Photo Credit: Freeimages.com/Lukasz Brzozowski
Article By Mark C. Pachucki, Ph.D.
The World Health Organization now recognizes social relationships as an important social determinant of health throughout our lives. Yet, the acknowledgement that social ties can shape our morbidity and mortality has been at times an uphill struggle. This is because the analysis of the effects of human relationships on our health sometimes requires either large or unusually complete datasets, and often, analytic techniques that make complicated demands on causal inference.
However, a rapidly-expanding body of research shows how examining interdependencies between two individuals who know one another can provide insights into how human health changes at an individual, small-group, and population scale. As such, a growing number of investigators are pursuing inquiry in this and related areas of complex systems science. As part of the University of Massachusetts, Amherst Department of Sociology and Computational Social Science Institute, Mark C. Pachucki, Ph.D. investigates how social relationships shape health throughout the life course.
Acknowledging that social ties can shape our morbidity and mortality has been at times an uphill struggle
The transition to adolescence marks an especially formative stage of human development wherein interpersonal awareness and changes in social interactions are happening at the same time as immensely complex hormonal, cognitive, and behavioral transformations. Because of this simultaneity, identifying mechanisms that may produce ill health can be difficult.
In a recent study in Social Science & Medicine, Pachucki, Emily J. Ozer, Ph.D., Alain Barrat, Ph.D. and Ciro Cattuto, Ph.D. investigated the role of changing social relationships in early adolescent depressive symptoms and self-esteem. In the longitudinal network-behavior research design, we gathered information on simultaneously changing social structure and individual attributes among a bounded community of 6th grade participants (11-12 years). Then, we analyzed the resulting data using a sophisticated class of statistical social network model.
One of the novel aspects of this study was the dynamic measurement of children’s interaction patterns during a three-month period. We chose the school lunchroom as the site to observe networks, because, by and large, students tend to sit with their closest friends during their limited free times during the day. We identified social interactions using an unobtrusive electronic proximity-sensing radio frequency identification (RFID) tag worn on a neck lanyard. Tags sent data back to small wall-mounted receivers that tracked which tags were in close proximity every twenty seconds. Thus, given students’ propensity to switch tables (and thus lunch partners), we were able to observe who spent time with whom in fine detail. To these data we added information on children’s socioeconomic background, health attributes, and health behaviors.
A novel aspect of this study was the dynamic measurement of children’s interaction patterns during a three-month period
We observed that girls who had more interaction partners tended to have fewer depressive symptoms and greater self-esteem within measurement periods, but no parallel trend for boys. However, after controlling for socioeconomic and demographic background as well as changes in network structure, socially-connected study participants did not become more similar in terms of depressive symptoms nor self-esteem over time. This was an important finding because of the rigorous analytic methods employed, yet somewhat surprising because this stage of the life course is thought to be ripe for evidence of interpersonal social influence among peers. Scrutiny of network tie changes found that almost two-thirds of 6th-graders interaction partners changed each month, which is more turnover in this age group than has been shown using more traditional means of enumerating networks. We also found, as others have, that gender similarity and similar levels of popularity play important roles in friendship formation among this age cohort, and that friends of a given child tend to become friends as well (transitive closure).
Girls with more interaction partners had fewer depressive symptoms, greater self-esteem. There was no parallel trend for boys
We are unambiguously living in an era of computational social science, where massive streams of real-time data on human behavior are increasingly amenable to analysis to improve our understanding of how social processes contribute to health from birth until our twilight years. This example of research on objective measurement of minute-by-minute socialization behaviors is one way to illustrate the subtleties of adolescent relationship dynamics.
Yet given the ubiquity of portable devices and advances in Bluetooth and RFID technologies, conducting this type of research is becoming more feasible and widespread. Reliance upon social network data to construct relational datasets that reflect who people actually interact with – rather than simply who people state their friends are – has great potential to clarify how socialization processes unfold and moreover, how social processes interact with biological processes of human development.
For instance, Pachucki, in collaboration with Lindsay T. Hoyt, Ph.D., is currently integrating information on changes in human relationships with information on pubertal development during adolescence and disparities in cardiovascular diseases during adulthood. Identifying how relationships matter to future health, but critically – at what stages of life they may matter the most – are worthwhile questions to pursue because they may help us to identify appropriate interpersonal levers for intervention to improve health and well-being.
Acknowledgements: Pachucki gratefully acknowledges the participation of the 6th-graders and teachers who agreed to participate in this research, and is grateful for the financial support of the Robert Wood Johnson Foundation Health & Society Scholars program.
About the author
Mark C. Pachucki, Ph.D. is an Assistant Professor in the Department of Sociology and Computational Social Science Institute at the University of Massachusetts, Amherst. Previously, Pachucki was Instructor in Medicine and Pediatrics on the faculties at Massachusetts General Hospital and Harvard Medical School, and a Robert Wood Johnson Foundation Health & Society Scholar at the University of California, San Francisco and UC Berkeley. Dr. Pachucki received his doctorate in Sociology from Harvard University in 2010 with a focus on social networks, medical sociology, and culture.