There are no bibliometric formulas currently available to measure the intrinsic quality of scientific publications. Nonetheless, publication assessment is an inescapable feature of academic and professional evaluation although it is not always done fairly. This paper proposes a scoring system that combines several of the variables most often used for evaluation: article length, inclusion in biomedical databases, impact factor of the journals publishing the articles, and number of citations received during the 2 years following publication. Articles can be classified in 20 categories and assigned scores depending on how the factors are combined. The scoring system's advantage is that it limits excessive weight given to extreme impact factors and corrects differences due to varying citing behaviors in different Science Citation Index categories. Finally, scores are classified by type of article, number of co-authors, and arthorship order. When applying this system, it would be sufficient to evaluate candidates' 5 best articles in order to establish quantitative differences between them, reducing administrative costs and the workloads of assessment committees.