This is an accepted version of a paper published in Journal of Nursing Education and Practice. This paper has been peer-reviewed but does not include the final publisher proof-corrections or journal pagination.Citation for the published paper: Hedberg, P., Lämås, K. Methods: A pilot study with an explorative approach including 30 nurse students. Students was randomized in three groups; 1) instructor-led training followed by self-training without feedback, 2) self-training with visual graphic feedback, and 3) self-training with voice advisory manikin (VAM). Outcomes were correct compression deep, frequency, hand position and release, and correct ventilation volume and flow. If performance was correct to 70%, students were considered to have reached approved level. The students also answered questions about theoretical knowledge about CPR.Results: In technical skills, group 2 had significant higher level of correct ventilation volume compared with the other group. Both group 1 and 3 did not reach the level of 70% correct performance. Group 1 and 2 had significant higher level of correct deep of compressions compared with group 3 which did not reach the 70% level. There was no difference in performance between groups in other parameters.Conclusion: This pilot study suggests that visual graphic feedback is promising and seemed to be more effective than self-training with voice advisory manikin and instructor-led training with followed self-training without feedback.