[Purposes] This paper aims to explore the identification and prevention methods on data falsification, so as to cause the editors of scientific journals to pay attention to insidious academic misconduct. [Methods] According to the classification and characteristics of data falsification, we took the typical cases in scientific articles as examples, and put forward the countermeasures. [Findings] Three common types of data falsification including fabricated data, false report on sample size, and tampering data are settled, and identification methods are provided in terms of author's background analyses, research methods description, modified contents comparison, and logical combing. It is recommended that editorial discriminating ability, authors' sense of scientific integrity, and punishment mechanism should be improved from the perspective of scientific journals to prevent data falsification in scientific articles. [Conclusions] This study provides a reference for further editorial work on identifying and preventing academic misconduct of data falsification in scientific articles.