Application of Kolmogorov-Smirnov Test to Evaluation of Population Distrbution by Age in Single Years

This study examines the evaluation of population distribution by sex and age in single years. The aim is to assess the validity of the assumption that when data on population by age in single years is correctly reported then the population is uniformly distributed among end digits. The data used is a secondary data derived from the census populations reported by age in single years from Nigeria, Ghana, Indonesia and United State of America. Measures of digit preference are computed using the Ramachandran Index (RI), one of the most commonly algebraic methods of evaluation of age data in single years. The Kolmogorov-Smirnov test was used to test the Goodness-of-Fit of the discrete uniform distribution to the distribution of population among end digits in the selected countries. The results of the analyses show that the null hypotheses were all rejected at 1% level of significance, indicating that the distribution of population among the 10 end digits is not uniform for both sexes in all the selected countries. In other words, the distribution of population by age in single years in none of the countries studied is completely free from digit preference/avoidance. The study therefore, recommends that the age data should be adjusted for the error of digit/avoidance before using them for further estimation of demographic parameters or at least used with caution.