This is bad data analysis, and "Bribery" is not the right conclusion here. It's more subtle than that.
I did my own data analysis. The article published the dataset, and I thought it was sloppy not to control for party affiliation. Also the average is TOTALLY the wrong statistic: the average is going to be dominated by a few politicians with massive contributions. The median, for instance, will capture a result that's more typical of the "bulk", rather than the outliers.
I mean, do your own data analysis and draw your own conclusions, but look. Here are the MEDIAN contributions by vote and affiliation, as far as I can see from 10 minutes with my spreadsheet:
D No: 10650
R No: 24300
D Yes: 11500
R Yes: 10500