The application of citation count regression to identify important papers in the literature on non-audit fees

Purpose - This paper aims to show that when conducting a literature review, important papers can be identified by regressing citation counts on prior publications’ metadata. Design/methodology/approach - The method developed in this paper applies citation count regression analysis to identify important papers that may be overlooked when conducting literature reviews on subject areas with a large population of studies. Findings - The developed method reduces a literature down to a small sample of important papers for further narrative analysis. Research limitations/implications - Although the most widely used citation count database was used for research, there is a risk that a paper is not indexed; thus, it would be out of the scope of the literature. Practical implications - The developed method allows both preliminary selection of important papers for literature review, and robustness and completeness checks for already conducted narrative reviews. Originality/value - This paper develops an automated search method for identification of important papers based on citation counts. This method allows for the reduction of big samples of research papers into smaller heterogenic subsamples. Like meta-analysis, this method is a quantitative technique that can enhance traditional narrative literature reviews.