Human CSC 540
Popular in Course
Popular in ComputerScienence
This 7 page Class Notes was uploaded by Mr. Cleve MacGyver on Sunday October 25, 2015. The Class Notes belongs to CSC 540 at University of North Carolina at Greensboro taught by Staff in Fall. Since its upload, it has received 11 views. For similar materials see /class/229060/csc-540-university-of-north-carolina-at-greensboro in ComputerScienence at University of North Carolina at Greensboro.
Reviews for Human
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/25/15
CSC540 Evaluatinq Usab tv with Users Extra Notes on Informal Testinq Fall 2002 Part I Why test interface with users Only real users can uncover some of the problems Types of informal testing with users 0 Observation techniques 0 Record user performing tasks 0 Record quotThinkaloud protocolquot while user performs tasks 0 Query techniques 0 Questionnaires 0 Interviews Before starting any kind of informal test you need to 0 Choose users to test 0 Choose tasks to test 0 Prepare quotinterfacequot to be tested 0 Choose type of data to collect Choosing people to test 0 best to get representative users 0 follow ethics 0 avoid distressing test users no tears no ambulance avoid embarassing test users get informed consent free to stop anytime maintain privacy in data Institutional Review Board approval required at most institutions including UNCG Choosing tasks to test 0 Don39t test separate tasks 0 Suites of related tasks as in a scenario Prepare quotinterfacequot to be tested 0 Pen and paper drawings of screens 0 OK as long as not testing functionality not yet supported eg not suitable to test ease of use of GUI for drawing 0 Prototypes o More realistic but can be expensive to implement o Wizard of Oz 0 Have person behind the scenes performing the unimplemented parts of the system 0 Example faking speech recognition with a human typist Choose type of data to collect 0 Process data data about steps user takes and why 0 quotBottomlinequot data 0 a summary of what happened as in usability specification eg 0 how long did task take 0 how many errors did user make Process data sometimes more useful than bottomline data for interface designers o if users have problems designers need to know where in task and why Recording technologies 0 pen amp paper cheap but requires experience not to miss details 0 audio more intrusive more accurate effort to transcribe o video more intrusive than audio collect fine grained data very high cost of transcription 0 automatic recording of user actions keystrokes mouse actions etc by system o for formal usability studies hidden cameras eyetracking equipment Warning about quotHawthorne effectquot people may alter their normal behavior when they know they are being observed Record 0 user performing tasks 0 and optional user quotthinkaloud protocolquot Part II Thinkaloud protocol users talk to you while they work provides quotprocess dataquot not quotbottomlinequot data realworld example Inexperienced users confused because they thought quotparameterquot in documentation meant quotperimeterquot hard to collect both quotprocess dataquot and quotbottomline dataquot simultaneously Instructions to users for thinkaloud protocol recording ask them to tell you what they are thinking while they work 0 what they are doing or trying to do 0 what they are reading on the screen 0 what they think is happening 0 questions they have while they are working 0 no deep or private thoughts Tell them you will not answer their questions unless they absolutely cannot continue working on task encourage them to talk but be careful not to influence their work 0 quotTell me what you are thinkingquot quotKeep talkingquot 0 NOT quotWhat do you think that button is forquot lnclass exercise record thinkaloud protocol while searching for csc540 Reserve List on library web page Part III Questionnaires 0 Advantages can reach many users easily 0 Disadvantages 0 hard to design so that answers are meaningful 0 those who answer may not be representative users eg web surveys o user39s may not be able to explain real reasons or preferences 0 Example most people preferred the 3rcl choice of 3 identical pantyhose recency bias Types of questions 0 Openended 0 Closed provides choices for answer 0 Alternatives eg YesNoDon39t know 0 Multipoint scale eg 1 5 where 1 no use and 5 very useful 0 Ranked eg which is more important to you candy mint or breath mint Design of questionnaires 0 Short as possible 0 Easy to reply eg email 0 Guarantee privacy for sensitive data eg income 0 Simple amp consistent amp unambiguous wording 0 Perform pilot study of questions 0 Plan statistical analysis before collecting data so the right kind of questions are included See web page on questionnaires link from course web page
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'