Abstract: [Purpose/Significance] The Open Access model has made great contributions to scientific communication and academic development, and the Open Access Citation Advantage Theory （OACA） derived from it has been proven by many studies. However, some academic journals have experienced the phenomenon of "reverse flip", that is, the flip from the open access model to the paid subscription model, which runs counter to the OACA theory. Analyzing the citation disadvantage caused by this phenomenon can provide suggestions for relevant subjects.[Design/Methodology] This study gathered 18,615 papers in 12 reverse-flip journals as the dataset, used matching and Difference in Difference （DiD） in causal inference, and analyzed the difference in citation counts between papers publish before and after the reverse flipping. [Findings/Conclusion] The study found that after reverse flipping, the citation counts of papers in these journals were significantly reduced, which confirms the citation disadvantage of papers in non-OA journals. But such disadvantage diminishes along with time. [Originality/Value] In this study, the phenomenon of "reverse flip" rarely mentioned by previous researchers is taken as the research object and the causal inference method is used to investigate its influence on the academic journals, which has made contribution to the theories related to Open Access. 

Key words: Open access, Reverseflip, Academic journal, Academic impact, Causal inference, Difference in difference