To determine school attendance and its predictors among children with cerebral palsy (CP) in Bangladesh using population-based data. This study utilized data from the Bangladesh Cerebral Palsy Register (BCPR), a population-based register of children with CP aged less than 18 years in Bangladesh. Sociodemographic, clinical, and educational data were documented, and descriptive statistics and multivariate regression analyses were used to identify potential predictors of school attendance. Between January 2015 and January 2019, 2725 children with CP were registered into BCPR of which 1582 were school-aged children (i.e. aged 6-18 years). The majority of those children had not attended school (73.9%); 50% (n = 239) children in Gross Motor Function Classification System (GMFCS) levels I to II did not attend schools. Adjusted odds ratios (OR) showed significantly higher odds of school attendance among children whose mothers had completed secondary education or higher (adjusted OR: 2.2; 95% confidence interval [CI]: 1.2-4.0) and received rehabilitation (adjusted OR: 2.1; 95% CI: 1.4-3.1). Conversely, lower odds of school attendance were observed among children aged 15 to 18 years (adjusted OR: 0.4; 95% CI: 0.2-0.8), those with bilateral CP (adjusted OR: 0.5; 95% CI: 0.3-0.8), GMFCS levels III to V (adjusted OR: 0.3; 95% CI: 0.2-0.5), Manual Ability Classification System levels III to V (adjusted OR: 0.5; 95% CI: 0.4-0.8), and speech impairment (adjusted OR: 0.3; 95% CI: 0.2-0.6). A large number of children with CP in Bangladesh, including half of those with milder forms, do not attend schools. These findings underscore a deficiency in awareness and support, encompassing the provision of resources and trained teachers, highlighting the necessity for policy-level changes. Sociodemographic and clinical predictors should be taken into account when devising educational programmes to enhance school attendance for children with CP in Bangladesh.