Kindergarten redshirting may affect a child's own outcomes and also has implications for school administration, classroom management, and peer learning. We use statewide micro-level census data to examine selection into redshirting, potential spillover effects, and its association with third grade outcomes. We find evidence of both negative and positive selection into redshirting, where children with disabilities are much more likely to be redshirted. We find small positive associations between redshirting and both math and reading achievement in third grade for students without identified disabilities. However, redshirting students with an identified disability score statistically significantly lower on mathematics assessments compared to similar non-redshirting students with identified disabilities. We do not find evidence of spillover effects from redshirting when students attend third grade classes with higher proportions of redshirted children. (author abstract)