SkillSwap Weekend - Usability Testing
- 7. Why make something usable? Necessary for survival
- 8. The customer is always right You can’t blame the user
- 13. How is usability defined? Learnability Efficiency Memorability Errors Satisfaction from Jakob Nielsen’s “Usability Engineering”
- 15. Your best guess isn’t good enough
- 19. Your best guess isn’t good enough
- 20. We don’t understand ourselves as well as we think we do
- 21. Smart phone habits What percentage of the time do you have your smart phone on you?
- 22. Smart phone habits What percentage of the time do you have your smart phone on you? ~60% Dey et al. 2011
- 23. You are not the user
- 27. Why not just ask? People lie.
- 28. What is usability testing? Method of evaluating the ease-of-use of a system through direct observation.
- 29. What is usability testing not?
- 30. ✔ ✘ User Usability Expert Behavior Opinion Direct Observation Indirect Structured Ethnographic via McGill University
- 31. Not focus groups. Not market research.
- 32. When to test usability? Early & often.
- 33. Low fidelity High fidelity
- 34. Wizard of Oz Technique
- 35. Practical approaches to usability testing
- 36. Running a usability test 1) Choose tasks 2) Choose users 3) Perform the test
- 37. Running a test: Choosing tasks Very specific Representative of target uses Most important parts of UI Not too short, not too long Start simple, end simple
- 38. 1) “Enter sales figures for six regions for each of four quarters, with these numbers.” 2) “Calculate totals and percentages of the data.”
- 39. Running a test: Choosing users Representative of target users Must be novices Optional: training Don’t repeat on different designs
- 40. Running a test: Perform the test “We’re testing the software. Not you.” One experimenter Refrain from interaction with user * No personal opinions You are a peer. Not an expert. Optional: Video recording Optional: Think-Aloud Technique
- 41. “Thinking aloud may be the single most valuable usability engineering method.”
- 42. Think-Aloud Technique Reveals how users view and interpret the UI Users continuously think out loud Wealth of qualitative data Unnatural/difficult for users Avoid rationalizations of thoughts
- 43. Good: “What are you thinking now?” from Jakob Nielsen’s “Usability Engineering”
- 44. Bad: “What do you think the message on the bottom of the screen means?” from Jakob Nielsen’s “Usability Engineering”
- 45. Demo of Think-Aloud Technique
- 46. Running a usability test 1) Choose tasks 2) Choose users 3) Perform the test
- 47. Kevin Schaefer [email protected] Thank you. u Gabi Marcu
Editor's Notes
- #2: Introduce myself + Gabi\n
- #3: Who has gone to the activities fair and signed up for a DLIST?\n\nFor the freshman, you go to activities fair and signup for 10 groups... then CMU sets in\n\nHavent been in this situation? You will be.\n
- #4: Who has gone to the activities fair and signed up for a DLIST?\n\nFor the freshman, you go to activities fair and signup for 10 groups... then CMU sets in\n\nHavent been in this situation? You will be.\n
- #5: Who has gone to the activities fair and signed up for a DLIST?\n\nFor the freshman, you go to activities fair and signup for 10 groups... then CMU sets in\n\nHavent been in this situation? You will be.\n
- #6: Who has gone to the activities fair and signed up for a DLIST?\n\nFor the freshman, you go to activities fair and signup for 10 groups... then CMU sets in\n\nHavent been in this situation? You will be.\n
- #7: Who has gone to the activities fair and signed up for a DLIST?\n\nFor the freshman, you go to activities fair and signup for 10 groups... then CMU sets in\n\nHavent been in this situation? You will be.\n
- #8: If people get lost, they’ll get frustrated.\nIf your website is hard to understand, they’ll leave.\nIf your app is confusing, they’ll delete it and give it 1 star.\n
- #9: \n
- #10: http://research.microsoft.com/en-us/um/cambridge/projects/sensecam/downloads/CambridgeBig.wmv\n
- #11: \n
- #12: \n
- #13: \n
- #14: Learnability: How easy is the system to learn?\nEfficiency: Expert users can reach high levels of productivity\nMemorability: Users can return to system without relearning\nErrors: Low error rate & easy recovery\nSatisfaction: Users should feel satisfied after using it\n
- #15: But... the talk is called usability *testing*\nWhy do we have to strategically test usability?\nWhy can’t we just build *usable* interfaces?\n(learnability, efficiency, memorability, error reduction, satisfaction)\n
- #16: It is impossible to design an optimal UI by giving it your best shot\nInfinite potential for making unexpected misinterpretations of UI\n^^ this is expected. Design with this expectation.\n\nEthernet example\n
- #17: CMU ethernet example\n\n
- #18: This is easy to miss. You have to test your designs.\n\n
- #19: This is easy to miss. You have to test your design.\nThis is obvious, but there are 5 things you wont see for every one you do\n
- #20: Design with this in mind. Expect to be surprised. Expect to iterate.\n
- #21: Another reason we have to test usability\n
- #22: \n
- #23: 50% on you (arms length)\n80% in the room\n20% off\n
- #24: HCI Cliche\nA designer can look at any screen and believe it makes perfect sense\n
- #25: I promise the designer of CMU’s academic audit had no trouble figuring out what classes they need to take to graduate. For the rest of us, not so much.\n\n
- #26: Another example of why designers cant evaluate their own UIs.\nLearning a UI is a one way street\nFedEX logo\nWhat has been seen cannot be unseen\n
- #27: One more for fun. \n\nPoint is: Once you learn how to interact with a UI, you can’t “un learn it”\n
- #28: Alright, alright. Usability is important. We need to study it. So why not just ask people?\n\nPeople are incapable of describing their actions\nDescribe yourself at grocery store\n
- #29: \n
- #30: \n
- #31: \n
- #32: \n
- #33: \n
- #34: \n
- #35: http://www.youtube.com/watch?v=NZR64EF3OpA\n\nyou don’t need to put the work into building the functionality until you are certain that it’s even needed\n
- #36: \n
- #37: \n
- #38: Last task should be easy to leave users satisfied.\n
- #39: Last task should be easy to leave users satisfied.\n
- #40: Training: Give Kinect example\n
- #41: Contamination/crime scene comparison?\n*** unless clearly stuck and frustrated\n
- #42: \n
- #43: *** unless clearly stuck and frustrated\n\nInformal comment: I dont like this\n
- #44: \n
- #45: \n
- #46: \n
- #47: Learnability: How easy is the system to learn?\nEfficiency: Expert users can reach high levels of productivity\nMemorability: Users can return to system without relearning\nErrors: Low error rate & easy recovery\nSatisfaction: Users should feel satisfied after using it\n
- #48: \n