In lieu of an abstract, here is a brief excerpt of the content:

Appendix 2.4: Data Source for Political Action Committees The data on contributions to 1998 state electoral campaigns were provided by the National Institute on Money in State Politics.1 Several caveats about these data must be noted. First, though the National Institute at times refers to the organizations it lists as political action committees (PACs), this name is not really appropriate in the sense in which it is used in the data on national PACs. That is, not all states legally define PACs or, even when they do, define them in the same manner. Indeed, the entities in the National Institute’s database include legally defined PACs, other groups that probably are PACs (e.g., the Alabama Dental Association), and individual businesses. Although, for simple convenience, we continue to refer to these entities as PACs, they should more appropriately be interpreted as “nonindividual, nonparty” contributors to political campaigns, as the National Institute more formally calls them. Second, the raw data generously provided by the National Institute through special data requests still required considerable cleaning before they were usable. That is, the state lists included large numbers of individual contributors and duplication of contributors. Recoding to eliminate these cases reduced the initial list of 222,592 PACs to 162,352 PACs. Thus, our experience should serve as a cautionary tale to researchers who are using the National Institute’s data without further refinement. Third, we removed party leadership PACs from our data set on the grounds that theoretically they are not interest groups, the subject of our study. And fourth, 1998 data were not available for eight states, usually due to their electoral calendars. In these cases, we used the next available election year, which was either 1999 or 2000.2 Nonetheless, as a result of all of these adjustments, we believe that our data set contains the best data on state health PACs in existence. And it matches up at the organizational level with our lobbying data set using 1998 lobby registrations; thus we capture both forms of organized interest activity for a single year. Entities on both lists (whether PACs or lobby registrations) were individually identified as having health interests using the coding rules employed by 188 | APPENDIX 2.4 Lowery and Gray (2007). The health PACs and lobby registrants were further coded by several subtypes of health interests using the same coding rules. Finally, the cleansed list of PACs was matched with the lobby registration data at the individual organization level to identify whether an organization was registered to lobby, contributed to political campaigns, or both. For the latter two categories, we also measured the size of the financial contribution the organization made to political campaigns for legislative offices, gubernatorial offices, all statewide offices, and judicial candidates. Notes 1. National Institute on Money in State Politics, “Political Giving Database,” www.followthemoney.org/index.phtml. 2.The exceptions are as follows: Arkansas (2000), Delaware (2000), Mississippi (1999), Nebraska (2000), New Jersey (1999), Oklahoma (2000), South Dakota (2000), and Virginia (1999). ...

Share