Publication Type: Paper/Poster Proposal
Abstract: Two methods, the Rizzo, Brick and Park (2004) and the last-birthday, were used to randomly select respondents from within households telephoned in two statewide RDD studies. The Rizzo et al. method employs both probability and quasi-probability selection techniques; the last-birthday method, quasi-probability selection. Although the last-birthday method is relatively unobtrusive and simple to administer, numerous studies have shown that the last-birthday method is about 75 to 91 percent effective in selecting respondents randomly from within households (Lavrakas et al., 1993; Lavrakas et al., 2000; O’Rourke and Blair, 1983). The Rizzo, Brick and Park (2004) method for selecting respondents within households is less obtrusive for the approximately 85 percent of households with fewer than three adult members, but uses the last birthday method to select among more than two adults and the Kish method if informants don’t know birthdays. In this comparison, interviewers were randomly assigned to each selection method and were seated in different parts of the telephone laboratory. About 278 surveys were completed using each method. Results of the first survey show that although surveys conducted via the last-birthday method had shorter average duration than those of the Rizzo et al. method, the differences are not statistically significant. The Rizzo et al. method used fewer calls per complete. Using AAPOR formulae, the response rates of the last birthday (23.9%) were slightly higher than those for the Rizzo et al. (21.4%) method. The two methods resulted in approximately the same ratios of men and women, 37 to 63 percent, compared with expected ratios of 45 to 55 percent. The average ages, levels of education and income of respondents in each group are comparable. Number of adults in households is similar. Responses on the substantive questions in the survey did not vary by within-household selection method.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: In the 2002 general election a field experiment was conducted in San Francisco. 5% of the call list for a GOTV mobilization drive was segregated into a control group; the rest were called as planned. Following the election, a comparison of the treatment and control groups revealed that the GOTV calls had virtually no effect on voter turnout. However, not everyone in the treatment group actually received the treatment – i.e., answering the telephone and listening to the message. Looking at only successfully completed calls, there was an apparent large effect, because these people voted at much higher rates than either the control group or members of the treatment group who did not receive the call. This is, of course, a selection effect – it is why we randomize. People who are available and willing to receive a GOTV call are different from people who are either not available or not willing. Indeed, there are even differences among those people who did not receive the call. I have data on the disposition of each call. While people who were successfully contacted voted at the highest rates, this is not the only distinction. The unwilling people vote at the next highest rate – people who answered the phone but hung up on the caller. Other distinctions exist as well. Among people who apparently were not home, people who had answering machines voted at higher rates than people who did not. This would be unsurprising if the callers left messages, but in fact they did not. There was also a very large distinction between wrong numbers and disconnected lines. This paper provides these data and explains some of the underlying theoretical reasons for these distinctions. It explains what the selection effect is, why it exists, and how experimental controls deal with it. It also discusses the methodological reasoning for analyzing data from both a random and non-random perspective to maximally leverage the information at one’s disposal.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: The establishment of the September 11 Victim Compensation Fund raises the question of when the government should compensate those suffering losses in terrorist attacks and other disasters. We address that issue in this paper, offering standards that may be used to distinguish among events. We also apply these standards to a number of real world cases.

Publication Type: Proceeding

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: With very few exceptions, quantitative studies of world politics have assumed that the major actors and actions are independent actors. Game theoretic models are legion, but rarely deal with more than two actors at a time. Some beginning attempts to model the interdependency in international relations have appeared in the geopolitical literature, but as yet network models have yet be widely applied in scholarly work on international politics. This is somewhat ironic, since it is evident at first blush that international politics is about the interdependence that appears around the world. Network models have been widely used to represent relational information among interacting units. Most work focuses on random graph models in which nodes represent individual actors and edges identify specific relations among them. What is clear is that we need to model the interconnections among the relevant actors in world politics. We develop probabilistic models of links among actors based on an unobserved ``social space'' and show how such an approach inherently captures various dependencies among actors. In particular, we detail the sender effects, receiver effects, the dyadic effects, as well as other actor and dyad level covariates and show how they may be estimated for complicated networks of individuals, organizations, and nation-states. We apply this in the context of extensive event data on the political interactions in Central Asia over the period from 1989-1999. Prior to the development of this approach, quantitative studies in world politics have been forced to assume the independence of the objects of study, or to find ways to attenuate statistical information to ``correct for'' temporal, spatial, and actor dependencies. The development of the latent space model couched entirely in a regression framework coupled with the ability to estimate it via simulation strategies allows the analyst to comfortably focus on the dependencies as well as the other important covariates without such draconian assumptions