Readings in DH, November 2014 (1)

Hsu, W. F. (2014). Digital Ethnography Toward Augmented Empiricism:  A New Methodological Framework. Journal of Digital Humanities, 3(1). Retrieved from

This paper addresses issues around digital ethnography, what does this mean as a method? Hsu’s paper is an attempt to address methodological aspects of digital ethnography, which she finds under-developed.

Computers offer not just new ways of looking at things, or new realms in which to examine them (the virtual, for example) they also change the scale at which we can sample when looking at culture(s). Hsu mentions approaches such as “ethnomining” and “Thick Data” (coined by Wang, a term that refers to the context of Big Data). However, this paper does not just address ideas of Big Data, computation methods can also be used to extract “boutique data”. (Aside: this is an accurate reflection of the kind of data that I am dealing with in my research.) Hsu suggests instead that small amounts of quantitative data (gathered digitally?) can be used to trigger questions that are then investigated using qualitative methods.

One thought that struck me as I read this article; I am not used to ideas of ethnography and empiricism together. This is probably because I associated empiricism specifically with quantification (and not with the collection of qualitative data). And because I always understood ethnography as a method of collecting data that defies quantification, as if that was its particular merit, I never associated empiricism with qualitative material. Yet empiricism is really about collecting data (of any type) based on experience or experimentation. However, it appears that qualitative researchers often feel that they don’t match up to empirical methods. [Berry (2011, 12) cites Latour as the authority that identified sociology as a discipline that seeks to become quantitative, but cannot achieve its aim.] Presumably the desire to carry out quantitative analysis is driven by the idea that such data is neutral (pure?). Yet any kind of data can be subject to collection biases (and so on). Hsu suggests that in conventional data analysis, quantitative data plays a passive role.

“To use a Foucauldian metaphor, data is disciplined by the principles of scientific method and its underlying positivist ideology” (Hsu 2014).

This sets up the opposition with new digital ethnographic data, where data is “active” and can facilitate the imagining, re-contextualizing, and extending of empirical knowledge. However, I remain uneasy about the idea that data, any data, can truly be regarded as passive.

Digital is not, however, only a means of gathering data, it is a means of exploring it, using the digital to discover new relationships. In Hsu’s case, she has explored ideas around music and place/space through web-scraping and mapping. This is a good example of where qualitative data can be used to gather spatial patterns in music (and trends amongst music fans) and where qualitative work can then take that further to go beyond space, to reveal details of relationships to place.

Digital is also, however, a means of data deception. Algorithms, working behind the scenes on websites, unseen, can affect how empirical knowledge is produced and transmitted, created. To counteract this, ethnographers need to:

“give serious considerations to software as infrastructure and materiality at the sites of …research”. (Hsu 2014) (My emphasis.)


Berry, D. (2011). The computational turn: Thinking about the digital humanities. Culture Machine, 12. Retrieved from

Hsu, W. F. (2014). Digital Ethnography Toward Augmented Empiricism:  A New Methodological Framework. Journal of Digital Humanities, 3(1). Retrieved from