Crossing the line: Effect of border representation in perceptual dialectology

Abstract

This study investigates the effect of differing representations of state boundaries on the draw-a-map task in perceptual dialectology in a region of the United States. The typical draw-a-map survey instrument represents state borders with solid lines. Would respondents react differently to maps with dashed-line state borders? More specifically, would respondents draw more dialect areas that cross state lines on maps with dashed-line state borders versus solid-line state borders? These questions are explored through two datasets, and similarities and differences emerge. For example, respondents of both map types draw more single-state dialect areas than multistate dialect areas, and respondents with dashed-line maps draw more dialect areas on average than respondents with solid state maps. While dataset 1 showed a significant association between map type and multistate dialect area with respondents using dashed-line border maps drawing more multistate dialect areas than respondents with solid-line maps, H(1) = 5.13, P = .017, this association was not significant in dataset 2, H(1) = .06, P = .798.