Why Name Generators with a Fixed Number of Alters may be a Pragmatic Option for Personal Network Analysis

Abstract

Social network analysis has grown exponentially in recent years, giving rise to methodological innovations in different scientific disciplines. In psychology, social network analysis has been incorporated into studies of individual personality differences and has generated novel areas, such as network psychometrics and network interventions. In community psychology, a recent review examined the use of network analysis in American Journal of Community Psychology publications (Neal & Neal, American Journal of Community Psychology, 2017, 60, 279). Based on their study, the authors advise researchers to avoid using the fixed-choice name generator when possible, as one of the five methodological recommendations proposed. In this essay, I explain how the recent increase of name generators with a fixed number of alters when studying personal networks is originally linked to an interest in describing structural properties. Second, I analyze the pragmatic contributions of this method: establishing a limit of alters a priori can entail advantages in terms of standardization and comparability of personal networks. Finally, to contextualize the methodological debate, I argue that personal networks represent the diversity of contexts in which the individual participates and are naturally integrated into community surveys.

Changes in Caregiver Personal Support Networks: Gender Differences and Effects on Health (CUIDAR-SE Study)

“… 2nd module: A name generation question that focused on identifying the people in their network (alters) belonging to the different areas of social life in which they are embedded (family, friends, co-workers, and neighbors, among others). Participants were asked to list a fixed number of 25 alters [ 29 , 30 ]. 3rd module: Variables regarding the composition of the personal network (alter char-acteristics) and variables related to social support (social support function).…”

“…2nd module: A name generation question that focused on identifying the people in their network (alters) belonging to the different areas of social life in which they are embedded (family, friends, co-workers, and neighbors, among others). Participants were asked to list a fixed number of 25 alters [ 29 , 30 ].…”

Migration‐Related Stressors and Suicidal Ideation in North Korean Refugee Women: The Moderating Effects of Network Composition

“…These people can include family, friends, coworkers, teachers, doctors, and so on.” The name interpreter then asked about the following information for each alter: type of relationship (spouse, parent, sibling, children, other family, coworker, friend, neighbor, church acquaintance, and other), gender, age, length of relationship (years), and nationality. We fixed the number of alters (i.e., network size) to five because prior research has stated that this can be useful for comparison and standardization in egocentric data while reducing respondent burden (Jariego, 2018) and that five alters are the most cost‐effective to record sociometric choices when collecting such data (Merluzzi & Burt, 2013).…”

Personal Support Networks of Young People with Mild Intellectual Disabilities during the Transition to Adult Life

“…Second module: Name generation question focused on identifying the people in their network (alters) belonging to the different areas of social life in which they are embedded (family, friends, fellow students, professionals, and neighbors, etc.). Participants were asked to name a fixed number of alters [58], specifically 25 alters, as this was considered an adequate number to obtain structural variables of the person's network [59]. Nevertheless, many of them struggled to name 25 alters-most of them managed to name between 13 and 25 alters.…”