Field Studies Ethnographic Studies Participant Observation Alfred Kobsa

Field Studies (Ethnographic Studies, Participant Observation) Alfred Kobsa University of California, Irvine

Characteristics • At offices, homes, construction sites, airports, … • Between a few hours and a couple of weeks • Primarily an observational technique, but can be combined with interviews, questionnaires, think aloud, etc. • Important: - Buy-in from stakeholders - good sampling plan (whom, where, when) - Good recording of observations (notes, pictures, videos, audio), and quick write-up of report - Develop “detached partnership” with subjects

Inquiry-based vs. observational studies • Observational empirical studies Reveal actual user behavior (which may differ from users’ stated behavior) Subjects may act differently than normal (intentional, Hawthorne effect…) Researcher may find it difficult to interpret observed behavior. Researcher is subject to simplification bias. • Inquiry-based studies Reveal aspects of users’ rationale that cannot be inferred from mere observation Respondents have difficulties recalling and formulating what they do/think, and may be biased through social desirability, conformity, acquiescence, prestige. Also simplification/translation problems. ☛ Both methods complement each other (they yield overlapping sets of insights and may reduce misunderstandings).

Different forms (Baxter et al. ) Observation only • Pure observation (☛ be “unnoticeable” to others) • Deep hanging-out (focused observation, collection of artifacts, becoming a user) Observation and Interaction • • Contextual inquiry (☛ strive for apprentice role; avoid sage, interviewer or guest role) Process analysis (☛ see task analysis) Condensed ethnographic interview (top-down, using results of inquiry) [Discount user observation (interviewer and note/picture taker)] Special techniques • • Artifact walkthrough (e. g. , documents, notes) Incident diaries (☛ useful for infrequent events, e. g. errors encountered) Experience sampling (randomly or strategically) Observer-less recording (video cameras, screen recordings, maybe sensors)