school.
Journal of Adolescence, 35, 1265e1276.
Bentler, P. M. (1995). EQS 6 structural equations program manual. Encino, CA: Multivariate Software, Inc.
Bierhoff, H.-W. (2002). Prosocial behaviour. Hove: Psychology Press.
Birch, S. H., & Ladd, G. W. (1998). Children's interpersonal behaviors and the teacherechild relationship. Developmental Psychology, 34, 934e946.
Bosacki, S. L., Marini, Z. A., & Dane, A. V. (2006). Voices from the classroom: Pictorial and narrative representations of children's bullying experiences. Journal
of Moral Education, 35, 231e245.
Van den Broeck, A., Lens, W., De Witte, H., & Van Coillie, H. (2013). Unraveling the importance of the quantity and the quality of workers' motivation for
well-being: A person-centered perspective. Journal of Vocational Behavior, 82, 69e78.
Byrne, B. M. (1995). One application of structural equation modeling from two perspectives: Exploring the EQS and LISREL strategies. In R. H. Hoyle (Ed.),
Structural equation modeling: Concepts, issues, and applications (pp. 138e157). Thousand Oaks, CA, US: Sage Publications, Inc.
Cadima, J., Leal, T., & Burchinal, M. (2010). The quality of teacherestudent interactions: Associations with first graders' academic and behavioral outcomes.
Journal of School Psychology, 48, 457e482.
Colarossi, L. G., & Eccles, J. S. (2003). Differential effects of support providers on adolescents' mental health. Social Work Research, 27, 19e30.
Cook, C. R., Williams, K. R., Guerra, N. G., Kim, T. E., & Sadek, S. (2010). Predictors of bullying and victimization in childhood and adolescence: A meta-
analytic investigation. School Psychology Quarterly, 25, 65e83.
Craig, W. M., Pepler, D., & Atlas, R. (2000). Observations of bullying in the playground and in the classroom. School Psychology International, 21, 22e36.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. New York: Plenum.
Deci, E. L., & Ryan, R. M. (2000). The ‘‘what’’ and ‘‘why’’ of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11,
227e268.
DeCoster, J. (1998). Overview of factor analysis. Retrieved May, 22, 2016 from http://www.stat-help.com/notes.html.
Forsberg, C., Thornberg, R., & Samuelsson, M. (2014). Bystanders to bullying: Fourth- to seventh-grade students' perspectives on their reactions. Research
Papers in Education, 29, 557e576.
Genta, M. L., Menesini, E., Fonzi, A., Costabile, A., & Smith, P. K. (1996). Bullies and victims in schools in central and southern Italy. European Journal of
Psychology of Education, 11, 97e110.
Gini, G., Pozzoli, T., & Bussey, K. (2015). The role of individual and collective moral disengagement in peer aggression and bystanding: A multilevel analysis.
Journal of Abnormal Child Psychology, 43, 441e452.
Grant, A. M., & Berry, J. W. (2011). The necessity of others is the mother of invention: Intrinsic and prosocial motivations, perspective taking, and creativity.
Academy of Management Journal, 54, 73e96.
Gregory, A., Cornell, D., Fan, X., Sheras, P., Shih, T.-H., & Huang, F. (2010). Authoritative school discipline: High school practices associated with lower
bullying and victimization. Journal of Educational Psychology, 102, 483e496.
Guay, F., Ratelle, C. F., Roy, A., & Litalien, D. (2010). Academic self-concept, autonomous academic motivation, and academic achievement: Mediating and
additive effects. Learning and Individual Differences, 20, 644e653.
Guay, F., & Vallerand, R. J. (1996). Social context, student's motivation, and academic achievement: Toward a process model. Social Psychology of Education, 1,
211e233.
Hagger, M. S., Sultan, S., Hardcastle, S. J., & Chatzisarantis, N. D. (2015). Perceived autonomy support and autonomous motivation toward mathematics
activities in educational and out-of-school contexts is related to mathematics homework behavior and attainment. Contemporary Educational Psy-
chology, 41, 111e123.
Hamarus, P., & Kaikkonen, P. (2008). School bullying as a creator of pupil pressure. Educational Research, 50, 333e345.
Hamre, B. K., & Pianta, R. C. (2001). Early teacherechild relationships and the trajectory of children's school outcomes through eighth grade. Child
Development, 72, 625e638.
Hardy, S. A., Dollahite, D. C., Johnson, N., & Christensen, J. B. (2015). Adolescent motivations to engage in pro-social behaviors and abstain from health-risk
behaviors: A self-determination theory approach. Journal of Personality., 83, 479e490.
Harel-Fisch, Y., Walsh, S. D., Fogel-Grinvald, H., De Matos, M. G., & Craig, W. (2011). Negative school perceptions and involvement in school bullying: A
universal relationship across 40 countries. Journal of Adolescence, 34, 639e652.
Hofstede, G. (2001). Culture's consequences: Comparing values, behaviors, institutions, and organizations across nations. Thousand Oaks, CA: Sage Publications, Inc.
Hoglund, W. L. G., Klingle, K. E., & Hosan, N. E. (2015). Classroom risks and resources: Teacher burnout, classroom quality and children's adjustment in high
needs elementary schools. Journal of School Psychology, 53, 337e357.
de Hooge, I. E. (2013). Moral emotions and prosocial behaviour: It may be time to change our view of shame and guilt. In C. Mohiyeddini, M. Eysenck, S.
Bauer, C. Mohiyeddini, M. Eysenck, & S. Bauer (Eds.), Recent theoretical perspectives and novel empirical findings: Vol. 2. Handbook of psychology of
emotions (pp. 255e275). Hauppauge, NY, US: Nova Science Publishers.
Hoyle, R. H., & Panter, A. T. (1995). Writing about structural equation models. In R. H. Hoyle (Ed.), Structural equation modeling: Concepts, issues, and ap-
plications (pp. 158e176). Thousand Oaks, CA: Sage Publications, Inc.
Hughes, J., Cavell, T., & Willson, V. (2001). Further support for the developmental significance of the quality of the teacherestudent relationship. Journal of
School Psychology, 39, 289e301.
Jimerson, S. R., Swearer, S. M., & Espelage, D. L. (2010). Handbook of bullying in schools: An international perspective. New York, NY, US: Routledge/Taylor &
Francis Group.
Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed). New York, NY: Guilford Press.
Ka€rn€
a, A., Voeten, M., Little, T. D., Poskiparta, E., Alanen, E., & Salmivalli, C. (2011). Going to scale: A nonrandomized nationwide trial of the KiVa antibullying
program for grades 1e9. Journal of Consulting and Clinical Psychology, 79, 796e805.
Knee, C. R., & Neighbors, C. (2002). Self-determination, perception of peer pressure, and drinking among college students. Journal of Applied Social Psy-
chology, 32, 522e543.
 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90 89

Koomen, H. Y., Verschueren, K., van Schooten, E., Jak, S., & Pianta, R. C. (2012). Validating the student-teacher relationship scale: Testing factor structure and
measurement invariance across child gender and age in a Dutch sample. Journal of School Psychology, 50, 215e234.
Korthagen, F. A. J., Attema-Noordewier, S., & Zwart, R. C. (2014). Teacherestudent contact: Exploring a basic but complicated concept. Teaching and Teacher
Education, 40, 22e32.
Lucas-Molina, B., Williamson, A. A., Pulido, R., & Pe rez-Albeniz, A. (2015). Effects of teacher-student relationships on peer harassment: A multilevel study.
Psychology in the Schools, 52, 298e315.
Mantzicopoulos, P., & Neuharth-Pritchett, S. (2003). Development and validation of a measure to assess head start children's appraisals of teacher support.
Journal of School Psychology, 41, 431e451.
Mashburn, A. J., & Pianta, R. C. (2006). Social relationships and school readiness. Early Education and Development, 17, 151e176.
Merritt, E. G., Wanless, S. B., Rimm-Kaufman, S. E., Cameron, C., & Peugh, J. L. (2012). The contribution of teachers' emotional support to children's social
behaviors and self-regulatory skills in first grade. School Psychology Review, 41, 141e159.
Milatz, A., Glüer, M., Harwardt-Heinecke, E., Kappler, G., & Ahnert, L. (2014). The StudenteTeacher Relationship Scale revisited: Testing factorial structure,
measurement invariance and validity criteria in German-speaking samples. Early Childhood Research Quarterly, 29, 357e368.
Moran, C. M., Diefendorff, J. M., Kim, T., & Liu, Z. (2012). A profile approach to self-determination theory motivations at work. Journal of Vocational Behavior,
81, 354e363.
Ng, J. Y. Y., Ntoumanis, N., Thøgersen-Ntoumani, C., Deci, E. L., Ryan, R. M., Duda, J. L., et al. (2012). Self-determination theory applied to health contexts: A
meta-analysis. Perspectives on Psychological Science, 7, 325e340.
Niemiec, C. P., & Ryan, R. M. (2009). Autonomy, competence, and relatedness in the classroom: Applying self-determination theory to educational practice.
Theory and Research in Education, 7, 133e144.
Nocentini, A., Menesini, E., & Salmivalli, C. (2013). Level and change of bullying behavior during high school: A multilevel growth curve analysis. Journal of
Adolescence, 36, 495e505.
Olweus, D. (1993). Bullying at school. Cambridge, MA: Blackwell.
O'Connor, E. E., Dearing, E., & Collins, B. A. (2011). Teacher-child relationship and behavior problem trajectories in elementary school. American Educational
Research Journal, 48, 120e162.
Pellegrini, A. D., Long, J. D., Solberg, D., Roseth, C., Dupuis, D., Bohn, C., et al. (2010). Bullying and social status during school transitions. In S. R. Jimerson, S.
M. Swearer, & D. L. Espelage (Eds.), Handbook of bullying in schools: An international perspective (pp. 199e210). New York: Routledge.
Pianta, R. C. (1992). Beyond the parent: The role of other adults in children's lives. San Francisco, CA, US: Jossey-Bass.
Pianta, R. C. (2006). Classroom management and relationships between children and teachers: Implications for research and practice. In C. M. Evertson, & C.
S. Weinstein (Eds.), Handbook of classroom management: Research, practice, and contemporary issues (pp. 685e709). Mahwah, NJ: Lawrence Erlbaum
Associates.
Pianta, R. C., Nimetz, S. L., & Bennett, E. (1997). Motherechild relationships, teacherechild relationships, and school outcomes in preschool and kinder-
garten. Early Childhood Research Quarterly, 12, 263e280.
Pianta, R. C., Steinberg, M. S., & Rollins, K. B. (1995). The first two years of school: Teacherechild relationships and deflections in children's classroom
adjustment. Development and Psychopathology, 7, 295e312.
Pozzoli, T., Ang, R. P., & Gini, G. (2012). Bystanders' reactions to bullying: A cross-cultural analysis of personal correlates among italian and Singaporean
students. Social Development, 21, 686e703.
Pozzoli, T., & Gini, G. (2010). Active defending and passive bystanding behavior in bullying: The role of personal characteristics and perceived peer pressure.
Journal of Abnormal Child Psychology, 38, 815e827.
Pozzoli, T., & Gini, G. (2013). Why do bystanders of bullying help or not? A multidimensional model. The Journal of Early Adolescence, 33, 315e340.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, In-
struments & Computers, 36, 717e731.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models.
Behavior Research Methods, 40, 879e891.
Reeve, J. (2006). Teachers as facilitators: What autonomy-supportive teachers do and why their students benefit. The Elementary School Journal, 106,
225e236.
Richard, J. F., Schneider, B. H., & Mallet, P. (2011). Revisiting the whole-school approach to bullying: Really looking at the whole school. School Psychological
International, 33, 263e284.
Rigby, K., & Johnson, B. (2006). Expressed readiness of Australian schoolchildren to act as bystanders in support of children who are being bullied.
Educational Psychology, 26, 425e440.
Roorda, D. L., Koomen, H. M. Y., Spilt, J. L., & Oort, F. J. (2011). The influence of affective teacherestudent relationships on students' school engagement and
achievement: A meta-analytic approach. Review of Educational Research, 81, 493e529.
Roorda, D. L., Vershueren, K., Vancraeyveldt, C., Van Craeyevelt, S., & Colpin, H. (2014). Teacherechild relationships and behavioral adjustment: Trans-
actional links for preschool boys at risk. Journal of School Psychology, 52, 495e510.
Roth, G., Kanat-Maymon, Y., & Bibi, U. (2011). Prevention of school bullying: The important role of autonomy-supportive teaching and internalization of pro-
social values. British Journal of Educational Psychology, 81, 654e666.
Rudasill, K. M., & Rimm-Kaufman, S. E. (2009). Teacherechild relationship quality: The roles of child temperament and teacherechild interactions. Early
Childhood Research Quarterly, 24, 107e120.
Rueger, S. Y., Malecki, C. K., & Demaray, M. K. (2008). Gender differences in the relationship between perceived social support and student adjustment
during early adolescence. School Psychology Quarterly, 23, 496e514.
Rueger, S. Y., Malecki, C. K., & Demaray, M. K. (2010). Relationship between multiple sources of perceived social support and psychological and academic
adjustment in early adolescence: Comparisons across gender. Journal of Youth Adolescence, 39, 47e61.
Runions, K. C., Vitaro, F., Cross, D., Boivin, M., Shaw, T., & Hall, M. (2014). Teacher-child relationship, parenting, and growth in likelihood and severity of
physical aggression in the early school years. Merrill-Palmer Quarterly, 60, 274e301.
Ruscio, J., & Roche, B. (2012). Determining the number of factors to retain in exploratory factor analysis using comparison data of known factorial structure.
Psychological Assessment, 24, 282e292.
Ryan, R. M. (1982). Control and information in the intrapersonal sphere: An extension of cognitive evaluation theory. Journal of Personality And Social
Psychology, 43, 450e461.
Ryan, R. M. (1995). Psychological needs and the facilitation of integrative processes. Journal of Personality, 63, 397e427.
Ryan, R. M., & Connell, J. P. (1989). Perceived locus of causality and internalization: Examining reasons for acting in two domains. Journal of Personality and
Social Psychology, 57, 749e761.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American
Psychologist, 55, 68e78.
Ryan, R. M., Patrick, H., Deci, E. L., & Williams, G. C. (2008). Facilitating health behaviour change and its maintenance: Interventions based on Self-
Determination Theory. The European Health Psychologist, 10, 2e5.
Sabol, T. J., & Pianta, R. C. (2012). Recent trends in research on teacherechild relationships. Attachment & Human Development, 14, 213e231.
Salmivalli, C. (1999). Participant role approach to school bullying: Implications for interventions. Journal of Adolescence, 22, 453e459.
Salmivalli, C. (2010). Bullying and the peer group: A review. Aggressive and Violent Behavior, 15, 112e120.
Salmivalli, C. (2014). KiVaeagainst bullying. Psychologist, 27, 258e259.
90 T. Jungert et al. / Journal of Adolescence 53 (2016) 75e90

Salmivalli, C., Voeten, M., & Poskiparta, E. (2011). Bystanders matter: Associations between reinforcing, defending, and the frequency of bullying behavior in
classrooms. Journal of Clinical Child & Adolescent Psychology, 40, 668e676.
Sarkova, M., Bacikova-Sleskova, M., Madarosova Geckova, A., Katreniakova, Z., van den Heuvel, W., van Dijk, J. P., et al. (2014). Adolescents' psychological
well-being and self-esteem in the context of relationships at school. Educational Research, 56, 367e378.
Satorra, A., & Bentler, P. M. (2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66, 507e514.
Spilt, J. L., Koomen, H. M., & Mantzicopoulos, P. Y. (2010). Young children's perceptions of teacherechild relationships: An evaluation of two instruments and
the role of child gender in kindergarten. Journal of Applied Developmental Psychology, 31, 428e438.
Tangney, J. P., Stuewig, J., & Mashek, D. J. (2007). Moral emotions and moral behavior. Annual Reviews of Psychology, 58, 345e372.
Taylor, G., Jungert, T., Mageau, G. A., Schattke, K., Dedic, H., Rosenfield, S., et al. (2014). A self-determination theory approach to predicting school
achievement over time: The unique role of intrinsic motivation. Contemporary Educational Psychology, 39, 342e358.
Teixeira, P. J., Carraca, E. V., Markland, D., Silva, M. N., & Ryan, R. M. (2012). Exercise, physical activity, and self-determination theory: A systematic review.
International Journal of Behavioral Nutrition and Physical Activity, 9, 1e30.
Thornberg, R. (2010). A student in distress: Moral frames and bystander behavior in school. Elementary School Journal, 110, 585e608.
Thornberg, R. (2011). ”She's weird!” e The social construction of bullying in school: A review of qualitative research. Children & Society, 25, 258e267.
Thornberg, R., & Jungert, T. (2013). Bystander behavior in bullying situations: Basic moral sensitivity, moral disengagement and defender self-efficacy.
Journal of Adolescence, 36, 475e483.
Thornberg, R., & Jungert, T. (2014). School bullying and the mechanisms of moral disengagement. Aggressive Behavior, 40, 99e108.
Thornberg, R., Pozzoli, T., Gini, G., & Jungert, T. (2015). Unique and interactive effects of moral emotions and moral disengagement on bullying and
defending among school children. The Elementary School Journal, 116, 322e337.
Thornberg, R., Tenenbaum, L., Varjas, K., Meyers, J., Jungert, T., & Vanegas, G. (2012). Bystander motivation in bullying incidents: To intervene or not to
intervene? Western Journal of Emergency Medicine, 13, 247e252.
Thornberg, R., Wa €nstro€ m, L., Hong, J. S., & Espelage, D. (2016). Individual and class socio-moral influences on how to act as a bystander in school bullying
situations. In Paper presented at the 44th Congress of Nordic Educational Research Association at Helsinki, Finland, 9e11 March, 2016.
Tsang, S. K. M., Hui, E. K. P., & Law, B. C. M. (2011). Bystander position taking in school bullying: The role of positive identity, self-efficacy, and self-
determination. The Scientific World Journal, 11, 2278e2286.
re, N. M., Senecal, C., & Vallieres, E. F. (1992). The Academic Motivation Scale: A measure of intrinsic, extrinsic,
Vallerand, R. J., Pelletier, L. G., Blais, M. R., Brie
and amotivation in education. Educational and Psychological Measurement, 52, 1003e1017.
Vieno, A., Gini, G., & Santinello, M. (2011). Different forms of bullying and their association to smoking and drinking behavior in Italian adolescents. Journal
of School Health, 81, 393e399.
Wang, C., Swearer, S. M., Lembeck, P., Collins, A., & Berry, B. (2015). Teacher matter: An examination of student-teacher relationships, attitudes toward
bullying, and bullying behavior. Journal of Applied School Psychology, 31, 219e238.
Weinstein, N., & Ryan, R. M. (2010). When helping helps: Autonomous motivation for prosocial behavior and its influence on well-being for the helper and
recipient. Journal of Personality and Social Psychology, 98, 222e244.
West, S. G., Taylor, A. B., & Wu, W. (2012). Model fit and model selection in structural equation modeling. In R. H. Hoyle, & R. H. Hoyle (Eds.), Handbook of
structural equation modeling (pp. 209e231). New York, NY: Guilford Press.
Wubbels, T., Brekelmans, M., den Brok, P., Wijsman, L., Mainhard, T., & van Tartwijk, J. (2015). Teacherestudent relationships and classroom management. In
E. T. Emmer, & E. J. Sabornie (Eds.), Handbook of classroom management (2nd ed., pp. 363e386). New York, NY: Routledge.