Based on a meta-analysis, this study aimed to examine cross-informant agreement between parents and teachers about Internalizing, Externalizing and Total Problems in preschool children using community and clinical samples and to test the effects of the type of sample, the measure used for assessments, and child sex on agreement between informants. The meta-analysis involved 23 studies assessing cross-informant agreement for preschool children. Informants were parents and teachers. The level of cross-informant agreement tended to be low. Meta-regression analyses showed that the child’s sex, the type of sample, and the measure used for assessments did not predict the level of cross-informant agreement on emotional and behavioural problems. The findings were in line with previous research results. Furthermore, the studied variables did not contribute to the prediction of agreement, suggesting the development of further studies that focus on other variables that may interfere with agreement in informants’ reports and will contribute to explaining different ratings of internalizing and externalizing problems in preschool-aged children.