In　China,　the　Chevron　alignment　sign　on　highways　is　a　vertical　rectangle　with　a　white　arrow　and　border　on　a　blue　background,　which　differs　from　its　counterpart　in　other　countries.　Moreover,　little　research　has　been　devoted　to　the　effectiveness　of　China＇s　Chevron　signs;　there　is　still　no　practical　method　to　quantitatively　describe　the　impact　of　Chevron　signs　on　driver　performance　in　roadway　curves.　In　this　paper,　a　driving　simulator　experiment　collected　data　on　the　driving　performance　of　30　young　male　drivers　as　they　navigated　on　29　different　horizontal　curves　under　different　conditions　(presence　of　Chevron　signs,　curve　radius　and　curve　direction).　To　address　the　heterogeneity　issue　in　the　data,　three　models　were　estimated　and　tested:　a　pooled　data　linear　regression　model,　a　fixed　effects　model,　and　a　random　effects　model.　According　to　the　Hausman　Test　and　Akaike　Information　Criterion　(AIC),　the　random　effects　model　offers　the　best　fit.　The　current　study　explores　the　relationship　between　driver　performance　(i.e.,　vehicle　speed　and　lane　position)　and　horizontal　curves　with　respect　to　the　horizontal　curvature,　presence　of　Chevron　signs,　and　curve　direction.　This　study　lays　a　foundation　for　developing　procedures　and　guidelines　that　would　allow　more　uniform　and　efficient　deployment　of　Chevron　signs　on　China＇s　highways.　(C)　2016　Elsevier　Ltd.　All　rights　reserved.