Dennis Fok, Richard Paap, and Bram van Dijk, "A Rank-Ordered Logit Model with Unobserved Heterogeneity in Ranking Capabilities", Journal of Applied Econometrics, Vol. 27, No. 5, 2012, pp. 831-846. The first dataset we use in the paper is about ranking gaming platforms. This was based on a short survey that we conducted ourselves among 91 students. All results can be found in the file game_data.csv, an ASCII file in DOS format that is zipped in fpd-data.zip. Unix/Linux users should use "unzip -a". The first line gives the variable numbers, and the following 91 lines give the answers of the respondents. Variables: [v1: ranking of the six gaming platforms] v1_1: rank number given to Xbox (1 to 6, 1 being most preferred, 6 being least preferred) v1_2: rank number given to Playstation (1 to 6, 1 being most preferred, 6 being least preferred) v1_3: rank number given to Playstation portable (1 to 6, 1 being most preferred, 6 being least preferred) v1_4: rank number given to GameCube (1 to 6, 1 being most preferred, 6 being least preferred) v1_5: rank number given to GameBoy (1 to 6, 1 being most preferred, 6 being least preferred) v1_6: rank number given to PC (1 to 6, 1 being most preferred, 6 being least preferred) [v2: which of the six platforms do the students own?] v2_1: owns an Xbox (1=yes, 0=no) v2_2: owns a Playstation (1=yes, 0=no) v2_3: owns a Playstation portable (1=yes, 0=no) v2_4: owns a GameCube (1=yes, 0=no) v2_5: owns a GameBoy (1=yes, 0=no) v2_6: owns a PC (1=yes, 0=no) [v3: how important are the following items for a gaming platform (1 to 5, 1 being not important at all, 5 being very important)] v3_1: importance of price of hardware (1 to 5, 1 being not important at all, 5 being very important) v3_2: importance of price of software (1 to 5, 1 being not important at all, 5 being very important) v3_3: importance of exchangeability of games (1 to 5, 1 being not important at all, 5 being very important) v3_4: importance of availability of games (1 to 5, 1 being not important at all, 5 being very important) v3_5: importance of graphical capabilities (1 to 5, 1 being not important at all, 5 being very important) v3_6: importance of portability (1 to 5, 1 being not important at all, 5 being very important) v3_7: importance of multiplayer capabilities (1 to 5, 1 being not important at all, 5 being very important) v3_8: importance of other capabilities (e.g. playing DVD's) (1 to 5, 1 being not important at all, 5 being very important) [v4: which genre of games do you like] v4_1: shooting/action (1=yes, 0=no) v4_2: strategy (1=yes, 0=no) v4_3: puzzle (1=yes, 0=no) v4_4: sports/race (1=yes, 0=no) v5_5: RPG (1=yes, 0=no) v5: Gender (1=female 0=male) v6: Age in years v7: Hours per week spent on gaming The second data set deals with political preferences. We are not allowed to distribute this data set. However, the data are available for academic research and education through DANS (Data Archiving and Networked Services), see http://easy.dans.knaw.nl. The data set is called "Culturele veranderingen in Nederland 2004 - CV'04" and can be found via http://persistent-identifier.nl/?identifier=urn:nbn:nl:ui:13-u38-wn0 . The variables we used are var1046a to var1046p (ranks of political goals), var001 (gender) and var002 (age). Respondents who did not give a (complete) ranking were discarded.