Introduction

Life expectancy on a global scale varies tremendously. Using data from the World Bank I have distinguished three of the most imperative factors that influence life expectancy. These are as follows: health expenditure as a percentage of GDP, access to improved drinking water sources, and access to improved sanitation facilities. The data has been refined to only include countries which have complete information on the variables I have just mentioned. This leaves us with 179 countries to observe. Furthermore, I have taken the average of the 179 countries between the years of 1995 to 2012.

Anaysis

For this analysis I have re-coded the life expectancy variable in order to fully understand the phenomenon we are studying. The average age of life expectancy for this data-set is 67.53. Thus the following analysis has been re-coded as such that the results can fall in a percentage either above or below the average life expectancy.

Table 1.

Dependent variable:
lifeaverage
san 0.107***
(0.019)
health -0.040
(0.110)
water 0.095***
(0.034)
Constant -16.449***
(3.294)
Observations 179
Log Likelihood -49.920
Akaike Inf. Crit. 107.839
Note: p<0.1; p<0.05; p<0.01


The table above shows that access to improved sanitation is statistically significant, with access to improved water conditions coming close behind. Health expenditure as a percentage of GDP is not significant at all in this analysis. Thus, for each year unit increase in life expectancy, the odds of access to improved sanitation increases by .107.


Table 2.

How to cite this model in Zelig: Kosuke Imai, Gary King, and Olivia Lau. 2015. “logit: Logistic Regression for Dichotomous Dependent Variables” in Kosuke Imai, Gary King, and Olivia Lau, “Zelig: Everyone’s Statistical Software,” http://gking.harvard.edu/zelig

Dependent variable:
lifeaverage
san 0.380
(0.306)
water 0.356
(0.297)
health -0.029
(0.111)
san:water -0.003
(0.003)
Constant -40.719
(27.821)
Observations 179
Log Likelihood -49.445
Akaike Inf. Crit. 108.891
Note: p<0.1; p<0.05; p<0.01


Table 2 is the same regression as table 1 but shows the interaction terms between access to improved sanitation and improved water.


Table 3.

How to cite this model in Zelig: Kosuke Imai, Gary King, and Olivia Lau. 2015. “logit: Logistic Regression for Dichotomous Dependent Variables” in Kosuke Imai, Gary King, and Olivia Lau, “Zelig: Everyone’s Statistical Software,” http://gking.harvard.edu/zelig

Dependent variable:
lifeaverage
san 0.097***
(0.021)
water -0.023
(0.119)
health -2.667
(2.586)
water:health 0.028
(0.028)
Constant -4.659
(11.728)
Observations 179
Log Likelihood -49.365
Akaike Inf. Crit. 108.730
Note: p<0.1; p<0.05; p<0.01


Table 3 is similar to the tables above, but includes the interaction terms between the health and water variable.


Graphs


The following graph shows as the percentage of people living past the average life expectancy increase, access to improved sanitation increases as well.

How to cite this model in Zelig: Kosuke Imai, Gary King, and Olivia Lau. 2015. “logit: Logistic Regression for Dichotomous Dependent Variables” in Kosuke Imai, Gary King, and Olivia Lau, “Zelig: Everyone’s Statistical Software,” http://gking.harvard.edu/zelig


The following graph shows that as the percentage of people living past the average life expectancy increase, access to improved drinking water increases as well.

How to cite this model in Zelig: Kosuke Imai, Gary King, and Olivia Lau. 2015. “logit: Logistic Regression for Dichotomous Dependent Variables” in Kosuke Imai, Gary King, and Olivia Lau, “Zelig: Everyone’s Statistical Software,” http://gking.harvard.edu/zelig

Conclusion

The results of this analysis tell us that water and improved sanitation hold greater weight than health expenditure when looking at life expectancy rates on a global scale. More specifically, access to improved sanitation is that strongest predictor when it when comes examining life expectancy. These results have been novice in the sense that common knowledge has alluded to the order of importance be in reverse for life expectancy. However these findings suggest the basic fundamentals of access to improved sanitation is extremely valuable when it comes to life expectancy on a global scale.