The goal of this project was to conduct a high-powered and faithful replication of Experiment 1 in Damisch et al. (2010) showing that superstition greatly improves golf performance. Damisch, L., Stoberock, B., & Mussweiler, T. (2010). Keep your fingers crossed!: How superstition improves performance. Psychological science, 21(7), 1014–20. doi:10.1177/0956797610372631 **Status** - Complete. All data collected, all materials and data posted to this project site, manuscript written, submitted, and revised for a special issue on replication in *Social Psychology*. Final registration of this project freezes in place all these materials--only corrections will be posted from this point out. The manuscript describing this project has now been published: Calin-Jageman, R. J., & Caldwell, T. L. (2014). Replication of the Superstition and Performance Study by Damisch, Stoberock, and Mussweiler (2010). Social Psychology, 45(3), 239–245. [http://doi.org/10.1027/1864-9335/a000190][1] **Materials** All materials have been uploaded. This includes a video demonstrating the protocol posted [here][2]. **Results**: First study - We completed the first replication. Raw data is uploaded in SPSS and Excel format. We found that participants in the lucky condition scored about the same (*M* = 4.73, *SD* = 1.96, *n* = 66) as those in the control condition (*M* = 4.62, *SD* = 2.13, *n* = 58). This was true despite strong impact on a retrospective rating of how lucky the participants felt. Second study - We completed a second replication designed to have higher impact. In this case, participants drew a golf ball from a sack. Unbeknownst to them, 1/2 were regular balls but 1/2 were emlazoned with four-leaf clovers. Raw data in SPSS format is now uploaded. We again found no effect of superstition activation. Those in the lucky group scored about the same (*M* = 4.12, *SD* = 2.01, *n* = 66) as those in the control group (*M* = 4.02, *SD* = 2.20, *n* = 54). Meta-analysis - Completed and results table uploaded **Registrations** **First registration: ** The project was produced as a registered replication for a special issue of Social Psychology on replication. First, a replication proposal was submitted detailing the precise methods and analyses to be used to faithfully replicate the experiment. This was reviewed and suggestions were incorporated. The first registration on this project locks in the methods, materials, and analysis strategy for the project. The complete project proposal is [here][3]. **Second registration**: Data collection was scheduled for September 2013. Shortly before this, however, the lead author of study of interest, Lysann Damisch, reviewed a video of our protocol and emailed back with comments and suggestions for increased fidelity of replication. Specifically, Damisch noted: - That a cover story was used in the original, indicating that adapting to new tasks like the golf task are predictive of future success (we thus added this cover story) - That the golf target was not placed against a wall (we thus moved it out 100cm, but preserved the 150cm distance from the tee) - That demographics had been collected via an online questionnaire (we thus switched our paper/pencil format to online) - That psychology majors and left-handers had been excluded from the study. We had already settled on an recruitment strategy, so we did not implement this suggestion. However, our participants were drawn from biology lab classes and had relatively few psychology majors. We collected information about major so that psychology majors could be excluded post-hoc. The second registration locks in these changes. Unfortunately, we didn't upload the updated online questionnaire until 10/6/2013, after data collection on Study 1. However, these changes were minor and were suggested by the lead author, and were committed to prior to any data collections. **Third Registration**: Based on results of our first replication attempt, we decided to try again using a higher manipulation impact. Specifically: - Participants drew their golf ball from a sack containing 4 regular balls and 4 marked with a shamrock (an Irish sign of good luck). The experiment said either "This is the ball you will use" or "Wow! You get to use the lucky ball". - We updated the second manipulation check to match this manipulation. Specifically, participants viewed a photo of each type of ball and reported which type they had selected. - After the golf task and after completing a rating of how lucky they felt (see original proposal), participants were also asked how much they believed in luck on a scale from 1-9. This was an item Damisch et al. had used during piloting to assess the belief in luck in their participant population. We added this to determine the match in the control condition with their study. It came after all relevant manipulation and DV measurement. - Just before the debriefing, we also asked participants to report their golf score. We did this just to save time on data entry, after all manipulation and DV measurement. The experimenter still recorded the score manually on a data sheet to enable cross-checking of participant's data entry. We maintained the same goal of at least 42 participants per condition to enable sufficient power given the estimated effect size of superstition on performance (see proposal). This second study has now been completed. **Final Registration** - Manuscript written, revised with feedback from editor, and now posted along with all data files, procedure video, etc. Project complete and final. No further edits will be made to this project unless errors requiring corrections are identified. [1]: http://doi.org/10.1027/1864-9335/a000190 [2]: https://openscienceframework.org/project/AWR6j/files/protocol_video_-_final.wmv/ [3]: https://openscienceframework.org/project/AWR6j/files/Replication_of_Superstition_and_Performance_Study_by_Damisch_-_r2d2.docx