Background: The myopia is a public health issue that attracts much attention. However, limited attention has been paid to the effect of primary school students' acceptance of health messages. Previous studies have found that framing effects and evidence types influence the persuasive effect of messages.Purpose: This study explored whether framing effects and evidence type influence the persuasive effect of myopia prevention messages among elementary school students and the influence of children's myopia prevention cognition was considered.Methods: A cross-sectional study was conducted among 1,493 elementary school students aged 9 to 13 in China from May to July 2020 by convenience sampling. Wilcoxon signed-rank test and multinomial logistic regression were used for data analysis.Results: Significant differences were found in the persuasive effect between statistical and non-statistical evidence messages (p < 0.001). Among non-statistical evidence messages, gain-framed messages showed a greater persuasive effect than loss-framed messages (p < 0.001). Among statistical evidence messages, loss-framed messages performed better than gain-framed messages (p < 0.001). Children's myopia prevention cognition exerted no significant effect on the persuasive effect of the messages (p > 0.05).Conclusion: This study demonstrated the influence of framing effect on the persuasive effect of myopia prevention messages among children aged 9 to 13 in China. Non-statistical evidence messages showed a better persuasive effect than statistical evidence messages. Different types of evidence influenced the persuasive effect of gain- and loss- framed messages. These findings have implications for strategies more or less likely to work in making myopia prevention messages for children.