New Data on Ideology and Money in Politics

by John Sides on July 25, 2013 · 3 comments

in Data

Adam Bonica writes:

I am pleased to announce the public release of the Database on Ideology, Money in Politics, and Elections (DIME). The database was initially developed as part of the project on Ideology in the Political Marketplace, which is an on-going effort to conduct a comprehensive mapping of the ideology of political elites, interest groups, and donors using the common-space CFscore scaling methodology. [JMS: For details, see here.). It includes records for over 100 million political contributions made by individuals and organizations to local, state, and federal elections spanning a period from 1979 to 2012. A corresponding database of candidates and committees provides additional information on state and federal elections. In addition, the database includes common-space ideal points for a comprehensive set of candidates for state and federal office, interest groups, and individual donors.

What this translates into:

The common-space CFscores allow for direct distance comparisons of the ideal points of a wide range of political actors from state and federal politics. In total, the database includes ideal point estimates for 51,572 candidates and 6,408 political committees as recipients and 13.7 million individuals and 1.3 million organizations as donors.

Here, “ideal points” means an estimate of ideology.  In essence, Bonica has developed an innovative way to measure the relative liberalism or conservatism of millions of political actors.  It’s a treasure trove, and I hope people will make use of it.  You can see some of the ways in which he’s used these data at his blog.

{ 3 comments }

Eli T. July 26, 2013 at 12:58 am

I hate raining on parades, but I seriously doubt that Bonica’s cf-scores are really measuring ideology.

True, the data is a treasure trove..,but it also lacks context. Fundraising is not ideological most of the time. In fact, as a former fundraising consultant who has worked for over a dozen MCsx.two consulting firms, and another dozen state and local campaigns, I would say 90%+ of fundraising is non-ideological.

I don’t want to give to much away as I am currently working on a paper that will likely be reviewed by commentors and/or contributors here and I am not at a top 10 university (yes, that clearly matters since nothing is really double-blind in the information age), but while Bonica’s AJPS is a well written article and his dataset is huge, as someone who has been in the trenches it is very,very unrealistic. The entire paper is based on the false assumption that ideology drives donations. While such an assumption may seem rational, that doesn’t make it correct.

Anonymous Coward July 26, 2013 at 11:18 am

I was going to jump down your throat, but the within-party correlations between cfscore and dw-nominate really aren’t great — sorry if I’m raining on *your* parade! Bonica cites values of .56 for House Democrats and .66 for House Republicans, those are for all years. Looked at session by session, the correlation for House Democrats drops from .63 in the 104th House to .33 in the 110th while those for H Republicans are pretty steady in the .58-.62 range.

Whatever dimension Bonica is uncovering has a clear partisan dimension, and draws an ideological component from that, but seems to be becoming less clearly ideological over time. At least for Democrats, anyway, which is just weird.

Adam Bonica July 27, 2013 at 4:24 am

Regarding the above comments – these are variants of the same critiques I’ve received about the model since I started on the project. I’d prefer not to give the impression that these issues were glossed over because the paper thoroughly addresses them. A substantial portion of the paper (and the bulk of the supplemental appendix) is devoted to addressing these issues and assessing measure validity.

In response to the comment about non-ideological contributions:

The importance of strategic giving in explaining the contribution behavior of individual donors is greatly overstated. When I control for the types of strategic considerations that are shown to be important determinants of corporate and trade PAC contributions, I find that they have almost no explanatory power for individual donors. In contrast, the ideological model fits the data very well.

In response to the comment about within party correlations:

I would caution against conflating DW-NOMINATE scores with the “true” measures of ideology. Roll call scaling is viewed as the gold standard for good reason. But in the end, they are measures of ideological voting, just as the CFscores are measures of ideological giving. Although comparisons with DW-NOMINATE are briefly used to establish face validity, the strength of the bivariate correlations can only tell you so much. The main evidence of face validity comes from an analysis that shows CFscores are able to predict voting outcomes nearly as well as roll call measures. Additional evidence that the model is doing more than just picking up partisanship is shown repeatedly throughout the paper.

Comments on this entry are closed.

Previous post:

Next post: