Social Network Analysis and Education. A Systematic Review of the Field (accepted)

Grunddaten zum Vortrag

Informationen zur Veranstaltung

Zusammenfassung

Rationale Social network analysis has been common in educational research from its advent in Jacob Moreno’s seminal 1934 Who Shall Survive?, and the number of published articles has increased in the last decade.  Despite the increased interest and application of the method, few sources provide an overview of its use in educational research (e.g. Robins, 2015). Researchers must independently identify relevant previous studies and methods, and may miss high-quality studies or inadvertently duplicate studies already carried out. The current study aims to address this gap by investigating the number and characteristics of studies using social network analysis in educational contexts. Methods Using a selected Boolean search strategy, empirical, English-language studies were searched for in the "Web of Science" database. Studies published between 2009 and 2019 with a focus on actors in primary or secondary schools were included. Over 2,000 studies were identified, approximately 500 of which were included the final sample after further review. Characteristics of each study such as year and source of publication as well as author(s) and their affiliation were analyzed using bibliographic evaluation procedures (e.g. “bibliometrix”). Also, the abstracts were used for text analysis (e.g. “keyword screening”, “topic modeling” etc.) to identify topics within the subject area. Content obtained through bibliometric information abstracts and titles are categorized and presented inspired by Watling Neal (2019), as follows: -          How? How do the number of social network analysis studies and the journals in which they are published change over time? -          Where? In which countries and regions are studies conducted)? -          Who? Who is publishing these articles? What is the collaboration network between authors look like? -          What? What topics within education do these authors study? Can clusters be identified? Results Results for a preliminary sample of 140 studies demonstrate the growth of social network analysis in educational research. Over the last decade, the number of scientific publications has grown by almost 4 percent annually. The articles are mostly published in small teams (only 13 documents are single-authored articles; on average 2.81 authors per document; collaboration index =3.06). Studies meeting the inclusion criteria were most frequently published in the “Journal of Youth and Adolescence” and “Social Networks”. The collaboration network currently consists of 393 actors. It has 44 components; obtaining a mean component size = 7.07. Using the “tm” R package five thematic clusters were identified from the most frequent terms in the abstracts: 1) “health” (studies about e.g. drug abuse, mental health issues, sports); 2) “behavioral deficits” (studies about e.g. anti-social, aggressive behavior, bullying and victimization); 3) “students with SEN” (studies about e.g. autism, children with disabilities); 4) “academic achievement” (studies about e.g. knowledge transfers, cognitive skills and performances) and 5) “Intergroup relations” (studies about e.g. homophily effects for gender or ethnicity). The distribution of authors, collaboration networks, journals, topics, and regions, as well as implications for the use of social network analysis in education studies will be discussed

Vortragende der Universität Münster

Marticke, Sophie

Professur für Methoden der empirischen Bildungsforschung (Prof. van Ophuysen)