Student Background And Relationships At The Start Of University Studies

The importance of social networks and social relationships has been documented in a variety of settings. For students at higher education institutions, network ties have been found to be an important factor in achieving a sense of social belonging in the institutions (Tinto, 1975). This is especially true for students who are disadvantaged and marginalized (Bhopal, 2011). Networks, and network structure, have also been found to predict whether students complete their studies or drop out of school (Magnus por Torfason & Margret Sigrun Siguroardottir, 2017; Wilcox, Winn & Fyvie-Gauld, 2005), and relationships at the start of university studies influence the evolution over the entire course of studies and beyond (Kossinets & Watts, 2009; Rios-Aguilar & Deil-Amen, 2012; Cho, Gay, Davidson & Ingraffea, 2007; Zander, Brouwer, Jansen, Crayen & Hannover, 2018).Given these and other ways in which social networks affect the experience of attending a higher education institution, it is important to understand the interplay between the background of students, and the social networks that support their journey within the institution. Social networks at the beginning of a course of study are particularly important, as students with larger networks in the beginning tend to build on their connections, are less likely to form new connections (Zander, Brouwer, Jansen, Crayen & Hannover, 2018), and tend to form connections through shared friends when they do (Kossinets & Watts, 2009), which can lead them to lose out on opportunities for beneficial and meaningful relationships. This might mean that students miss out on the opportunity to connect with students they might have more in common with, but were not part of their extended network when they began their studies. For the study presented here, we administered a set of surveys to first year students at the University of Iceland. We collected responses from two of the five major faculty areas within the university, the School of Social Sciences (SSS, N=292) and the School of Engineering and Natural Sciences (SENS, N=350). In the survey, which was conducted in the second week of the first semester, we asked students to choose those of their fellow students that they knew when they started their studies, and offered them a list of all students in their respective schools to choose from.We also collected data on other variables, including whether they had moved to the Reykjavik capital area in order to attend the university (which coincided with whether their prior residence was urban or rural). In a subsequent analysis, we joined the survey data with background data on the students, including, gender, age, and which upper secondary school (referred to as "framhaldsskoli" in Icelandic) they had attended prior to starting their university.Our results document the significant way in which the upper secondary level school affects the social networks of first year university students and also indicate that the patterns are in some respects different for the two schools within the university. In particular, we found that students at the SENS had on average a greater number of ties than their counterparts at the SSS.We found that students that lived in the Reykjavik capital area, before they started their studies, had on average larger social networks than students who had moved from more rural areas. Such a difference, resulting from the background of students, could indicate a challenge that students from outside the capital area face in their higher education. Further analysis of the difference between students from the capital area and other students highlighted that there is significant variation due to which upper secondary schools they had attended.In addition to differences in the number of ties, we found evidence of significant homophily (ties to similar others) in the students' social network. We found this with respect to gender and age, but the strongest indication of homophily was with respect to students' upper secondary school. Gender homophily was similar for SENS and SSS students, but for other variables we found greater tendency for homophily among SENS students. In particular, we found that a tie was over 20 times more likely to be reported between two SENS students if they had attended the same upper secondary school than if they had not.These results indicate that student characteristics and backgrounds have a significant impact on their initial social network at their higher education institution. It is important to be mindful of this in designing policy, but the results also point to the need for further study of the evolution of social networks among higher education students, and of the impact that the networks have on academic achievement and retention.