CMSC434 Intro to Human-Computer Interaction. Data Gathering and User Research Tuesday, February 12, 2013 Instructor: Jon Froehlich TA: Matt Mauriello

Transcription

1 CMSC434 Intro to Human-Computer Interaction Data Gathering and User Research Tuesday, February 12, 2013 Instructor: Jon Froehlich TA: Matt Mauriello

2 Hall of Fame Hall of Shame

3 Printing Scott Berkun s Reading, submitted by Floyd Hawkes

4 Printing Scott Berkun s Reading, submitted by Floyd Hawkes

5 Dialog Box for Creating Mail Filter Task: Create mail filter that automatically classifies and sorts mail for CMSC434. "Add Condition" is closer to the list of actions when the window is large than it is to the conditions; "Add Action" is far away from the list of actions The two actions are at a high level of cognitive abstraction ("add action") rather than a lower level (a plus sign grouped with the list) The two actions have similar "command sequences"/names Creating Mail Filter, submitted by Ted Smith

6 THURSDAY Team Brainstorms

7 TeamAssignmentDeadline Team Project Proposal Due: Friday, February 22 nd, 11:59:59PM

8 ProposalGrading Worth 5% of your overall grade in the class You will be graded on how well you execute the aforementioned seven sections You will be additionally graded on creativity and writing quality I will also factor in peer review to help assess the quality of the work and the output from each team member

9 ProposalSubmission Do not this submission! Instead, post it directly on CMSC434 Wiki Each team must make a wiki for their project This wiki can be very simple; later in the semester you will make a project website.

10

11 ProjectNotebook

12 [Buxton, Sketching User Experiences]

13 [Buxton, Sketching User Experiences]

14 [Buxton, Sketching User Experiences]

15 [Greenberg, Sketching User Experiences: The Workbook]

16 DigitalNotebooks Microsoft OneNote Evernote

17 So, where are we? We ve defined high level problems but o What specific problems exist to be solved here? o How do we know that these problems exist? o How do we know when we ve solved the problem?

18 TodaysMainPoints Do not just trust your instincts to make design decisions Observe target users in context to inform your design

19 Instincts! The intellect has little to do on the road to discovery. There comes a leap in consciousness, call it intuition or what you will and the solution just comes to you and you don t know where and why. AlbertEinstein Patent Clerk / Nobel Prize Winner

20 Give you the tools to combine o observations and other data o inspirations o your intellect o your intuition

21 No one sets out to make a bad design

22 Xerox1983 Existing copiers judged as too complicated by customers, but why? Lucy Suchman, UC Berkeley PhD in Anthropology, suggests videotaping interactions

23 [Suchman, CHI 1983] Task: Make two-sided copies of 100 page original

24 AverageUsers? AllenNewell ACM Turing Award Winner RonKaplan ACM Fellow Observation showed that difficulties were not due to lack of sophistication of users, but due to problems reading (making sense of) an unfamiliar artifact

25 XeroxCommercial [Suchman, CHI 1983]

26 How the customer explained it How it was understood How it was designed How it was coded How it was documented How it was marketed What the customer really needed: [

27 Getting the design right and the right design [Buxton, Sketching User Experiences] BillBuxton Designer / Researcher / Writer / Thinker

28 UserResearch finding out about people, problems, and needs We ve defined high level problems but o What specific problems exist to be solved here? o How do we know that these problems exist? o How do we know when we ve solved the problem?

29 UserResearchMethods Formative Build Evaluative

30 UserResearchMethods Formative Build Evaluative Ethnography Interviews Surveys Cultural Probes Focus Groups Diary Studies Experience Sampling Studies Studying Similar Products Interaction Logging of Past Product / Early Prototype Studying Past Product Documentation Ethnography Interviews Surveys Focus Groups Diary Studies Experience Sampling Studies Interaction Logging

31 Statistical Macro techniques (many people) Surveys Video Ethnography Saying Explicit opportunities and needs Latent opportunities and needs Doing Focus Groups Observational Techniques [Designing Interactions by Bill Moggridge, 2007] Micro techniques (few people) Interpretive

32 KeyPoint: Triangulation All user research techniques have their own limitations Use multiple techniques to fully understand a design scenario o Choose techniques that account for the weaknesses of each other o Choose techniques to cover both depth and breadth of the user experience

33 Why can t we just ask users what they want?

34 Users Words are Unreliable People are notoriously bad at predicting what they would use or would prefer when it is only hypothetical They can much better respond to actual, concrete things, or make comparisons This highlights the importance of observation and of prototypes!

35 Plus, the risk is that you might be simply refining an unoptimal solution Windows CE Pocket PC 2000 April 2000 Pocket PC 2002 Oct 2001 Win Mobile 2003 June 2003 Win Mobile 5 May 2005 Win Mobile 6 Feb 2007

36 Plus x2

37 It s not the consumers job to know what they want. SteveJobs Designer / Inventor / Creative Genius

38 If I had asked my customers what they wanted, they would have said a faster horse. HenryFord Inventor / Car Salesman

39 So, often it s better to watch what people do than to only design solely for what they say

40 You can observe a lot by just watching YogiBerra MLB Player / Quote Machine