In this paper, we investigate gender assignment to recently borrowed English loanwords in Dutch, introducing several innovations to the field of gender assignment to anglicisms. For example, we use multiple mixed-effects logistic regression to determine which factors underlie gender assignment in Dutch. This model indicates that there is variation in the degree of homogeneity in the speech community (that is, agreement among respondents) concerning the gender assigned to an anglicism; therefore, we analyze the contexts in which homogeneity is the lowest. Our analysis reveals that the degree of consensus does not solely depend on how established an anglicism is. In contrast to what has been argued in previous studies, gender assignment to anglicisms in Dutch is not a categorical process: Gender variation increases when respondents are faced with a conflict between the default article de and some factor that favors its neuter counterpart, het.