BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Project  Ernestine: Validating a GOMS analysis for predicting and 
 explaining real-world performance - Alan Blackwell and William Newman
DTSTART:20060302T110000Z
DTEND:20060302T123000Z
UID:TALK4817@talks.cam.ac.uk
CONTACT:Alan Blackwell
DESCRIPTION:Gray\, W. D.\, John\, B. E.\, & Atwood\, M. E. (1993). Project
 \nErnestine: Validating a GOMS analysis for predicting and\nexplaining rea
 l-world performance. Human-Computer Interaction\,\n8(3)\, 237-309. Project
  Ernestine: Validating a GOMS Analysis for\nPredicting and Explaining Real
 -world Performance\n\nAvailable online at:\nhttp://www.rpi.edu/~grayw/pubs
 /papers/GJA93_HCIj.html\n\nOriginal abstract:\n\nProject Ernestine served 
 a pragmatic as well as a scientific\ngoal: to compare the worktimes of tel
 ephone company toll and\nassistance operators on two different workstation
 s\, and to\nvalidate a GOMS analysis for predicting and explaining real-wo
 rld\nperformance. Contrary to expectations\, GOMS predicted and the\ndata 
 confirmed\, that performance with the proposed workstation\nwas slower tha
 n with the current one. Pragmatically\, this\nincrease in performance time
  translates into a cost of almost $2\nmillion dollars a year to NYNEX. Sci
 entifically\, the GOMS models\npredicted performance with exceptional accu
 racy.\n\nThe empirical data provided us with three interesting results:\np
 roof that the new workstation was slower than the old\, evidence\nthat thi
 s difference was not constant but varied with call\ncategory\, and (in a t
 rial that spanned four months and collected\ndata on 72\,450 phone calls) 
 proof that performance on the new\nworkstation stabilized after the first 
 month. The GOMS models\npredicted the first two results and explained all 
 three.\n\nIn this paper\, we discuss the process and results of model\nbui
 lding as well as the design and outcome of the field trial. We\nassess the
  accuracy of GOMS predictions and use the mechanisms of\nthe models to exp
 lain the empirical results. Lastly\, we\ndemonstrate how the GOMS models c
 an be used to guide the design\nof a new workstation and evaluate design d
 ecisions before they\nare implemented.
LOCATION:Primrose Room\, Microsoft Research Cambridge
END:VEVENT
END:VCALENDAR
