The Abbreviated Math Anxiety Scale (AMAS) is one of the most popular instruments measuring math anxiety (MA). It has been validated across several linguistic and cultural contexts. In this study, we investigated the extent of administration method invariance of the AMAS by comparing results (average scores, reliabilities, factorial structure) obtained online with those from paper-and-pencil. We administered the online version of the AMAS to Polish students. Results indicate that psychometric properties of the AMAS do not differ between online and paper-and-pencil administration. Additionally, average scores of the AMAS did not differ considerably between administration forms, contrary to previous results showing that computerized measurement of MA leads to higher scores. Therefore, our results provide evidence for the usefulness of the AMAS as a reliable and valid MA measurement tool for online research and online screening purposes across cultures and also large similarity between administration forms outside an American-English linguistic and cultural context. Finally, we provide percentile and standard norms for the AMAS for adolescents and adults (in the latter case for both online and paper-and-pencil administration) as well as critical differences for the comparison of both subscales in an individual participant for practical diagnostic purposes.