Translation studies is a field of interdisciplinary research, and one of the factors that has driven its rapid development in recent times is methodological innovation. Translation scholars are one of the most dynamic research communities working across the boundaries of the arts and humanities, social and natural sciences. From comparative literature, bilingual and multilingual education to textual statistics, technology localisation, and machine learning modelling of large multilingual translation databases, for decades, we have been working passionately, tirelessly to advance the understanding of cross-cultural, cross-lingual translation. Despite the availability of constantly improving automatic machine translation technologies, translation studies, which explore the underlying principles, methods, and mechanisms of human translation activities, have strived around the world. This reflects increasing demands from different cultures, societies, and communities for high quality human translations and human-centred translations, which cannot be replaced by machine translation algorithms. Translation is never a straightforward activity. It centrally reflects the complex, subtle, and context-dependent nature of human communication. Existing translation theories tend to be operational at the macro level, exploring the social and cultural impact on the selection and use of certain translation strategies conceptualised as language-independent norms. This theoretical approach has proven effective when the reception of translation is a collective social behaviour. In recent studies, more importance is given to the impact of individual differences among readers on the design and development of translation resources, for example, for the purpose of public health education and communication. Translations which are adaptive to the varying reading habits and abilities of individuals tend to be successful with regards to their communicative effectiveness. From the communication of health risks to climate change, translation is playing an important role in reducing health and environmental inequalities, misunderstandings, and confusion in a world of uncertainties. New research questions that have emerged in our time include how to translate credible, critical information, for example, on health and environmental issues more effectively to global readers. Research papers in this special issue illustrate that the development of multilingual resources for environmental communication represents another contribution that the translation community is making to the broader academy and to societies. For the purposes of health education and promotion, health translation requires higher understandability, readability, and accessibility. To translate effectively requires an in-depth understanding of the practical needs, reading habits, and reading abilities of the readers as individuals. This represents a missing link in many current translation practices. Even with machine translation, which exhibits increasing accuracy and fluency, few studies have addressed the pressing needs for human-centred translations. This special issue aims to foster scholarly debate around the value of new research evidence and the development of research methods to effectively process, analyse, and interpret translations. This is a continuation of the empirical translation studies envisaged by earlier scholars. The ongoing pandemic provides the social background for this issue on the data turn in translation studies, around which we discuss the development of integrated quantitative and qualitative approaches to address socially oriented research questions. Three of the articles in this special issue investigate the question of translation understandability. The first is Silvia Rodriguez-Vasquez, Abigail Kaplan, Pierrette Bouillon, Cornelia Griebel, and Razieh Azari’s contribution, entitled “La traduction automatique des textes faciles à lire et à comprendre: une étape comparative” [Machine translation of texts that are easy to read and understand: a comparative study]. The use of controlled languages (CL) has long been associated with machine translation (MT). Early MT systems such as TAUM-METEO, which translated weather reports between French and English, owed their success to the absence of ambiguity, as well as the restricted range of vocabulary and syntactic structures, in the …