Currently, information about the psychometric properties of the Beck Anxiety Inventory (BAI) in family caregivers of children with cancer is not available; thus, there is no empirical evidence of its validity and reliability to support its use in this population in Mexico or in other countries. This study examined the psychometric properties of the BAI in family caregivers of children with cancer and pursued four objectives: to determine the factor structure of the BAI, estimate its internal consistency reliability, describe the distribution of BAI scores and the level of anxiety in the sample and test its concurrent validity in relation to depression and resilience. This cross-sectional study was carried out with convenience sampling. A sociodemographic questionnaire, the BAI, the Beck Depression Inventory and the Measurement Scale of Resilience were administered to an incidental sample of 445 family caregivers of children with cancer hospitalized at the National Institute of Health in Mexico City. Confirmatory factor analysis using the maximum likelihood method was performed to determine the factor structure and exploratory factor analysis using axis factorization with oblique rotation was conducted. The two-, three- and four-factor models originally proposed for the BAI did not hold. The exploratory factor analysis showed a model of two correlated factors (physiological and emotional symptoms). Confirmatory factor analysis revealed a lack of discriminant validity between these two factors and supported a single-factor model. The internal consistency of the scale reduced to 11 items (BAI-11) was good (alpha = 0.89). The distribution of BAI-11 scores was skewed to the left. High levels of symptoms of anxiety were present in 49.4% of caregivers. The scale was positively correlated with depression and negatively correlated with resilience. These findings suggest that a reduced single-factor version of the BAI is valid for Mexican family caregivers of children with cancer.