Assessing Multicollinearity in Population Data Using the Farrar-Glauber Approach

Abstract

The purpose of this study is to determine whether multicollinearity exists, how severe it is, where it occurs, and what pattern it exhibits. Additionally, it aims to rectify any multicollinearity between the rates of death, fertility, life expectancy, and maternal mortality. The study covers distribution and trend of multicollinearity in the population rate of Nigeria between 2005 and 2024.  The intercorrelation between the different explanatory variables as determined by multiple correlation coefficients was ascertained using tests (chi-square, F-test, and T-test). According to the findings, the degree of multicollinearity was high, as indicated by the Pearson value of 0.9 and the R² value of 0.999 (99%).