Impact of Income, Life Expectancy and other factors on the Happiness Index

This study analyses the impact of population, Per Capita Personal Income (PCPI), Literacy Rate (%), Murder Rates, High School Graduation Rate, Average Temperature °F, Land size (Sq. Miles), Poverty rate (%) on the Life Expectancy (Years) in the USA. The aim of this study is to understand how different factors affect the life expectancy in the different states in the USA. In any society, the Life Expectancy is determined by a myriad of factors such as genes, food availability (scarcity), exercise, access to clean water, access to clean air, access to health facilities among many others. The correlation coefficients between Life Expectancy and the predictors are as follows: Population (0.24), Per Capita Personal Income (0.67), Literacy (0.04), Murder (-0.46), High School Graduation rate (0.43), Average Temperature (-0.29), Land Size (0.07) and Poverty rate (-0.75). The reduced model shows that four quantitative predictors and the categorical predictor (region) are significant in explaining the variability in the life expectancy (F = 30.47, p-value = 1.088 x 10-14 ≈ 0.000). This indicates that we reject the null hypothesis that there is no significant regression and conclude that there is significant regression between the Life Expectancy and the predictor variables. The explanatory variables that are significant (at 95% level of confidence) in explaining the variability in the life expectancy are region, population, per capita income, average temperature and poverty. The value of the adjusted coefficient of determination (R-squared) in the reduced model is 0.8049 which implies that the model has a good fit. 80.49% of the variability in the life expectancy is explained by the predictors in this study holding all other factors constant. It is important for policy makers to make use of the findings of this study in order to improve the life expectancies across all the states. Keywords: Life Expectancy, Predictors, Regression, Response, Significant, Explanatory