Thursday, 31 January 2013

A bright new look on people’s opinions

Vanessa Torres van Grinsven is a researcher at the University of Utrecht.

Opinion surveys: they are all around us. Who hasn’t received a request – by mail, telephone, email or through one of those annoying pop-up windows on websites? 

But, why do researchers even bother? Why are our opinions, our feelings and experiences even important for researchers?


Our opinions matter for a whole lot of purposes: brands perhaps would like to know if and how we appreciate their products. Decision-makers would like to know how we feel about certain policy issues. Social researchers would like to find out, how, for example, people that are ill experience the healthcare system, or how elderly feel about the care they are receiving at retirement homes. One way to find out about these opinions, feelings and experiences is to ask people, like by sending out surveys or doing interviews. 

There are though also other possible ways for doing opinion research. Nowadays, more and more people are becoming increasingly active on social media, like with platforms such as Twitter, Facebook, and fora on websites. They write, for example, about what they are doing (“Trying a new recipe for a delicious pie”), what they have experienced (“Guess what happened to me today…”), how they feel (“I feel lucky today”) and what caused those feelings (“…because I got promoted at my job J”). Much of what people write is publicly available – open for anyone to read.

These writings closely resemble day-to-day communications. The same way we have a chat with our neighbour in the morning when we rush out to school, our jobs, etc., we write things on our Facebook and Twitter accounts - small, casual, day to day chats.

Taken all together, these writings have become a useful source for social researchers that want to find out how people really feel and think about things, and how they experience them. I recently undertook research to explore people’s feelings and ideas about, and experiences with, a national semi-governmental organization. I wanted to do that without burdening them with the job of filling in a survey or having to take time to have an interview with me. Doing an analysis of postings on social media gave seemingly relevant insights into these feelings, ideas and experiences.

For my analysis, I was able to use all public posts for a period of about two years. Like this, I could apply the “principle of total accountability” [1], that is, accounting for all the instances of the phenomena under investigation, which aids in its objectivity and representativeness.

Though being an innovative approach, recently more and more researchers are using social media as an important data source. These data may not be representative for a whole society. But, at least, using and analyzing these new types of data can offer us a first image upon which to further construct our research.

On the other hand, social media postings may possibly be an even more realistic account of people’s opinions than surveys or interviews, as all these posts were written down spontaneously, without anybody asking for it. There is no influence of the researcher on what is posted. Besides, according to media researchers, the modern social media can be seen as a new cultural forum, and as such they may be a repository and a resource articulating and negotiating meanings and world views on behalf of the culture at large.

Modern social media open up new doors as a datasource for researchers, and create additional possibilities to listen to people and what they really want.


[1] Leech, G. (1992) ‘Corpora and Theories of Linguistic Performance’, in J. Svartvik (ed.) Directions in Corpus Linguistics: Proceedings of the Nobel Symposium 82, Stockholm, 4–8 August 1991, pp. 105–22. Berlin: Mouton de Gruyter.