Replication data for: Comparing name generator designs in rural panel studies

Abstract

We conducted a two-wave personal network study in a rural Romanian community, interviewing the same participants (n = 68) using two name generators. Wave 1 employed a fixed-choice generator (n = 25) focused on emotional closeness; Wave 2 used a free-choice generator based on frequent interaction. We compared tie characteristics and assessed re-elicitation across waves. Alters who were kin, co-residents, or emotionally close were more likely to be re-elicited, regardless of generator type. These findings underscore the role of relational attributes in personal network elicitation and highlight design considerations for network studies in resource-limited, culturally distinct settings.

Key-words: personal networks, name generators, rural fieldwork, tie retention, panel studies, longitudinal design

This data package corresponds to the paper "Comparing name generator designs in panel personal network data: implications for alter re-elicitation in a rural field setting" (by MG Hâncean, J Lerner, C McCarty). The data were produced within the 4P-CAN project, HORIZON-MISS-2022-CANCER-01, project ID 101104432, programme HORIZON https://4p-can.eu/ This data package includes personal network information collected across two survey waves (Wave 1 and Wave 2). The study tracks focal participants (“egos”) and the individuals they name as social contacts (“alters”). There are 68 egos (study participants or cases) within the data package. Three data files are included in the present data package: Wave 1 (Wave 1 information; myData1.rds), Wave 2 (Wave 2 information; myData2.rds) and a data file that includes measurements from Wave 1 and Wave 2 (myData.rds). Each data file includes detailed information about egos and their alters, such as demographic characteristics (age, sex, education), family ties, co-residence, and interaction features such as closeness, meeting frequency, and relationship status

Key features:

• Ego-level variables include age, sex, and education;

• Alter-level variables include age, sex, education, relationship to the ego, and co-residence;

• Dyadic attributes include closeness, meeting frequency, and relationship status;

• Metadata include wave information, operator (interviewer) ID, and retention status across waves.

This data package is structured to support the replication of the analysis reported in the manuscript "Comparing name generator designs in panel personal network data: implications for alter re-elicitation in a rural field setting" (by MG Hâncean, J Lerner, C McCarty).

For the code, check the Supplementary Material ("supplementary_material_updated.html").

"Comparing name generator designs in panel personal network data: implications for alter re-elicitation in a rural field setting" (by MG Hâncean, J Lerner, C McCarty) is available as part of the "Replication data for: Comparing name generator designs in rural panel studies", MG Hâncean, J Lerner, C McCarty. Zenodo DOI: https://doi.org/10.5281/zenodo.15322439 (2026). Unlike the previous version, the current version includes updated versions of the supplementary material and README note. We note that the data files are the same as in the previous version of the "Replication data for: Comparing name generator designs in rural panel studies". 

This data package also includes the files used for the additional analyses (i.e., myData1_ego.rds, alterDataW11.rds, mynet_participants11.rds, myattr_participants.rds). All the files included in this data package are available in the .rds file format. These can be manipulated using RStudio (version 2024.12.1+563). For further details, please check the README file.