Objective : This study aims to suggest the basis for the development of evaluation items used to measure the community participation of stroke patients. Methods : Opinions were collected from 32 community and stroke-related expert panels using the Delphi method. The first survey gathered opinions through items reconstructed using closed-ended·open-ended questions for community participation, while the second survey determined the goodness of fit and its importance. The third survey was a final check of the opinions by providing a response and results to the second survey. Results : A total of 38 items were selected from the first Delphi results. For the second Delphi results, five items showing a value of .33 or lower for the content validity ratio were removed to make a total of 33 items. For the third Delphi results, the 33 items were deemed to be fit, and were therefore selected. Finally, the average content validity ratio of the Delphi results was .93, the stability was .11, the convergence was .40, and the consensus was .83, which are regarded as high scores. Conclusion : Through this study, items for measuring the degree of participation of stroke patients in their community were developed. In is expected that these results can be used as a basic set of data to promote more participation by stroke patients in their communities.