quiz: Multidimensionnal rank data : quiz

In Rankcluster: Model-Based Clustering for Multivariate Partial Ranking Data

Description

This dataset contains the answers of 70 students (40 of third year and 30 of fourth year) from Polytech'Lille (statistics engineering school, France) to the four following quizzes:

Literature Quiz

This quiz consists of ranking four french writers according to chronological order: A=Victor Hugo, B=Moliere, C=Albert Camus, D=Jean-Jacques Rousseau.

Football Quiz

This quiz consists of ranking four national football teams according to increasing number of wins in the football World Cup: A=France, B=Germany, C=Brazil, D=Italy.

Mathematics Quiz

This quiz consists of ranking four numbers according to increasing order: A=pi/3, B=log(1), C=exp(2), D=(1+sqrt(5))/2.

Cinema Quiz

This quiz consists of ranking four Tarentino's movies according to chronological order: A=Inglourious Basterds, B=Pulp Fiction, C=Reservoir Dogs, D=Jackie Brown.

Format

A list containing:

data

a matrix of size 70*16. The student's answers are in row and the 16 columns correspond to the 4 rankings (for the 4 quizzes) of size 4 (ranking representation).

The ranking representation r=(r_1,...,r_m) contains the ranks assigned to the objects, and means that the ith object is in r_ith position.

For example, if the ranking representation of a rank is (4,3,1,2,5), it means that judge ranks the first object in 4th position, second object in 3rd position, ...

frequency

a matrix of size 63*17. Each row corresponds to one of the 63 differents observed rankings (ranking representation). Each row contains 4 ranks of size 4 and a last column for the frequency.

m

a vector with the sizes of the ranks for each dimension.

Source

Julien Jacques

Examples

data(quiz)
 

Example output

WARNING : Since Rancluster 0.92, the ranks have to be given to the package in the ranking notation (see convertRank function), with the following convention :
- missing positions are replaced by 0
- tied are replaced by the lowest position they share

Rankcluster documentation built on Aug. 26, 2019, 3 p.m.