Project title

Preregistration of analyses

How / if does the net worth of a billionaire affect the number of children they have?

This question will deal with analyse how the amount of wealth a billionaire has affects the number of children they have. For this, we will clean the data and look into the variables \(Children\) and \(NetWorth\). First, we will clean the data by replacing all blank entries in Children by zero. Then, we will create a bar chart to help us visualize the relationship between the two variables so that we can assess the likelihood of there being a relationship visually.

Initially, we should decide on whether or not the net worth of a billionaire actually affects the number of children they have. For this, we would do a hypothesis test to see if it’s likely that there is a relationship that cannot be explained purely by chance.

\[ \newcommand{\indep}{\perp \!\!\! \perp} \newcommand{\nindep}{\not\!\perp\!\!\!\perp} H_0: net~worth \indep number~of~children\\ H_1: net~worth \nindep number~of~children \]

If it is determined that there is a “unexplainable” relationship between the two variables we will then perform a regression analysis and assess the correlation coefficient so as to see how related the variables are. If the correlation coefficient is high we will be able to then create a predictive model and test it’s accuracy.

Analysis #2

This question deals with how a billionaire’s education status affects their total networth. We can determine by looking at the NetWorth and Education column. We can replace all the blanks with 0 or drop na. Then we can create a line graph that shows the how the two variables are related to each other.

We can determine this by performing a hypothesis test. \[ H_0: education \indep net~worth\\ H_1: education \nindep net~worth\ \]

If it is determined that there is a “unexplainable” relationship between the two variables we will then perform a regression analysis and assess the correlation coefficient so as to see how related the variables are. If the correlation coefficient is high we will be able to then create a predictive model and test it’s accuracy.