Bivariate Analysis

Only available on StudyMode
  • Download(s) : 355
  • Published : November 27, 2010
Open Document
Text Preview
Bivariate Analysis
Objective-To find the effect of money spend as a percentage of GDP(X) on education on the infant mortality rate (Y). Priori Reasoning – Educated people are well aware of the needs of a pregnant mother and take good care of her. Hence higher the education spend better is the infrastructure and the corresponding education level and lower will be the death of the infants. Null Hypothesis

H0: There exists a relationship between percentage money spent on gdp and the infant mortality rate. Alternate Hypothesis
H1: There is no relation between percentage money spent on gdp and the infant mortality rate Independent Variable – Percentage of GDP spent on education Dependent Variable – Infant deaths/1,000 Live births

Significance level: 98%
TYPE OF DATA-Cross sectional data.
Data was taken for 122 countries.
Independent var| Dependent var| Model| Equation| R square| Df| F| Significance level| Y| x| Linear| Y=a+bx| .017| 1| 2.003| .16|
Log y| x| Log Linear| Logy=a+bx| .031| 1| 3.774| .054| Y| x,x^2| Quadratic| Y=a+bx+cx^2| .041| 2| 2.551| .082| Y| X,x^2,x^3| Cubic| Y=a+bx+cx^2+dx^3| .082| 3| 3.497| .018|

Independent var| Dependent var| a| b| c| d| ta| tb| tc| Td| Y| x| 39.158| -1.863| -| -| 5.761| -1.415| -| -|
Log y| x| 3.55| -.0431| -| -| 10.937| -1.943| -| -| Y| x,x^2| 3.904| -.334| .023| -| 8.727| -2.179| 1.876| -| Y| X,x^2,x^3| 5.447| -1.32| .195| -.008| 6.669| -2.884| 2.553| -2.282|

Based on the results we can see that the data is not a fit with any of the models, hence we cannot say that the % of gdp spent on education and infant mortality rate are linearly related. There are several other factors that affect infant mortality rate in a country.

Multivariate Analysis

The Objective-To find the effect of different variables on crime rate in different countries A Priori Reasoning – Crime rate...
tracking img