The noun phrase (NP) is at the heart of several studies investigating regional variation in varieties of English. While so far the bulk of research has focused on isolated structural features, the present study is a comparative analysis of NP complexity across varieties of English. NP complexity is compared across five regional varieties and four text categories, based on data from the International Corpus of English. The study adopts a multinomial regression approach, which takes into consideration the interaction of three potential predictors: syntactic function, text type, and variety. The results underline the need for text-type-sensitive studies and add to an understanding of syntactic contact phenomena in varieties of English. More specifically, it finds marked differences in the predictive power of the variables and illustrates how focusing on the interaction of syntactic functions, text type and regional variety contributes to a systematic description of variation in the NP in world Englishes.