The Pairs ice-skating competition at the 2002 Winter Olympics in Salt Lake City generated considerable controversy. Allegations were made that the French judge was influenced by outsiders. Ultimately Gold Medals were awarded to both a Russian and a Canadian pair.
Here are the data set up as a 4-facet analysis in Olympics.txt. The original ratings are numbers like 5.4. These have been multiplied by 10 to produce the integer ratings that Facets expects. Not all intermediate categories in the functional range of 2.9 to 5.9 were observed, but the unobserved intermediate category 3.0 is maintained in the rating scale.
Use the Output Tables menu to look for interactions between judges and skater pairs.
For a two-facet Rasch analysis of these data, investigating rater behavior, see www.winsteps.com/winman/example15.htm
Facets specifications and data (in file Olympics.txt):
Title = "Pairs Skating: Winter Olympics, SLC 2002"
Facets = 4 ; Facets are Judges, Skating Pairs, Programs, Skills
Convergence= 12.0, .001 ; convergence when largest residual is less than 12.0
; and biggest logit change is less than .001
Positive = 1, 2, 3, 4 ; all Facets positive:
Models =
?,?,?,?, ISU ; keep unobserved intermediate categories with scores up to 6.0
*
Rating scale = ISU, R60, Keep ; 'Keep" maintains unobserved intermediate categories in the rating scale
59 = 5.9 ; all observed ratings multiplied by 10
*
Labels=
1, Judge ; leniency
1= Rus ;Mrs. Marina SANAIA : RUSSIA
....
*
2, Pair ; ability
1= BS-Rus ; 1 BEREZHNAYA Elena / SIKHARULIDZE Anton : RUS
2= SP-Can ; 5 SALE Jamie / PELLETIER David : CAN
....
*
3, Program ; easiness
1 = Short
2 = Free
*
4, Skill ; easiness
1 = Technical Merit
2 = Artistic Impression
*
Dvalues=
1, 1-9 ; Facet 1 is the 9 judges
*
Data =
1 1 1 58 58 57 58 58 58 58 58 57
1 1 2 58 58 58 58 59 58 58 58 58
....
