The effect of deep soil deposits in Phoenix Basin on the seismic hazard for a 2500 years return period was evaluated using four different methods, including: 1) the results of a seismic hazard analysis for reference site conditions and National Earthquake Hazard Reduction Program (NEHRP) site factors; 2) the results of a seismic hazard analyses using Next Generation Attenuation (NGA) relationships; 3) one-dimensional equivalent linear site response analyses, and 4) onedimensional non-linear site response analyses. Use of the NEHRP site factors yielded the lowest values of the four methods for the spectral accelerations required for design. Non-linear site response analyses, considered the most accurate method, yielded the highest values. Results of the equivalent linear analyses were systematically lower than the results of the non-linear analyses. Results of the seismic hazard analysis conducted using the NGA relationships fell in between the results of the equivalent linear and non-linear analyses at periods of 0.2 s or greater. The results suggest that, for the case of an area of relatively low seismicity that was considered herein, the NGA relationships are capable of accounting for deep soil basin site-specific response effects directly in a seismic hazard analysis. However, it must be recognized that the analysis conducted herein was for an area of relatively low seismicity and the differences among the various types of analyses may change significantly if higher levels of seismicity or softer ground conditions are considered.