ACSPRI Conferences, RC33 Eighth International Conference on Social Science Methodology

Font Size:  Small  Medium  Large

Snowball Sampling in Online Social Networks

Mahin Raissi, Robert Ackland

Building: Law Building
Room: Breakout 5 - Law Building, Room 020
Date: 2012-07-11 03:30 PM – 05:00 PM
Last modified: 2012-06-12


As social interactions are increasingly conducted in social networking sites (SNS) such as, there are huge opportunities for social network researchers. SNSs record detailed information on the attributes of users and their relationships, with considerable overlap with real life social networks.
But there are many methodological and ethical issues associated with research using data from SNSs, ranging from how to collect the data to reliability and generalizability of results.
One of the core challenges with using SNS data for social network analysis is that it is typically impossible to define the population sampling frame, and hence researchers need to resort to non-probability based sampling methods such as snowball sampling. This paper outlines an approach for conducting snowball sampling using
We argue that some SNS features may actually help researchers to overcome some of the ethical and methodological challenges of snowball sampling. Using non-probability based sampling results in some biases in data and results especially when the collected data is networking behaviour. Some of SNS features can help researchers to avoid some inherent conflicts between ethical and methodological issues of snowball sampling like conflict between participants’ anonymity and non-representative volunteers recruited into the sample. In this way, SNS can encourage voluntary participation by acting as a legitimate proxy between researcher and participants (ensuring privacy); facilitate ensuring anonymity (compared to real life snowball sampling) by assigning unique ID to users and possibility of robot mediated data collection. We further contend that SNSs can potentially mitigate snowball sampling biases somehow by reducing dependence on respondents via providing archived real activities (rather than self-reported) and possibility of broadcasting the invitation to participation (act as a media) and the possibility of checking volunteers eligibility.
This paper will explore and discuss these challenges in the context of an Australian government-funded study into the role of online social networks on successful ageing.