By Taylor Sloan,
In honour of the first few snowflakes of the year (at least here in Alberta where winter starts in mid October and lasts until mid May) I’d like to talk about the snowball method of research and hope to get you thinking about the way we understand history as told by those who lived through it.Â
Snowball sampling is a method of data collection often used in the social sciences or in statistics gathering. The pool of participants is grown by having an initial set of participants recruit or refer other participants. It can also refer to meta or survey research which examines the prevalence or use of a particular theory within a discipline, similar to the way historiographical work looks into developments of schools of thought. The name of this method is meant to invoke the image of a snowball rolling down a hill, having size and momentum as it collects more snow. Also called chain sampling or referral sampling, it can be an excellent tool for qualitative study where respondents are asked to describe events or experiences and the effects these have had on their lives. Helen Ahn Lin uses this method in her sociological investigation into the narratives of racism and experiences of hate crimes by Asian Americans (Lin 2009). While Lin’s work deals with contemporary experiences her explanation of who she chose the initial subjects is quite similar to how a local historian (whether professional or layperson) might go about gathering oral historical accounts from order community members. Lin describes choosing her interviewees based on “travel convenience and the belief that they have important stories to share. “(Lin 2009) Â
The snowball method can offer the benefits of increasing an otherwise small sample size, particularly if the demographics are hard to reach out to. The disadvantage is that the proximity of subjects to one another can interfere with the objectivity of the study. Sample errors or biases can also be much more difficult for the researcher to detect and account for. Snowball sampling frequently uncovers characteristics not previously noted about a particular population, which can inspire routes of further research to determine causality or underlying causes of correlation. Â
When one attempts to uncover the past through living histories and oral histories, it's important to keep the benefits and risks of snowball effect in mind. Typically if you want to reach out to understand a period of recent history by speaking to those who lived through it you may begin with only a few respondents initially interested in speaking, but these individuals you can hope to will bring more of their cohorts forward. This allows for more perspectives to be heard and patterns can begin to emerge between different accounts. However it is important to realize that as participants generally refer family members, friends and close acquaintances, which can lead to a relatively homogeneous sample group. A further issue particularly to the oral historian is that those first few respondents who start the ball rolling me further are also more likely those already used to having their voices heard. Within the collection of recollections which makes up oral history responds may through reminiscing be reminded of more details This is why it is so important to seek oral histories from a broad range of social demographics and socioeconomic backgrounds, if you want to understand not just a segmented experience of history.Â
We are slowly getting to a place where in Canadian society we are seeking to hear more from the sides of history which have previously been unheard or ignored. Being mindful of the limitations to objective research, snowball sampling deserves a place in the social historians tool kit for its capacity to uncover significant narratives and empower a chain of voices.Â
Citations
Ahn, L. (2009) Beyond the Imitate Message: Understanding Hate Crimes as Message Crimes. In Perry, B & Levin B. Hate Crimes: Understanding and Defining Hate Crimes. Praeger Perspectives pp 107-121.