Society has become thoroughly mediatised. Every aspect and part of society from the economy, politics and education to civil society and everyday social relations is immersed by media. Today we have the internet, smart phones, apps, social network services, blogs, email, and other social media platforms. Social media has brought changes to the way we inform, communicate with others, learn, play and socialize. Most people are quite well connected and communicate with others and while being

more »

... ted they obtain, organize, produce and share information on a regular basis. These common routinely activities generate a large amount of information and knowledge of different forms, much of it created by 'ordinary people'. This information is generally referred as digital social data which is potentially of great interest to social scientists (Sloan and Quan-Haase, 2017). From the emergence of the internet, both quantitative and qualitative research have been interested in analysing digital data for its endeavour. As social media have enlarged the size and variety of the traces of social actors' actions and expressions, the analytical possibilities available for social science researchers have been reshaped too. This has brought to the fore the necessity of methodological innovations and interdisciplinary collaboration for the study of social media data. Conversely, this digital turn has generated lively debates about its potential to know the contemporary social world. While at the beginning of internet research scholars tended to study social life online as a separated from 'real life', researchers disputed this and argued the need for online social life to be viewed as an integral part of social life (Beneito-Montagut, 2011). Nevertheless, linking social media data to what is going on offline is still one of the challenges of digital social science research. Most of the methodological innovations have been made regarding the analysis of big data and technological tools to make sense of it, but not much has been said about how to qualitatively analyse social media data. This chapter deals with the analysis of social media data and offers a framework which considers the limits and challenges that this kind of data poses. Rather than offering a prescriptive method, it attempts to document the reflective process to incorporate social media data to qualitative research projects considering how we extract meaning from them. So, it is presented as a toolkit to start thinking about the analytical process instead of a universally applicable model. The aim is to link the analysis of social media data to well-stablished research paradigms, instead of reinventing them. Modes and media. Multimodality and multimediality The first section brings some definitions to set together the differences and common grounds of two terms: multimodal and multimedia. We will also put these two concepts in relation to the digital context and explain the reasons why we use multimedia. Yet suggesting multimodal research as a useful analytical framework to analyse social media data. We first define modes and media. Modes are ways of re-presenting information −the semiotic means used to compose a text. Note that text is understood, in a wider sense, as any socio-cultural artefact or act that embeds communication. Examples of modes comprise words, sounds, still and moving images, animation and colour (see Bella Dicks' chapter). Media, on the other hand, are the tools and material resources used to produce and disseminate texts. Media are technologies, channels, practices and platforms which we use to disseminate share, obtain, store, organize and create information about the world around us through communication and interaction. Examples of media, both broad and narrow casting, include books, radio, television, computers, smartphones, paint brush and canvas, and the body. The internet and digital tools are usually referred as digital media. 1 Hence, although media and modes are different and independent of each other, the media used affects the ways in which meaning can be realized through modes -or each media affords certain possibilities for communication or action. Accordingly, these concepts are independent and interdependent from each other at the same time. A multimodal text is characterized by the use of several different modes. In multimodal research, communication and representation is more than language, it refers to semiotic approaches as in the analysis of colour, shape, image, gestures and gazes in communicative situations. It assumes that our languages and their modes (written, oral, visual, gestural) attach meanings (semantic, symbolic and affective) and are world making. Its major influence comes from social semiotics. Multimodal analytical approaches have provided concepts, methods and a framework for the analysis of multimodal data. 1 Also, they used to be called new media.