This study of the effect and characteristics of Selective Availability (SA) is described in this thesis. Two approaches, utilising differential and absolute positiioning methods, are used in the study. In the frst approach, a real GPS data collected at two stations are used. In processing one is assumed as a reference station and the other as a remote station. The positions of remote station and the errors at selected ages of correction are computed using the available programs. The variation of errors at each age of correction are presented in graphs. For the second approach, the nature of errors in the data at a single point are analysed to study the elective Availability level and any chages in its characteristic. In the research, data collection has been carried out two days a month in period of November, 1991, to April, 1992. A comparison of the results computed from data with and without SA in the above period shows the differences in terms of the effect and characteristics of SA. The spectral technique is used to identify the characteristics of SA in terms of frequency. This technique of analysing any GPS or surveying data is considered as a new approach. The concept of this technique has been investigated and used to analyse the characteristics of SA. There are three programs used for the processing the collected data. A program of RINEXPOS developed at the Department of Surveying, The University of Newcastle upon Tyne is available for positioning computations. The program which was developed by the Author is used to compute the errors adn its graphical representation. The MATLABTM Signal Processing programs are purposely used to carry out the spectral analysis.