Abstract: Fifteen residential districts from r1 to r15 were selected in Xinxiang City in this study. Seven indicators, i.e., vegetation landscape (x1), lighting landscape (x2), hard landscape (x3), shelter landscape (x4), simulation landscape (x5), water landscape (x6) and place landscape (x7), were selected. AHP method was used to establish the impact factor assignment standards. We established one regression equation between impact factors (xi) and residential landscape environment attraction (y) with the multiple regression analysis. The multivariate linear regression equation is y=-26359+6181x1+1 583x2+1247x3+1123x4+026x5+017x6+2230x7, with the contribution rate sorting as x1> x7> x2> x3> x4> x5> x6. The model holds good reliability and strong predictive ability. It indicates that measurable evaluation on residential environment landscape attraction is feasible. From the contribution rates of the model, it can be concluded that, the following three landscapes should be placed on the important status in the environmental landscape design in northern cities, such as vegetation landscape fulfilling the needs of the ecological and aesthetic, place landscape fulfilling the needs of various activities and lighting landscape which is of great importance to night landscape. The hard landscape and shelter landscape with higher average unit cost should be a choice, whereas water landscape and its application should be very careful.