The aim of this study was to summarize evidence about the relations between parent math support in children aged 3-5 years (from several countries in America, Asia, and Europe) and concurrent and longitudinal math outcomes. The (bio)ecological model of human development guided our hypotheses. The design and reporting of this meta-analysis used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). We screened 2,163 abstracts, from which 135 full-text studies were assessed for eligibility. Forty-five studies fulfilled our inclusion criteria and were retained (representing 244 effect sizes); 90 studies were discarded as they did not include preschool children or measures of both parent math support and children's math skills. Meta-analysis using Correlated and Hierarchical Effect (CHE) models showed a significant positive association between parent math support and child math skills for both concurrent and longitudinal studies. On average, higher parent math support was associated with better math skills in children, albeit these being small effect sizes. We found non-significant or inconclusive moderator effects related to formal vs. informal parent math support, type of children's math skills, participants' characteristics (e.g., child age and child/parent gender), parent education, and study characteristics. There was a significant moderator effect of the specificity of parent math support, with global assessments showing higher correlations with math outcomes than specific assessments. The publication bias analysis showed small-study effects for longitudinal but not for concurrent studies. Conclusions are drawn regarding the importance of promoting parent math support and informing future intervention studies.