Methodology (from the cited article)

The Internet makes campaign finance more transparent than ever. With Federal Election Commission data online, anyone with an Internet connection can access donor information through sites such as www.tray.com. One can confine a search to individual donors, candidates, or, as was done in this survey, employers of donors. Since this information is online, every reader of this survey has the ability to check the validity of the data presented.

Reports from the Federal Election Commission display information for donors contributing $200 or more to federal campaigns. My methodology in researching donations from employees at U.S. News and World Report’s top twenty-five national universities consisted of counting only contributions to the two major campaigns and their auxiliaries containing the candidate’s last name in the title.

I’ll use Carnegie Mellon as an example of how I arrived at my findings. I listed the dollar amounts of each donation given to George W. Bush. The total came to $1,000 from three donations. I did the same for Carnegie Mellon donations by employees of that university to John Kerry. The total came to $29,325 from 55 donations (Since some donors gave multiple contributions, it is important not to use donors and donations interchangeably.). From this, I found that the split in the raw number of donations is 55 for Kerry and 3 for Bush and that the giving ratio is 29 to 1 in favor of Kerry. For simplicity’s sake, the giving ratios have been rounded to conform to a ratio of some variable to one.

This is an inexact science. FEC reports don’t list donations under $200. Not everyone fills out the reports honestly or completely. Searches based on key words “Bush” or “Kerry” omit some contributions that in effect go to the campaign. Searches based on name variations of top universities—e.g., “Penn,” “UPenn,” “U. of Pennsylvania,” “University of Pennsylvania”—may leave out some more obscure name variants for schools, and thus, leave out some information.

At one of U.S. News and World Report’s top-ranked schools, the University of California-Berkeley, an FEC search proved ineffective in determining ratios for employee giving. The number of schools within the University of California system made it impossible to know exactly which school employed individual donors in certain cases. Thus, the data included here focuses on the entire University of California system, rather than just its Berkeley campus.

This is a survey of the giving patterns of employees at major universities. University employees include not only professors and administrators, but also librarians, janitors, coaches, and anyone else employed by the school. Although students who donate to campaigns often provide the FEC with the names of the educational institutions they’re affiliated with, we’ve excluded donors identified as students unless they provided information demonstrating that their college or university also employed them, e.g., as a teaching assistant.

In addition to students, hospitals associated with universities pose problems. Does one omit or include, for instance, the data from employees of Cornell University’s hospital in New York City—approximately 200 miles from Cornell’s campus in Ithaca, New York? I omitted it, and did so in similar cases when institutions—such as hospitals and clinics—bearing a university’s name were not housed on-site, or had only a tangential relationship with the school. While employees of major hospitals and health plans bearing a university’s name were excluded from the survey, the faculty of medical, law, and business schools affiliated with the main university were included.

The differences in methodology, and the limits posed by computer searches, ensures that researchers inspecting the same data will compute slightly different dollar amounts coming from each school to each candidate. It is, however, unlikely that anyone looking at the same reports would come up with ratios that differ in any substantial way from the ratios that I’ve computed.