Yes. Here's one, for example: https://www.researchgate.net/publication/229989870_Does_Community_Service_Rehabilitate_better_than_Short-term_Imprisonment_Results

Yes. Here's one, for example: https://www.researchgate.net/publication/229989870_Does_Community_Service_Rehabilitate_better_than_Short-term_Imprisonment_Results_of_a_Controlled_Experiment https://www.researchgate.net/publication/229989870_Does_Comm... This meta-study pulls together RCTs and experimental programs and covers non-custodial (which includes a range of options besides community service) v custodial, and finds them to be at worst equivalent for recidivism https://onlinelibrary.wiley.com/doi/full/10.4073/csr.2015.1 https://onlinelibrary.wiley.com/doi/full/10.4073/csr.2015.1