DOI: 10.36347/sjahss.2016.v04i11.006

Pages: 1399-1410

This paper examined the spatial dependency in the voting pattern of the 2015 Nigeria’s Presidential election. Data for the study was obtained from the Independent National Electoral Commission as published on its website. The study was particularly interested in the percentage of votes won within each State and the Federal Capital Territory by the two main political parties in the election: the All Progressive Congress (APC) and the People’s Democratic Party (PDP). Shape files containing polygons of the various States, the Federal Capital Territory, Abuja and the Nigerian international boundary were obtained from the Geographical Information Systems (GIS) Unit of the Department of Geography and Environmental Science, University of Calabar, Calabar. The data were pre-processed and analyzed mainly within the GIS environment. Both the join count statistics and the Moran’s I were used to analyze and test for spatial autocorrelation in the data set. The results of join count statistics revealed that the estimated number of connections between the “presence” and “absence” of a winning by APC was 42.88 with a standard deviation of 4.35. The observed join count was 24. For the bilateral test of significance(H1: OPA ≠ EPA), |Z_obs| > |Z_crit|, hence the evidence was not enough to retain the Ho. Therefore, the observed distribution is significantly different from a random distribution. For the unilateral test, (H1: OPA< EPA), Z_obs<Z_crit.