About the RSReduction
It presents capabilities of main functions of
startAuc(attribute, D)– compute the AUC values of every single item in the rating scale.
totalAuc(attribute, D, plotT = FALSE)– sort AUC values in the ascending order and compute AUCs of running total of first k items, k = 1, …, n, where n is the number of attributes. Setting the argument plotT as TRUE the plot of new AUC values is created. The horizontal line marks the max new AUC.
rsr(attribute, D, plotRSR = FALSE)– the main function of the package reducing the rating scale according the procedure illustrated by Fig. 1.
Setting the argument plotRSR as TRUE the plot of ROC curve of the sum of attributes in reduced rating scale is created.
All functions work on the
data.frame containing columns of items (attributes) and
one decision column with two categories, e.g. (0,1). The rows in
represents examples in the sample.
About the data set used in RSReduction
The data set comes from the project Social Diagnosis - Objective and Subjective Quality of Life in Poland (2015). It contains 20 questions about the happiness in life. The decision vector is the answer happy (1) or unhappy (0).
- W. Koczkodaj, F. Li, A. Wolny–Dominiak. The R Journal (2018) 10:1, pages 43-55.
- W. Koczkodaj, T. Kakiashvili, A. Szymanska, J. Montero-Marin, R. Araya, , J. Garcia-Campayo, K. Rutkowski, and D. Strzałka. How to reduce the number of rating scale items without predictability loss? Scientometrics, 2, 2017.
- W. Koczkodaj, A. Wolny-Dominiak. RatingScaleReduction package: stepwise rating scale item reduction without predictability loss, arXiv, 2017.