ArticlePublisher preview available
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Social contagion theories suggest that adolescents in relationships with same-age high school dropouts should be at a greater risk of dropping out themselves. Yet, few studies have examined this premise, and none have considered all potentially influential same-age intimates, focusing instead on only either friends or siblings. Moreover, a key influence in adolescents’ social worlds, romantic partners, has been ignored. The goal of this study was to provide a comprehensive view of dropout contagion by considering occurrences of dropout among friends, siblings, and romantic partners. Data came from a sample of Canadian adolescents (N = 545) comprising one third of high school dropouts; a second third of carefully matched at-risk but persevering schoolmates; and a last third of average, not-at-risk students. As predicted, adolescents were at greater risk of dropping out when a member of their network had recently left school (i.e., in the past year, OR = 3.11; 95% CI [1.78, 6.27]), with independent associations of nontrivial sizes for occurrences of dropout among friends, romantic partners, and siblings (ORs between 1.97 [95% CI 1.25, 3.41] and 3.12 [95% CI 1.23, 11.0]). Moreover, adolescents seemed particularly at risk of quitting school (OR = 4.88; 95% CI [2.54, 12.5]) when their networks included more than one type of same-age intimate (e.g., a friend and a sibling) who had recently dropped out. Findings suggest that social contagion of dropout is a pervasive phenomenon in low-income schools and that prevention programs should target adolescents with same-age intimates who have recently left school.
Social Contagion and High School Dropout: The Role of Friends,
Romantic Partners, and Siblings
Véronique Dupéré
University of Montreal Eric Dion
University of Quebec at Montreal
Stéphane Cantin, Isabelle Archambault, and Eric Lacourse
University of Montreal
Social contagion theories suggest that adolescents in relationships with same-age high school dropouts
should be at a greater risk of dropping out themselves. Yet, few studies have examined this premise, and
none have considered all potentially influential same-age intimates, focusing instead on only either
friends or siblings. Moreover, a key influence in adolescents’ social worlds, romantic partners, has been
ignored. The goal of this study was to provide a comprehensive view of dropout contagion by considering
occurrences of dropout among friends, siblings, and romantic partners. Data came from a sample of
Canadian adolescents (N545) comprising one third of high school dropouts; a second third of carefully
matched at-risk but persevering schoolmates; and a last third of average, not-at-risk students. As
predicted, adolescents were at greater risk of dropping out when a member of their network had recently
left school (i.e., in the past year, OR 3.11; 95% CI [1.78, 6.27]), with independent associations of
nontrivial sizes for occurrences of dropout among friends, romantic partners, and siblings (ORs between
1.97 [95% CI 1.25, 3.41] and 3.12 [95% CI 1.23, 11.0]). Moreover, adolescents seemed particularly at
risk of quitting school (OR 4.88; 95% CI [2.54, 12.5]) when their networks included more than one
type of same-age intimate (e.g., a friend and a sibling) who had recently dropped out. Findings suggest
that social contagion of dropout is a pervasive phenomenon in low-income schools and that prevention
programs should target adolescents with same-age intimates who have recently left school.
Educational Impact and Implications Statement
An adolescent’s decision to leave high school without a diploma is unlikely to be made in a social vacuum,
especially in areas characterized by low graduation rates. Our interview-based findings indicate that in
comparison with students who stay in school, adolescents who drop out are at a greater risk of having a
friend, a romantic partner, or a sibling who had themselves recently left school without a diploma.
Dropout thus appears to be socially contagious over a short period of time. In order to prevent this
outcome, it may be necessary to limit its spread from one adolescent to another.
Keywords: high school dropout, friends, siblings, romantic partners, social contagion
High school dropout is associated with a host of negative
lifelong consequences, and thus represents a primary target for
prevention in many Western countries (Lansford, Dodge, Pettit, &
Bates, 2016;OECD, 2018). Even though national high school
graduation rates are generally on the rise and have reached 90% in
some countries (Snyder, de Brey, & Dillow, 2019), dropout rates
remain high in many disadvantaged communities (Nieuwenhuis &
Hooimeijer, 2016). Understanding the factors responsible for the
high prevalence of dropout in these communities appears espe-
cially urgent in today’s context of rising socioeconomic inequality
This article was published Online First June 18, 2020.
XVéronique Dupéré, School of Psychoeducation, Youth in Difficulty
University Institute of the Integrated University Health and Social Services
Center of the South-Center-of-the-Montreal-Island (CCSMTL), Public
Health Research Center of the CCSMTL, University of Montreal; Eric
Dion, Department of Special Education, University of Quebec at Montreal;
XStéphane Cantin and Isabelle Archambault, School of Psychoeduca-
tion, University of Montreal; Eric Lacourse, Department of Sociology,
University of Montreal.
Financial support for the preparation of this article was provided to
Véronique Dupéré by Canada’s Social Sciences and Humanities Research
Council; Fonds de recherche du Québec Santé and Société et culture and
Institut de recherché en santé publique de l’Université de Montréal.We
thank all the staff and students of the participating schools and school
boards, as well as other school professionals who made the project possi-
ble.
Correspondence concerning this article should be addressed to Véro-
nique Dupéré, École de psychoéducation, Université de Montréal, C.P.
6128, Succursale Centre-Ville, Montréal, QC H3C 3J7, Canada. E-mail:
veronique.dupere@umontreal.ca
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Journal of Educational Psychology
© 2020 American Psychological Association 2021, Vol. 113, No. 3, 572–584
ISSN: 0022-0663 http://dx.doi.org/10.1037/edu0000484
572
... It is worth noting that all studies relied on self-reported information provided by the respondents. In most cases, adolescents were asked in a school context to identify a number of (best) friends to construct a social network ( Some studies used indirect measures of deviant behavior by asking respondents to report on peer deviant behavior (e.g., Dijkstra et al. 2010;Lin et al. 2018;Dupéré et al. 2021). An advantage of this approach is that it could uncover additional information that a person would not reveal about oneself, which could limit the social desirability bias (Lin et al. 2018). ...
... al. 2019), which could result in respondents withholding incriminating information about their friends, being unaware of certain behaviors (e.g.,Chen et al. 2021;Weerman and Smeenk 2005;Young et al. 2011), or even exaggerating the deviant behavior of their friends(Young and Weerman 2013).The impact of social network compositions of adolescent networks on deviant influence While most studies used network structure measures with nodes and edges, Bayram Özdemir et al. (2018) and Smith and Ecob (2013) focused on social influence by measuring the similarity of criminal behavior between individuals and their friends.Moreover, few studies have primarily examined descriptive information such as the number of edges(Dupéré et al. 2021;Lin et al. 2018;Widdowson et al. 2020) and the group size(Rees and Pogarsky 2011).Peer influence and popularityAlmost 25% of the studies have analyzed the popularity of the individual nodes (e.g.,Dijkstra et al. 2012;Burt and Rees 2014;Stogner et al. 2014;Turanovic and Young 2016;Copeland et al. 2019). In addition, five studies have examined the Bonacich centrality(Copeland et al. 2019;Osgood et al. 2013;Reynolds and Crea 2015;Schreck et al. 2004;Stogner et al. 2014). ...
... or high school dropout(Dupéré et al. 2021), while another may focus on violent offending and victimization(Turanovic and Young 2016).Moreover, the results indicated variations in the utilization of either direct deviance measures or indirect measures among studies. This limits the comparability since the interpretation of deviance and perceived deviance can yield different conclusions(Young et al. 2015). ...
... Both also limit chances of creating diverse and beneficial peer networks. Whether friends, romantic partners and siblings choose to leave school may also influence an individual's decision to do so [23]. The studies affirm that there are significant interrelationships between opportunity structures and post-school transitions in communities of cumulative disadvantages. ...
... Growing up in a community with a large proportion of low-income families coincides with and reinforces known contributors to dropping out of high school, such as emotional stressors, low levels of parental education, involvement in social exclusion or bullying, and a higher proportion of friends, siblings and romantic partners who also drop out of high school [2][3][4]23]. Moreover, social exclusion increases the likelihood of future unemployment as competition for unskilled work in Western countries increases [1,25,38]. ...
Chapter
Full-text available
Lowering high-school drop-out rates is regarded as a significant challenge in Western countries, as high levels of drop-out create problems both for the individuals affected and for society at large. Dynamics that may contribute to increased drop-out rates are heightened in communities with a prevalence of young people with low socio-economic status, yet the potential benefits of enhancement of strategic skills as a preventative measure directed at youth from disadvantaged communities have received scant attention. This chapter draws on a community-based policy scheme that provided leadership experience for youth from a disadvantaged community by hiring them to fill temporary job vacancies and employing them as leaders. Data are drawn from focus-group interviews with youth aged 13–19 who participated in the 2022 scheme. The data suggest that the acquisition of leadership skills through this temporary vacancy scheme holds the potential to counteract individual ramifications of social exclusion conducive to drop-out. Acquired skills included counteracting destructive group behavior and promoting inclusionary group dynamics. Participants also gained skills in decision-making, completion of tasks, and gaining and maintaining formalized authority. Such learning processes hold the potential to increase self-efficacy, to develop personal agency and support advantageous post-school transitions.
... Drawing on these theories, a number of studies have documented that siblings who are alike in terms of demographic characteristics are more likely to be a stronger source of contagion and role models and pave the way for one another. For example, siblings with a smaller age difference have a higher chance of influencing each other regarding dropping out of high school [31], leaving the parental home [7], and entry into parenthood [14]. Same-gender siblings are also more similar with respect to internal migration [30] and patterns of family formation [15]. ...
... Same-gender siblings are also more similar with respect to internal migration [30] and patterns of family formation [15]. One of the reasons sibling similarity is related to life-course resemblance is their relationship quality and emotional closeness [13,31,32], that siblings similar in age shared more experiences during childhood and could thus enhance relationship closeness in adulthood [33]. Other studies indicated that gender similarity is associated with increased sibling support and interaction [34,35]. ...
Article
Full-text available
Empirical evidences on intragenerational transmission of life course have been demonstrated and that interpersonal similarity may moderate the effect. In particular, siblings who are more similar in their demographic characteristics are more likely to follow each other’s life course transitions. Focusing on parental home-leaving and building upon the social influence processes and similarity-attraction effects, this study investigates whether the association between siblings’ departures from the parental home increases when they are similar in the Big Five personality traits, like similarity in demographic traits. We use 28 waves of a longitudinal sample from “Understanding Society: The U.K. Household Longitudinal Study”. The results of the multilevel discrete-time event-history analysis (N = 3,717 children) indicate that the association between leaving of a sibling and oneself was strengthened when they had a similar level of extraversion, particularly when they were both introverts. This implies that although introverted adolescents and emerging adults might take less initiative regarding social relationships and be more hesitant in their transition to adulthood, when a similarly introverted sibling makes such a transition, they are more inclined to do so. To conclude, the study uncovers the relationship between siblings’ personality similarity and their resemblance in nest-leaving, which helps explain young adults’ home-leaving decision in an era when delayed leaving is observed.
... The integrative stress process (the life course theory of dropout) (Dupéré et al., 2015) explains dropout as a consequence of the coalescing of several events triggering the process of leaving school. This can be, for example, the chronic absence of students with long-term illnesses or mental health problems (Dupéré et al., 2015;Fortin et al., 2006) or a type of social infection when close friends leave school (Dupéré et al., 2021). This configuration of stressors can include both immediate factors and factors that may not appear serious but that persist throughout life and can make a student more at risk of failing at school (Dupéré et al., 2021). ...
... This can be, for example, the chronic absence of students with long-term illnesses or mental health problems (Dupéré et al., 2015;Fortin et al., 2006) or a type of social infection when close friends leave school (Dupéré et al., 2021). This configuration of stressors can include both immediate factors and factors that may not appear serious but that persist throughout life and can make a student more at risk of failing at school (Dupéré et al., 2021). ...
Article
Full-text available
The goal of this review study was to map the current state of knowledge of early school leaving typologies (ISCED 3) and to provide an overview of the findings. Using systematic mapping, ten texts published in peer-reviewed journals between 2000 and 2021 were identified and analyzed. Although the researchers applied different theoretical concepts, five significant, recurring, and distinguishable types of early departures were identified in these studies. The interest in this specific area has been growing in recent years and the number of qualitative research studies on the phenomenon is also increasing. At the same time, however, there is not enough knowledge convincingly explaining the circumstances leading to early school leaving for the individuals in the largest group-the quiet type.
... CSE traits also involve favorable perceptions of an individual's responsiveness to the immediate environment . Although social contagion theory does not directly provide insights into specific personal traits that could impact the ways in which different employees react to leaders' actions and external environments, this theory has subtle implications, and related studies have suggested that individuals who exhibit differing characteristics may perceive such contagion differently (Bade, 2022;Xia, 2023;Dupéré et al., 2021;Özkan et al., 2023). People who exhibit high levels of CSE are dispositionally prone to feel that they have control over external environments (Booth et al., 2020;Hu et al., 2021;Wang & Xu, 2019;Tang et al., 2023;Chen et al., 2024); thus, they are expected to be intrinsically motivated to pursue their goals and to be resilient when encountering challenges and obstacles . ...
Article
Full-text available
Work‒life balance is becoming increasingly important to employees in contemporary society, who face demands from the different roles that they play in life. Drawing on social contagion theory, this paper investigates how employees’ work‒life balance can be improved by examining relational perspectives of leadership and human energy. The findings from a three-wave survey of 431 employees demonstrate that relational leadership exerts a positive effect on employees’ work‒life balance through employees’ relational energy. Core self-evaluation strengthens the mediating effect of relational energy on the relationship between relational leadership and work‒life balance such that this mediating effect is stronger at high level of core self-evaluation. This paper offers both theoretical contributions by confirming the novel energizing pathway associated with employees’ work‒life balance and the corresponding boundary condition (i.e., employees’ core self-evaluation) and managerial implications pertaining to how employees’ work‒life balance can be enhanced through managers’ leadership practices.
... As explained by SDT, motivations drive human actions and therefore learning effectiveness, social interactions, campus life experiences, physical wellness, technological learning and green attitude significantly drive student engagement and consequently drive students' motivation to attend lectures enthusiastically. Studies find that the student dropout phenomenon is influenced by several intrinsic factors, namely, low student engagement, decreasing motivation and insufficient social interaction (Dup er e et al., 2018(Dup er e et al., , 2019(Dup er e et al., , 2021. ...
Article
Full-text available
The current research aims to examine the factors that influence student engagement after the COVID-19 pandemic. This study applied a survey , using an online questionnaire as a data collection method. The target respondents were both undergraduate and postgraduate students who experienced online learning during the pandemic and returned to face-to-face learning after the pandemic. The sampling method was a judgmental sample, with a total of 292 students participating in this survey. The measuring items had seven-point semantic differential scales and Likert scales. To analyze the data, this study used descriptive statistics, exploratory factor analysis and multiple regression analysis to examine the effects of eight identified factors on student engagement. Then, the effect of student engagement on the intention to attend lectures was tested. The results showed that six variables, namely, learning effectiveness, social interaction, campus life experience, physical wellness, technological learning and a green attitude, positively influence student engagement. However, flexibility and efficiency did not have a positive influence. Lastly, student engagement significantly affected the intention to attend lectures. The findings of this study complemented the student engagement literature. The results showed that higher education institutions must focus on managing student engagement by considering the students’ motivational factors of learning.
... The materials and analysis code for this study are not available due to ethical reasons related to the confidential nature of the interview material. This study's design was reported in previous publications Dupéré et al., 2021). The analyses were not preregistered. ...
Article
Full-text available
“Diseases of despair,” most prominently depressive and substance-related problems, diminish the prospects of many young adults, especially those with lower levels of education. Yet many young adults in that situation avoid these problems. Close relationships are thought to be a key factor underpinning risk and resilience among this group. To examine this premise, this study explored links, beyond potential confounders assessed in adolescence, between strengths and strains in the social domain and markers of despair in the early and mid-20s in a Canadian sample overrepresenting youth without postsecondary credentials (N = 543, 52% male, 23% non-White). Having a good general ability to maintain supportive relationships was associated with fewer depressive and substance-related symptoms. Furthermore, support and less strain in relationships with parents and friends were associated with fewer depressive symptoms in the early and mid-20s, while for romantic relationships, associations with depressive symptoms became more consistent and robust in the mid-20s. Therefore, support and strains in romantic relationships gradually gained prominence as a determinant of depressive symptoms as youth advanced in adulthood. For substance-related problems, significant associations emerged for strains in friendships only, in both the early and mid-20s. These results suggest that multiple aspects of relationships pose a risk for or protect from the development of diseases of despair in early adulthood and that their relative importance changes depending on outcomes and developmental timing. Thus, facilitating meaningful social connections and reducing the prevalence and impact of relationship strains might contribute to preventing diseases of despair among youth with lower educational attainment.
Article
In the evolving education landscape, this survey investigates the integration and transformation of educational paradigms using social network analysis (SNA). This paper examines the fundamentals of SNA, including nodes, edges, centrality metrics, and network dynamics, for a comprehensive understanding of the education domain. It guides researchers through various applications of SNA in education, such as student-teacher networks and institutional collaborations, highlighting the advantages and challenges of these complex interactions. The paper assesses the methodologies used in educational SNA, including data collection strategies and the associated ethical considerations. The survey also discusses various case studies and applications where SNA facilitates well-informed decision-making, enhanced academic collaboration, and the evaluation of student performance. This paper focuses on the transformative potential of SNA and acknowledges the limitations, ethical dilemmas, and technological challenges in the field. It concludes with a forward-looking perspective on the future of SNA in education, showcasing supportive technological advancement. This survey highlights the evolution of SNA since its incorporation into educational research and practices.
Article
Full-text available
In almost all OECD countries, a significant proportion of students drop out of their studies and, despite a long tradition of research, the influence of a partnership has hardly been analyzed in this context. This paper examines whether the expectations and demands (the realistic and idealistic aspirations) of the partner influence the student’s dropout intention. In a first step, relationships are investigated on the basis of a pooled data set. The results clearly show that there is a negative correlation such that when the partner has higher realistic aspirations the student is less likely to have high dropout intentions. In a second step, possible dynamics associated with the course of studies are considered. By means of fixed-effects regressions it can be shown that when the partner’s realistic aspirations increase, the dropout intention decreases. This effect proves to be stable. With regard to idealistic aspirations, however, none of the models show negative effects. The analysis shows that the partner is important for the course of studies and should also be considered.
Article
Full-text available
Prior studies have noted several risk and protective factors for school dropout; however, only a few have examined longer-term vulnerabilities alongside temporary risk and protective factors. Consequently, we focused on the role that both stable and time-varying psychosocial risk and protective factors play in dropout intentions and actual dropout, using a 4-year longitudinal design. We investigated to what extent dropout intentions and dropout can be predicted by an interplay between negative life events, general self-efficacy, and perceived social support. We distinguished between time-averaged levels of self-efficacy and social support, and within-person change in self-efficacy and social support over time. This enabled us to establish whether dropout intentions and dropout were sensitive to fluctuations in perceived self-efficacy and social support over time when controlling for person-specific levels of these psychosocial resources. Calculating multilevel models with data from a prospective cohort study (N = 4,956, 43% male), we found that negative life events were significantly associated with an increase in dropout intentions and the likelihood of school dropout. Furthermore, time-averaged levels of self-efficacy and social support, and a within-person (situational) increase in these characteristics relative to their time-averaged levels, were related to lower levels of dropout intentions but did not prevent dropout. The positive relationship between negative life events and dropout intentions was attenuated for individuals who perceived higher levels of self-efficacy than usual. Our findings suggest future research should further investigate time-averaged and situational psychosocial drivers of school dropout in combination.
Article
Full-text available
Despite the fact that grade retention is now seen as controversial in many quarters, it remains common practice in numerous countries. Previous research on the effects of grade retention on student development has, however, generated ambiguous results, particularly in terms of motivational outcomes. This ambiguity has been attributed in part to a lack of high-quality studies including a longitudinal design, a suitable comparison group, and adequate statistical control of preretention differences. Based on longitudinal data of N = 3,288 German students over 3 years of secondary school, we examined differences in their academic self-concept, scholarly interests, learning motivation, and achievement motivation between those being retained in the sixth grade (n = 61) and those of the same age being promoted annually. To account for confounding variables, we applied full propensity score matching on baseline measures of the dependent variables, as well as various other covariates that have been found to be associated with the risk of retention (e.g., cognitive ability, academic performance, and family background variables). Results reveal a steep decline in students’ academic self-concept, interests, and learning motivation during the last months spent in the original class, just before retention. For those measures that were available, negative effects were still partly significant after 1 year, but had diminished 2 years after grade retention. Contrary to predictions suggested by the big-fish-little-pond effect, we found no positive effects of retention on students’ academic self-concept.
Article
Full-text available
Survey response rates are especially low among young people, those with low levels of formal education and those who are not well integrated into society. These effects are amplified in panel studies, where similar factors influence the propensity of continued participation in survey waves. As a result, conducting a panel study on young, unemployed people is particularly challenging. The paper describes the difficulties associated with selection effects and panel mortality in this particular group and the remedies applied in the JuSAW project, where we conducted a panel survey among young adults aged 18-28 who became unemployed between May and September 2014. Based on our experiences and a detailed analysis of sample balance and panel stability, we develop recommendations for future studies.
Article
Full-text available
This article provides an overview of research on social contagion in the context of education. We highlight the importance of students' social interactions in school, considering contagion between peers and contagion from teachers to students, using a motivation perspective. The framework of contagion is introduced broadly, followed by a focused review on social contagion in school environments, both peer and teacher related. Then we introduce methodology for mapping behavior change to networks that are a direct representation of school cohorts. We argue that these different lines of research can be coherently interpreted from a motivation perspective, suggesting the critical role of motivation in social contagion, in the context of education. We highlight the limited amount of research on positive contagion effects and we call for further investigation into ways in which to increase the contagion of positive, academic behaviors. Finally, the neuroscience behind social contagion, both for the mechanisms and the interactions, is discussed.
Article
Children's neighborhood contexts are defined by rising socioeconomic inequality and segregation. This article reviews several decades of research on how neighborhood socioeconomic conditions are associated with children's development. The nonexperimental literature suggests that the most salient neighborhood socioeconomic condition depends on the outcome—disadvantage for social, emotional, and behavioral outcomes and advantage for achievement-related outcomes. Moreover, children's cumulative exposure to neighborhood socioeconomic conditions over the first two decades of life, and possibly especially in childhood, may matter most for later development. These findings are partially supported by the few experimental studies available, and across study designs, neighborhood effects are typically modest. In order to improve our understanding of this topic, we recommend methodologically rigorous designs—experimental and nonexperimental—and comparative approaches, particularly ones addressing the complexities of development in neighborhood contexts. To guide this research, we provide an integrated framework that captures a broad and dynamic perspective including macro forces, neighborhood social processes and resources, physical features, spatial dynamics, and individual differences. Expected final online publication date for the Annual Review of Developmental Psychology, Volume 1 is December 16, 2019. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
Article
Background Retrospective reports of lifetime experience with mental disorders greatly underestimate the actual experiences of disorder because recall error biases reporting of earlier life symptoms downward. This fundamental obstacle to accurate reporting has many adverse consequences for the study and treatment of mental disorders. Better tools for accurate retrospective reporting of mental disorder symptoms have the potential for broad scientific benefits. Methods We designed a life history calendar (LHC) to support this task, and randomized more than 1000 individuals to each arm of a retrospective diagnostic interview with and without the LHC. We also conducted a careful validation with the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition. Results Results demonstrate that—just as with frequent measurement longitudinal studies—use of an LHC in retrospective measurement can more than double reports of lifetime experience of some mental disorders. Conclusions The LHC significantly improves retrospective reporting of mental disorders. This tool is practical for application in both large cross-sectional surveys of the general population and clinical intake of new patients.
Article
The present research examines social influences on self‐reported frequency of drunkenness in a longitudinal sample of 1,439 adolescents (46% female, 90% White, mean age = 14 at baseline) with social network measures from friends, romantic partners, and romantic partners’ friends. We build on past research by addressing multiple mechanisms of social influence—peers’ frequency of drunkenness, alcohol‐related attitudes, and unstructured socializing—across relationship types. Adolescents’ drunkenness frequency increased when their friends’ and partners’ friends’ drunkenness frequency increased and when their romantic partners’ positive alcohol‐related attitudes increased. Furthermore, the association between unstructured socializing and frequency of drunkenness was stronger for older than younger adolescents. Results advance understanding of the social transmission of alcohol use in adolescence and inform intervention efforts.