• Modelling determinants of a cost accounting system: Mixed methodology and logistic regression

      Nagirikandalage, P; University of Chester
      Mixed methodology is becoming increasingly significant in several scientific research areas. Empirical management and cost accounting research attempt to integrate quantitative and qualitative methods and combine theories generally associated with incommensurable paradigms. Furthermore, mixed methods research could provide a more comprehensive understanding of cost accounting research by establishing a prevailing means of validation of research findings. However, this has also been criticised considerably in the social science aspects especially due to failings of presenting a vibrant philosophical foundation to produce valid knowledge statements and also in circumstances of a concept of triangulation is emerged as a mean of validation. As a methodological note on the analytical aspects, logistic regression model has been used in various studies of management and cost accounting research. However, there are criticisms over the presentations of the logistic model which has led to a misinterpretation of research findings. As per the usage of these methodologies in various contexts are concerned, scholars in management and cost accounting have argued that Sri Lanka seems to be more profound in methodology but the methodology should be determined by the research question and it is not given. Sri Lanka is perceived to be an empirical laboratory for management research as management practices in this country are different or distinctive. Hence, reporting on distinctiveness of practices will be very appealing to international audiences. The purpose of this paper is to illustrate how the mixed methodology has been adopted and how the logistic regression model was used to model the determinants for the demand for cost accounting systems in Sri Lanka as a developing country. A cost accounting system (CAS) has been used for decision support, financial planning and control as well. Empirical evidence has shown that different factors have influenced on demand for CAS but again has shown mixed results and there is a lack of evidence from the developing country or emerging economy context as well. Hence, this research study attempts at bridging the gap between the literatures by modelling the determinants for the demand for a CAS within an emerging economy such as Sri Lanka. Logistic regression model has identified that the market competition, size, desire and need of the management, quality of the report generation and changing dynamics as significant predictors for the demand for a CAS. Thematic analysis has been adopted to analyse the qualitative data gathered to achieve an in-depth understanding of CAS. This paper allows understanding how mixed methods research is conceptualised across these studies. The findings show a range of perceived strengths and weaknesses/ limitations identified and opportunities and risks attributed to this approach as well.