Measurement of age-of-acquisition in morphologically rich languages: Insights from Kannada and Filipino

Abstract

In this study, we present age-of-acquisition (AoA) ratings for 885 Kannada and Filipino words as a new resource for research and education purposes. Beyond this, we consider the methodological and theoretical considerations of measuring AoA in morphologically rich, specifically agglutinative, languages, to study child language acquisition. Parents, teachers, and experts provided subjective ratings of when they thought a child acquired each word. Results were generally consistent between the two languages. Mixed-effects models demonstrated that word characteristics, including parts-of-speech category, word length, and age band of first occurrence in a print corpus, were significantly related to AoA ratings, whereas rater characteristics, including participant type, age, gender, and number of languages spoken, had generally non-significant associations with AoA ratings. The number of morphemes was significantly associated with AoA ratings in some analyses; however, crosslinguistic differences in the directionality of the relationships suggested the need to investigate underlying drivers of morphological complexity such as morpheme frequency, transparency/consistency, and function. The age-of-acquisition ratings were internally reliable and demonstrated consistency with the first occurrences of words in print and known trends in child language research. The results demonstrate the potential of these resources and open new directions for AoA research in morphologically rich languages.