This paper uses binary logistic regression to develop two models of firms’ Occupational Health and Safety disclosures, one based on disclosure / non-disclosure, the other based on above / below the median levels of disclosure. Industry and auditor are found to be important components of both models, whilst operating revenue contributes to the former and company age to the latter. These findings support to some extent previous studies into Corporate Social and Environmental Disclosure, but also identify new factors that need to be further investigated. The paper contributes to our current understanding of Corporate Social Disclosure through its focus on the little considered area of Occupational Health and Safety and also raising the possibility of non-parametric statistics as a better statistical methodology for such research.


Published in Full


Further information about this conference or APIRA may be accessed here

Conference Papers may be accessed here

Included in

Business Commons