I further ran a Bonacich Centrality test on these corporations to find which corporations are connected to those who also have the most connections. In other words, being connected to neighbors with more connections makes one powerful. This would further indicate that the power elite theory is important, as these corporations would have the most information flowing amongst them. After running the test, I found that Kraft Foods Inc had the highest Bonacich score (-2.127). Following this, General Electric (-2.104), Coca Cola Co. (-1.957), Home Depot (-1.907), and Goldman Sachs (-1.845). This first seems surprising that these corporations have the largest Bonacich centrality, but take a look back at the diagram above. These corporations’ ties
…show more content…
However, these corporations who have joint contributions do not all have interlocking directors, in the same industry, or have headquarters in the same state. So, I decided to run a MR-QAP analysis on this, with Joint Contributions as the dependent variable and Industry, State, and Interlocking Directors as the independent variables. MR-QAP analysis is used to predict one relation, knowing the other. In this case, I’m attempting to predict political contributions, knowing the other attributes and this analysis will determine if those independent variables are good factors to use to predict whether a corporation will donate to a particular candidate. After running this analysis, I found that the P-value of the same-industry variable was the highest, at 0.006, and the director interlock P-value was 0.10995. Although the director interlock value was larger and therefore would typically be irrelevant, for a sample size this small, it becomes a relevant factor in this network. This confirms that industry is a good predictor to determine which corporations donate to similar candidates, but that network factors, such as director interlocks, do play a significant role in the process. However, I found that the R squared value for this MR-QAP analysis was very small (0.0078), meaning that there are many other variables which affect political contributions. However, as I solely focused on industry, headquarter locations, and director interlocks, I was not able to determine which other factors would come into
However, these corporations who have joint contributions do not all have interlocking directors, in the same industry, or have headquarters in the same state. So, I decided to run a MR-QAP analysis on this, with Joint Contributions as the dependent variable and Industry, State, and Interlocking Directors as the independent variables. MR-QAP analysis is used to predict one relation, knowing the other. In this case, I’m attempting to predict political contributions, knowing the other attributes and this analysis will determine if those independent variables are good factors to use to predict whether a corporation will donate to a particular candidate. After running this analysis, I found that the P-value of the same-industry variable was the highest, at 0.006, and the director interlock P-value was 0.10995. Although the director interlock value was larger and therefore would typically be irrelevant, for a sample size this small, it becomes a relevant factor in this network. This confirms that industry is a good predictor to determine which corporations donate to similar candidates, but that network factors, such as director interlocks, do play a significant role in the process. However, I found that the R squared value for this MR-QAP analysis was very small (0.0078), meaning that there are many other variables which affect political contributions. However, as I solely focused on industry, headquarter locations, and director interlocks, I was not able to determine which other factors would come into