The aim of this paper is to compare the validity of six recent symmetric mapping functions. The mapping function models the elevation angle dependence of the tropospheric delay. Niell Mapping Function (NMF), Vienna Mapping Function (VMF1), University of New Brunswick- VMF1 (UNB-VMF1) mapping functions, Global Mapping Function (GMF) and Global Pressure and Temperature (GPT2)/GMF are evaluated by using ray tracing through 25 radiosonde stations covering different climatic regions in one year. The ray-traced measurements are regarded as “ground truth”. The ray-tracing approach is performed for diverse elevation angle starting at 5° to 15°. The results for both hydrostatic and non-hydrostatic components of mapping functions support the efficiency of online-mapping functions. The latitudinal dependence of standard deviation for 5° is also demonstrated. Although all the tested mapping functions can provide satisfactory results when used for elevation angles above 15°, for high precision geodetic measurements, it is highly recommended that the online-mapping functions (UNBs and VMF1) be used.The results suggest that UNB models, like VMF have strengths and weaknesses and do not stand out as being consistently better or worse than the VMF1. The GPT2/GMF provided better accuracy than GMF and NMF. Since all of them do not require site specific data; therefore GPT2/GMF can be useful as regards its ease of use.