1. Home
  2. Understand your audience
  3. Research methods
  4. Overview of Research Methods
  5. Social media research

Social media research

What is it?

Social media research refers to a set of tools and techniques of extracting and analysing data from social media channels and the Internet. These include, among other things, aggregating social data related to specific events or issues and supporting analysis along spatial, temporal, thematic, and sentiment/psychosocial dimensions.

Given its nature, social media research represents a useful research resource to understand what social media channels are being used, in what way, by whom and what is being said. Thus, social media research relates more to a method of monitoring and (naturally) observing the content and information produced in the Internet and popular social media channels, instead of a method for doing systematic reseach in the traditional way.



When to use it?

Social media research seems particularly suited to:

  • Learn how social information flows in different channels, target groups of individuals and/or geographical areas.
  • Identify new topics and trends in people’s interactions.
  • Gain insight of consumers' sentiments, attitudes, opinions, and thoughts about one or more specific issues. Possible examples are: products/brands; health threats; natural or non-natural disasters; politicians and political issues.
  • Gain insight into organization’s communication strategies, actions and other aspects with regard to one or more specific issues.
  • Have a real-time overview of people’s reactions during an episodic event(s) (e.g., a food crisis), that can occur for a few hours to various weeks.
  • Monitor consumers' reactions to a recurrent or chronic event(s) during an extended period in time (e.g., discourse with regard to salt intake, cholesterol, diabetes, etc.).
  • Identify potential influencers and opinion makers across the main social media platforms.




The use of social media research provides a number of advantages relative to other research methods, namely:

  • Opportunity to collect and analyse data in real-time.
  • Opportunity to collect a wide diversity of messages/information (e.g., from more spontaneous to more thoughtful reactions) produced by a heterogeneous group of people.
  • Opportunity to collect data corresponding to different moments in time, very quickly (either longitudinally or by selecting different collection points in time).
  • Online data/conversations seem to yield a higher level of honesty and transparency.
  • Allows access to a large sample of people, from different geographical areas, with relatively little practical restraints.
  • No need for data transcription.



Disadvantages and limitations

Social media research has some disadvantages and limitations relative to other types of research, namely:

  • Repeated messages (same message posted more than once by the same person or other people, such as retweets and shares, for example) can over represent the sample of messages and lead to biased interpretations, depending on the research goal.
  • Social media quantitative analysis does not completely draw on the richness of consumers' responses and, thus, they provide only a superficial analysis.
  • Deeper and richer qualitative analysis can represent a very time consuming task, given the size and the unstructured nature of the data.
  • Deeper and richer qualitative analysis may require investing in specific software programmes.
  • Loss of paralinguistic phenomena. Non-verbal communication plays a significant role in conveying meaning and emotion and its loss can affect the interpretation of messages. In accordance, although the analysis of some forms of written expression of emotions (e.g., emoticons) can give some insight in this regard, they still provide an incomplete view of the phenomena.
  • Generalization to the broader population is constrained, unless complemented with additional methods/data analysis procedures (e.g., Epidemiological studies).
  • Samples might be biased by excluding people that do not use social media or internet regularly and/or that are illiterate.
  • Samples might be biased by including the people that are more interested in the event or issue studied (self-selection).
  • The access to data is only possible for information that the user shares as public or provided an informed consent is given.