Interaction Human-Computer COMS21301 Methods 1: Observation and Ethnography
Dr. Mike Fraser [e-mail]
Introduction
n
n
Observation-based studies have a longer history than you might think Just standing around watching and trying to remember what your impressions were is not generally considered enough for rigorous data collection you will need to actually note down or collect materials somehow
Ethnography
n n
Long history with its roots in anthropology Is it a method? q certainly NOT a methodology q does not predefine a theoretical or conceptual status q has been used in support of many methodological standpoints
The study of interaction
n
Ethnography & the analysis of work
q q
q q q
assumes human activities are socially organised commits to inquiring into patterns of interaction and collaboration relies on an observer going into the field learning the ropes through questions listening, watching, talking, etc., with practitioners
The basic principles of ethnography (1)
n
Ethnography is naturalistic
q
q
studies should be studies of real people and their activities, operating in their natural environment doesnt deal with artificial worlds & controlled versions of work (but what is artificial?)
Michael Lynch: "Stop talking about science. Go to a laboratory - any laboratory will do - hang around a while, listen to conversations, watch the technicians at work, ask them to explain what they are doing, read their notes, observe what they say when they examine data, and watch how they move equipment around..."
q
The basic principles of ethnography (2)
n
Ethnography is prolonged (usually)
q
q q
no logical reason why an ethnography should take a long time some practical reasons: The main reasons for prolongation
n n n
ethnographers have no clear idea what they will find domain may be technical focus may change
The basic principles of ethnography (3)
n
Ethnography understands the world from the point of view of those who inhabit it
q
q
q
it is behavioural, i.e. interested in the detail of the behaviour to a greater or lesser extent it is not behaviourist, i.e. it does not consider the behaviour itself as the appropriate level of analysis The appropriate level is the significance of the behaviour for those who undertake it.
The basic principles of ethnography (4)
n
Ethnographic data is eclectic
q q
resists formalisation can include:
n
n n
general descriptions of behaviours, descriptions of physical layouts, close descriptions of conversation thoughts and feelings, work sequences, anecdotes examples, common occurrences, hypotheses?
q
difficult to distill data down to an 'essential' form analysis is critical
Ethnographic data
n
n n
n n n
descriptions of conversation and movement, stories, opinions, anything! granularity issues ethnographer organises the data as showing something readability issues contain a number of technical terms The ethnographer usually provides 'examples' of something
Types of Ethnography
n
Consecutive/concurrent ethnography
q
ongoing ethnographic study taking place before, or at the same time as, systems development to provide a general but informed sense of the setting for designers to verify or validate a set of already formulated design decisions to inform initial design thinking
n
Quick and dirty ethnography
q
n
Evaluative ethnography
q
n
Re-examination of previous studies
q
Consecutive v concurrent ethnography
n n
n
n
Sequenced process Ethnographic study precedes system development Thorough insight into the subtleties of the organisation Declining rate of usefulness for fieldwork to contribute to design
n n
n
n
Concurrent process Ethnographic study goes hand-in-hand with system development Highlights key surface features of organisation High contribution of fieldwork to design per hour spent
'Quick and dirty' ethnography
n n
n
n
n
Provides knowledge of large scale organisations work setting in relatively short space of time Pay off greater as time expended on fieldwork provides a lot of information Knowledge can be built upon for more focused study of detailed aspects of work Provides broad overview which can sensitise designers to issues of acceptability and usability of an envisaged system Can provide an informed sense of what the work is like in a way that is useful for designers
Evaluative ethnography
n n
n
n n
n
More focused version of quick and dirty Does not necessarily involve a prolonged period of fieldwork Sanity check of already formulated design proposal Useful in tweaking existing systems Useful for informing design of next generation of systems Modest redesign through periodic ethnographic field studies of system use may have considerable benefits
Re-examination of previous studies
n
n
n
n
New approaches, new methods, new systems lack experience and a set of case studies which can be used either to sensitise study or to inform preliminary design Especially useful when prime goal is obtaining sight of general infrastructural principles A way of sensitising designers to social nature of settings Performs useful role in making designers aware of what to avoid and what the more specific issues might be
Practical issues with doing ethnography (1)
n
Do not read methodology books
q
Ethnography is not an esoteric procedure nor is it searching for things that are hard to find Much of ethnography does involve hanging around, but this is not the point, but a means of achieving the objective of uncovering the social organisation of the workplace
n
Hanging around is not the point
q
n
Shut up and listen
Practical with issues doing ethnography (2)
n
Access
q q
q
q q
q
Getting in Gatekeepers can be managerial sections, admin staff, shop floor workers Some areas might be regarded as off-limits to observers Observer might be associated with vested interests Ethnographers must gain acceptability - open and honest, showing respect Gaining credibility may include working shifts, sharing conditions, a non-intrusive demeanour (without being too self-effacing), sharing dress codes
The limitations of ethnography
n
What ethnography cannot do
q
q
q
Is that it then. Over-expectations as to how much can be found out in the time available Well, we knew that. False expectations as to the kind of information that fieldwork can reveal, and complaints that it does not provide blinding insights into the organisation of work Tell me what to build. False expectation that fieldwork can Directly answer design problems
Doing ethnography (1)
n
Beginning fieldwork
q
You arrive, tape recorder in hand, with a grin rigidly planted on your face. You probably realise that you have no idea how your grin is being interpreted, so you stop and nervously attempt a relaxed pose. Then you realise you have no idea how that is being interpreted. Soon you work yourself into the paralysis of the psychiatrist in the strip joint - she knows she cant react, but she knows she cant not react. It is little wonder that sometimes people hide in a hotel room and read mysteries. (Agar, 1980)
Doing ethnography (2)
n
Initial phase of organisational ethnography
q q
q
q
q q
familiarising yourself with the organisation characterised by the frantic collection of every piece of information that you are exposed to make notes on everything that you see and hear, on what youre told directly and on what you hear on the other side of the room sketch plans of office spaces and desktops; glance at official documents and scribbled notes tape record anything that you can start anywhere you can
Focus of the study
n
n
n
n
Choice between the innocent ethnography and informed ethnography How theoretically, strategically (etc.) informed should you be at the outset? One choice is progression from one to the other Full participation or observation only?
What data should you collect? (1)
n
n
n
n
Data collection is the least of the problems of ethnography Information is laying around in plain sight but no one has bothered to collect it up There is nothing special to look for, nothing to find that is hidden an illustrative list of the sorts of things that can be collected and recorded:
What data should you collect? (2)
n n n n n n n
conversations; descriptions of activities diagrams of places; descriptions of places sociograms: who talks to whom and when jokes; interviews; job descriptions memos, notices, graffiti happenings; transcripts of meetings; forms war stories
Technical support for ethnography (1)
n
n
Technologies must support analysis, not vice versa Video
q q q
q
Data can be analysed repeatedly off-site Time-consuming data collection and analysis Difficulty setting up and using equipment in some domains Missing visual actions that are peripheral to the video field-of-view
Technological support for ethnography (2)
n
Tape recording
q
q q
less intrusive than video but data is less detailed awareness of being recorded not a problem useful to record long e...