Characterizing occupational accidents patterns using Multiple Correspondence Analysis (MCA)

Celeste Jacinto, Helena Carvalho, Sílvia Silva

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Scope & Objective: this research maps accident profiles through Multiple Correspondence Analysis (MCA), using a large national database of occupational accidents from the Portuguese Office for Strategy and Planning (GEP), over a period of 3 years. Data analysis covered a number of relevant variables, namely: “Specific Physical Activity”; “Deviation”; “Contact”; “Part of the Body Injured” and “Type of Injury”. MCA allowed the authors to describe the associations and provided a graphical display of the multidimensionality of the space, representing all the categories of the variables in a sub-space with the minimum number of dimensions possible. Results: results revealed four main profiles, namely, (1) accidents related to “man-machine” interaction (i.e., work with machines or tools); (2) accidents related to “bad movements and overexertion of force”; (3) accidents related to “trips and falls” and, (4) the forth profile can be interpreted as “undifferentiated” accidents. These patterns where further explored in order to identify specificities associated with (1) organizational variables employer/company (i.e., “economic activity”, “size of enterprise”) and (2) data about the employee who suffered the accident (i.e., “age”, “sex”). The main findings support the usefulness of MCA in accident analysis.
Original languageEnglish
Title of host publicationThe 8th International Conference on Working on Safety (WOS 2015)
Place of PublicationPorto, Portugal
Pages101-102
Number of pages2
VolumeBook of Abstracts
Publication statusPublished - 2015

Keywords

  • Multiple Correspondence Analysis
  • Accident patterns
  • Occupational safety

Fingerprint Dive into the research topics of 'Characterizing occupational accidents patterns using Multiple Correspondence Analysis (MCA)'. Together they form a unique fingerprint.

Cite this