Open AccessDo transformation methods matter?
Author(s) -
Lenka Hudrlíková,
Jana Kramulová
Publication year - 2013
Publication title -
metodološki zvezki
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.127
H-Index - 7
eISSN - 1854-0031
pISSN - 1854-0023
DOI - 10.51936/sqxo6989
Subject(s) - multitude , sustainability , consistency (knowledge bases) , transformation (genetics) , preference , quality (philosophy) , czech , process (computing) , perspective (graphical) , phenomenon , computer science , risk analysis (engineering) , econometrics , management science , mathematics , economics , statistics , business , ecology , political science , artificial intelligence , law , chemistry , biology , operating system , biochemistry , epistemology , quantum mechanics , physics , gene , philosophy , linguistics
The general aim of a multitude of research projects is to assess a social, economic or environmental process or phenomenon by various indicators that are often measured in different units. In such situations, the data transformation and/or normalisation are inevitable. The present paper focuses on benefits and drawbacks of different normalisation methods. Further, it compares the results produced by several methods from the consistency and quality of the measurement perspective. The case of Czech NUTS 3 regions sustainability indicators is introduced. The authors employ 40 indicators divided into three sustainability pillars, attempting to conclude which method is the most suitable for further statistical analysis under the preference of dimensionless numbers.