User research - different approaches and methods
This is the second part of my fourth lecture at HITLab, Canterbury University, Christchurch, New Zealand about user research. I am presenting the three levels of understanding user needs and the methods that correspond with investigating these needs. The idea is to show how different methods enable a designer to dig for different insights and how to conduct exemplary studies for each type of the method.
user research: dierent approaches and methods aga szstek(at)gmail.com what people level of knowledge research methods Fraukje Sleesvijk Visser what people level of knowledge tell think explicit interviews research methods Fraukje Sleesvijk Visser what people level of knowledge tell think do act explicit observational interviews observations research methods Fraukje Sleesvijk Visser what people level of knowledge tell think do act know feel dream explicit observational subconscious latent deep interviews observations generative techniques research methods Fraukje Sleesvijk Visser before discussing the dierent methods it is crucial to know about the concurrent and retrospective experiences - when people report on their current feelings the feelings themselves are accessible to introspection allowing for accurate reports - aective experiences are eeting and become not available once the feeling is gone - once the feeling disappears the aective experiences need to be reconstructed - when people report on their current feelings the feelings themselves are accessible to introspection allowing for accurate reports - aective experiences are eeting and become not available once the feeling is gone - once the feeling disappears the aective experiences need to be reconstructed - reports of past feelings are based on semantic knowledge - when asked how theyusuallyfeel during a particular activity, people draw on their general beliefs about that activity - the actual experience does not gure prominently in these reports because the experience itself is no longer accessible - the same knowledge is a basis for predicting future feelings, for which episodic information is not available to begin with so, lets discuss the methods what people level of knowledge tell think explicit interviews research methods introspective versus retrospective methods introspective methods Analysis of present behaviour and gathering of preferences by combining observations and interviews in context: - cognitive walkthorugh: heuristic identication of problems connected to the current ways of behaviour and interaction - think aloud: heuristic identication of problems arising from the current state of interaction through verbal articulation of user observations while interacting with the solution - co-discovery: a method where two users collaborate in order to resolve the problems found in the process of interaction with the current solution - Contextual Inquiry: gathering of detailed information about behaviour and motivations of users while the user is actually performing a given task retrospective methods Support for retrospective assessment of user preferences towards a given solution: - interviews: a discussion with the user aiming to collect rich material about his / her interactions and attitudes - focus groups: group discussion about a given problem or solution - retrospective analysis of behaviour: and interview supported by a qualitative recoding of user behaviour (e.g., video or audio recordings) when to apply? - exploration phase - to get a better understanding - to nd inspiration for further research - requirements validation - to check whether study assumptions are correct - evaluation - to discuss and verify design decisions types of studies - fully structured: the entire study takes place according to the earlier prepared scenario - good for comparison across individuals - more extensive answers comparing to written surveys - inability ask unplanned questions - fully structured: the entire study takes place according to the earlier prepared scenario - good for comparison across individuals - more extensive answers comparing to written surveys - inability ask unplanned questions - semi-structured: there are certain guidelines to the elements of the study but the order is unimportant - pre-specied questions - questions as a starting point for discussion - digression from the script is acceptable - fully structured: the entire study takes place according to the earlier prepared scenario - good for comparison across individuals - more extensive answers comparing to written surveys - inability ask unplanned questions - semi-structured: there are certain guidelines to the elements of the study but the order is unimportant - pre-specied questions - questions as a starting point for discussion - digression from the script is acceptable - unstructured: going with the ow, only the initial question and a possible list of topics is dened - good for open ended exploration - dicult to conduct and analyse study questions - close-ended - simple questions get simple answers - On a scale of 1-10, 10 being best, how did you like the prototype? - Did you like the prototype? - easy to analyse, but may not be that informative - open-ended - invite elaboration and discussion - build conversation and stories - What did you think about the prototype? - avoid - long questions, jargon, leading questions, unconscious biases, negative questions - ask why? - even when you think you know the answer, ask people why they do or say things - encourage stories - whether or not the stories people tell are true, they reveal how they think about the world - look for inconsistencies - sometimes what people say and what they do are dierent. These inconsistencies often hide interesting insights - observe nonverbal cues - be aware of body language and emotions. - do not be afraid of silence - if you allow for silence, a person can reect on what theyve just said and may reveal something deeper preparing the study - brainstorm questions - identify and order themes - pilot the study - write an interview guide - prepare for capturing the interview - prepare logistic backup conducting the study - you are the host - be friendly, respectful and non-judgmental - do not bias through body language - be exible (unless fully structured) but keep things on track - ask for clarications if something is unclear - listen (dont talk too much) and read between the lines - be consistent - outline - briey introduce testing goals - complete paperwork (informed consent) - simple questions rst, hard questions later capturing the study - have someone to help taking notes or do the audio or video recording - capture insights, non-verbal responses, etc. - balance between detail and conciseness - summarize immediately after the interview - do transcriptions - respect privacy and anonymity debrieng - ask for any nal comments - provide more detail about research goals - oer a brief summary of ndings - turn o recording devices - saythanks! - reect and summarize notes immediately what people level of knowledge tell think do act explicit observational interviews observations research methods - roots in ethnographic research - observing people in action - recording what is being observed - helping the researcher learn perspectives held by participants - often used in conjunction with other methods (e.g. interviews, focus groups, content analysis) - observations: observing and recording of user behaviour without a direct contact between the user and the researcher - shadowing: direct observation of user behaviour by the researcher without intervening with his / her actions - interaction logging: logging previously dened behaviours and system states that pertain to the specied user behaviour - diaries: systematic reporting by the user his / her interactions in a form of a diary or a blog - experience sampling: a systematic way of having participants provide samples of their ongoing behavior when participants' reports are dependent on either a signal, pre-established intervals or the occurrence of some event - day reconstruction method: capturing daily experiences of the users at the end of each day where participants are asked to listed all activities of the day that somehow related to the product or a situation that is under investigation data collection - eld notes - video - audio - transcripts - behaviour matrix - situational drawings pros and cons - exible and open-ended - directly measures behavior - gives the researcher insider view - allows for morphing of the study - hard to be an insider - invasive and intrusive - dicult to record - subjective - Hawthorne eect - time consuming - not generalizable - does not measure cognitive or aective aspects phases of observational research - phase 1: develop positive relationships with participants, gatekeeper, etc. - phase 2: improve design, refocus, redene questions after initial observations (may be ongoing) - phase 3: select additional participants as necessary - phase 4: follow up and probe deeper ethical considerations - how much do you disclose? - make your intentions clear - get an informed consent - develop a code of practice - maintain condentiality (Mack, et al., 2005) being an eective observer - know what you are researching - check in with your lens, biases, experiences and expectations -- have a blank mind - go where people are engaged in their daily lives and nd the right informants - understand the culture - rehearse how you will explain the purpose of your study - decide how you will do the observation and collect the notes - document what you observe without expectations - summarize and expand on eld notes as soon as possible - be open to the research problem reformulation what people level of knowledge tell think do act know feel dream explicit observational subconscious latent deep interviews observations generative techniques research methods using design as a research tool cultural probes - tools for better understanding of peoples latent needs and dreams - small craft objects used in empathic engagements with individuals around issues centered on self- identity and personal signicance - their form should reect the research question and its context - asking questions by applying a delicate, provoking and creative means - and oering intriguing ways to consider the asked questions and to look for surprising answers understanding what people do with their invoices AgaSzstek,MarcinPiotrowski, what is agood and a bad invoice? what happens to an invoice once it arrives? what is the best way to pay an invoice? how should the email accompanying an invoice look like? which elements should actually stay on an invoice? what should happen if you forgot to pay? technology probes - simple, exible, adaptable interactive installations - combining collecting data about the use of the technology in a real-world setting, testing the technology and inspiring users (and designers) to think of new kinds of technology - installed in user environment with the goal to observe them over a period of time - helping to identify design partners determining which kinds of technologies would be interesting to pursue - open-ended, inspiring new activities and encouraging experimentation managing interruptions | project supervised by Aga Szstek dealing with stress at work design probes - collaboration with a singular person based on a very deep engagement in the topic that is personally signicant - designing objects, which reect individual experiences and their meaning for the user - objects that support reection and auto-reection - such objects often become a part of an intimate relationship between the designer and the user Top Trumps probe cards to mark objects which are perceived to have magical power helping to keep a straight posture | Magda Rydygier project supervised by Aga Szstek preparing the study - brainstorm the context to the given problem (e.g. through contextmapping exercises) - brainstorm on user expectations and the ideal experience - prepare a list of questions regarding the current experience - decide on the format of the cultural probe (digital elements, physical element and intangible elements) - select the criteria for choosing participants and prepare awards - develop the probes and the instruction - organize a session with participants to explain to them the elements of the probe and the study procedure - provide informed consent forms - personally give the probes to the participants running the study - the study typically lasts no longer that one week (but this is very individual to the research problem) - usually the study happens without an active involvement of the researcher - if the study involves an online diary check the entries every day and comment on them - give participants a phone number and an email address where they can contact you in the case of any problems - at the end of the study invite all participates either to a joint session (a Focus Group or a creative workshop) or for individual interviews - record the sessions for later reference summary - the research method should reect the information you are trying to obtain - dont be afraid to combine the methods and experiment with them - think of creating visual materials for your studies they engage the users bringing you more insight - user research is a moment when you approach users with open mind try not to bias them and put your ideas into their mouth - the quality of the data you collect will likely be directly reected in the ideas you will create later on references Gaver, William W., et al. "Cultural probes and the value of uncertainty." interactions 11.5 (2004): 53-56. Gaver, Bill, Tony Dunne, and Elena Pacenti. "Design: cultural probes." interactions 6.1 (1999): 21-29. Mattelmki, Tuuli. Design probes. Aalto University, 2006. Hutchinson, Hilary, et al. "Technology probes: inspiring design for and with families." Proceedings of the SIGCHI conference on Human factors in computing systems. ACM, 2003. Buijs, Jan. The Delft innovation method: A design thinker's guide to innovation. Eleven International Publishing, 2012.