Interviews are the most fundamental tool for research, providing direct contact with participants and collecting their attitudes, beliefs, perceptions and experiences at first hand. Interviews are always best conducted in person, ideally at a location relevant to the behaviour to be understood, in order to understand the contextual cues that drive behaviours as well as personal expressions and body language that communicate rich additional information beyond what is actually said.Interviews generally have some structure, but ideally should be allowed to flow in a conversational way rather than following the more formal structure of a questionnaire. This can be as loose as a guide to the topics and issues to be covered, and less structure can often reveal much more than more structured interviews that can be perceived as formal and impersonal. One tip is to start with more general topics and then focus on more specific questions later in the interview (including any sensitive questions which should only be asked when rapport has been built).
Interviewers need to be sensitive and flexible to allow participants to share, and also aware of how questions are phrased in order to leave options open without leading the participant. TapestryWorks often use visual stimulus to capture participant’s implicit feelings before delving into the motivations and emotions associated with behaviours.
Interviews are most often conducted one-on-one, but for some topics it may be appropriate to interviews pairs of people (e.g., couples for behaviours and decisions that relate to households or friends with shared interests where this can help them to express themselves by sharing opinions).
Interviews are the most fundamental tool for researchers to provide feedback on design, and should always be the start point of any investigation, even if for informal understanding of a topic or domain. Interviews humanise research in a way that surveys cannot. Although interviews can be conducted online, which can be particularly useful for exploring topics in depth and across many days, online approaches can never completely replace looking participants in the eye and seeing the full context of their behaviours and lives.
Universal Methods of Design by Martin & Hanington