Lines at the polls raise the cost of voting and can precipitate unequal treatment of voters. Research on voting lines is nonetheless hampered by a fundamental measurement problem: little is known about the distribution of time voters spend in line prior to casting ballots. We argue that early, in-person voter check-in times allow us identify individuals who waited in line to vote. Drawing on election administrative records from two General Elections in Florida—1,031,179 check-ins from 2012 and 1,846,845 from 2016—we find that minority voters incurred disproportionately long wait times in 2012 and that in-person voters who waited excessively in 2012 had a slightly lower probability—approximately one percent—of turning out to vote in 2016, ceteris paribus. These individuals also had slightly lower turnout probabilities in the 2014 Midterm Election, ceteris paribus. Our results draw attention to the ongoing importance of the administrative features of elections that influence the cost of voting and ultimately the extent to which voters are treated equally.

Good education requires student experiences that deliver lessons about practice as well as theory and that encourage students to work for the public good—especially in the operation of democratic institutions (Dewey 1923; Dewy 1938). We report on an evaluation of the pedagogical value of a research project involving 23 colleges and universities across the country. Faculty trained and supervised students who observed polling places in the 2016 General Election. Our findings indicate that this was a valuable learning experience in both the short and long terms. Students found their experiences to be valuable and reported learning generally and specifically related to course material. Postelection, they also felt more knowledgeable about election science topics, voting behavior, and research methods. Students reported interest in participating in similar research in the future, would recommend other students to do so, and expressed interest in more learning and research about the topics central to their experience. Our results suggest that participants appreciated the importance of elections and their study. Collectively, the participating students are engaged and efficacious—essential qualities of citizens in a democracy.

Seasonal variations in sodium concentrations have been measured on surface-pit snow samples and on firn and ice core samples from the Ross Ice Shelf, Antarctica. Site locations include J-9 (82° 22’ S., 168° 40’ W.), Roosevelt Island dome (79° 22’ S, 161° 40’ W), C-7-1 (78° 30’ S., 177° 00’ W.), and C-7-3 (78° 20’S., 179° 51’ E.). The predominant source for the Na is sea salt, indicated by greater concentration levels at seaward sites. Al concentrations of the order of only 10–9 g/g show that the input of continental dust is comparable to that at inland Antarctic locations, and that dust contributes only a negligibly small fraction of the Na on the shelf. Maximum Na concentrations occur in the winter or early spring, as is the case for Greenland ice. The annual accumulation-rate at J-9, determined by counting Na concentration peaks with depth, is 90 kg m–2 year–1, in agreement with rates determined radiometrically. Annual cycles in Na concentration are also detectable at depth in the J-9 ice core. It is suggested that Na concentration is a useful diagnostic criterion for distinguishing between East Antarctic ice (10–8 g Na/g), West Antarctic ice (30 × 10–9 g Na/g), and ice that fell as snow on the shelf itself (> 30 × 10–9g Na/g). The transition between snow that is chemically characteristic of the ice-shelf regime to snow of an inland regime is expected to occur near the 500 m elevation contour. This position is up to 200 km inland of the grounding line. A model is presented for the large-scale decrease in Na concentration with distance inland within the ice-shelf regime. Since deeper ice in Ross Ice Shelf cores originated generally further from the ocean, the up-stream origin of shelf ice may be estimated from the chemical profile. The Little America V ice-core chemistry profile shows no discontinuity as would be expected if a recent surge of West Antarctic ice had occurred.

An empirical model of firn densification from the surface to the zone of pore close-off has been constructed. Fundamental rate equations have been derived for the first two stages of densification. In the first stage, for densities less than 0.55 Mg m−3, the densification rate is proportional to the mean annual accumulation times the term (ρi − ρ), where ρ is the density of the snow and ρi is the density of pure ice. The densification rate in the second stage, where 0.55 Mg m −3 < ρ < 0.8 Mg m−3, is proportional to the square root of the accumulation rate and to (ρi− ρ). Depth–density and depth–age calculations from this model are compared with observation. Model accumulation rates are within about 20% of values obtained by other techniques. It is suggested that depth intervals of constant density in some Antarctic cores may represent a synchronous event in the 1880 ’s when ten times the normal accumulation fell within a year or two.

The practice of using point estimates produced by the King ecological inference technique as dependent variables in second-stage linear regressions leads to second-stage results that, in general, are inconsistent. This conclusion holds even when all assumptions behind King's ecological technique are satisfied. Second-stage inconsistency is a consequence of the fact that King-based point estimates of disaggregated quantities contain errors correlated with the true quantities the estimates measure. Our findings on second-stage inconsistency, as well as a fix that we propose, follow from econometric theory in conjunction with an analysis of simulated and real ecological data sets.

We address in Herron and Shotts (2003; hereinafter HS) an increasingly common statistical practice called EI-R, in which point estimates generated by the King (1997) ecological inference technique are used as dependent variables in second-stage linear regressions. Although HS argue that EI-R slope estimates are inconsistent and suffer from attenuation bias, this characterization of EI-R is excessively conservative. Indeed, we show here that EI-R estimates can suffer from sign reversals, attenuation bias, and augmentation bias, that inference based on these estimates can be misleading, and that EI-R is unfixable and should not be used.

Research designs that call for the estimation of time trends in legislator ideal points must allow for changes in underlying policy spaces. Simply put, a set of legislator ideal points from one such space cannot necessarily be compared with a set of ideal points from another. Rothenberg and Sanders's (2000, American Journal of Political Science 44:316–325) analysis of shirking in Congress does not internalize this point insofar as it assumes that legislator ideal points from adjoining congressional sessions lie in a common space. While it may be the case that members of Congress shirk ideologically when approaching retirement, the regression estimates at the heart of Rothenberg and Sanders neither support this possibility nor constitute evidence against it.

Since Herron and Shotts (2003a; hereinafter HS), Adolph and King (2003; hereinafter AK), and Herron and Shotts (2003b; hereinafter HS2), the four of us have iterated many more times, learned a great deal, and arrived at a consensus on this issue. This paper describes our joint recommendations for how to run second-stage ecological regressions, and provides detailed analyses to back up our claims.

Many political science research articles that use limited dependent variable models report estimated quantities, in particular, fitted probabilities, predicted probabilities, and functions of such probabilities, without indicating that such estimates are subject to uncertainty. This practice, along with the reporting of “percentage correctly predicted,” can overstate the precision of reported results. In light of this, the present article describes a variety of measures of uncertainty that authors can include alongside estimates generated by limited dependent variable models. It also proposes an alternative to “percentage correctly predicted” and illustrates its calculations with congressional cosponsorship data from Krehbiel (1995).

While it is very common for Congressional researchers to use interest group ratings as measures of legislator policy preferences, this paper argues that the manner in which such ratings are calculated implies that they may poorly approximate the underlying legislator preferences on which they are based. In light of this, the paper develops a technique designed to adjust interest group ratings so that they more closely correlate with legislator preferences. It argues based on Monte Carlo simulations that the technique produces adjusted ratings that improve on unadjusted ratings, and it applies the adjustment technique to historical ratings published by the Americans for Democratic Action.

This article uses spatial voting theory to analyze the properties of linear regressions that employ interest group ratings as measures of legislator policy preferences. Such regressions, in general, yield inconsistent results. In particular, least-squares estimation of a bivariate regression which contains an interest group rating as a regressor produces an inflated slope estimate. Instrumenting for the rating with a second rating, as proposed by Brunell et al. (1999), does not fix this problem, and this is because errors in both sets of ratings are correlated. Finally, estimation of a trivariate regression that contains an interest group rating and a party indicator on its right-hand side yields inconsistent slope estimates and, in particular, a party coefficient estimate of unreliable sign. Hence, regressions including both ratings and party indicators are not useful tools in the debate on whether party affiliation has an independent impact on legislator behavior.

In mid-2011, the Florida state legislature passed House Bill 1355 (HB 1355) and in so doing placed new regulations on community organizations that historically have helped eligible Floridians register to vote. Among the legal changes promulgated by this bill were new regulations on the operations of groups like the League of Women Voters and a new oath, warning of prison time and fines, that voter registration agents were required to sign. Such changes raised the implicit costs that eligible Florida citizens faced when registering to vote, and we show that voter registrations across the state in the second half of 2011 dropped precipitously compared with registrations in the second half of 2007. This pattern is evident among registrants in general, among registrants age 20 and younger, and among individuals who registered as Democrats. Outside of HB 1355, we know of no credible explanations for these results. Our findings thus show how restrictions on the way that third-party organizations register voters can have tangible effects on actual registrations. Given that registration prior to an election is a civic necessity in Florida and in many other states, such restrictions have the potential to affect electoral outcomes as well.

We consider the relationship between the preferences of American voters and the preferences of the U.S. legislators who represent them. Using an Internet-based, national opinion survey in conjunction with legislator voting records from the 109th and 110th Congresses, we show that members of Congress are more extreme than their constituents, i.e., that there is a lack of congruence between American voters and members of Congress. We also show that when a congressional legislator is replaced by a new member of the opposite party, one relative extremist is replaced by an opposing extremist. We call this leapfrog representation, a form of representation that leaves moderates with a dearth of representation in Congress. We see evidence of leapfrog representation in states and House districts and in the aggregate as well: the median member of the 109th House was too conservative compared to the median American voter, yet the median of the 110th House was too liberal. Thus, the median American voter was leapfrogged when the 109th House transitioned to the 110th. Although turnover between the 109th and 110th Senates occurred at approximately the same rate as between the 109th and 110th Houses, the Senate appears to be a more moderate institution whose median member does not move as abruptly as that of the House.

We show that the butterfly ballot used in Palm Beach County, Florida, in the 2000 presidential election caused more than 2,000 Democratic voters to vote by mistake for Reform candidate Pat Buchanan, a number larger than George W. Bush's certified margin of victory in Florida. We use multiple methods and several kinds of data to rule out alternative explanations for the votes Buchanan received in Palm Beach County. Among 3,053 U.S. counties where Buchanan was on the ballot, Palm Beach County has the most anomalous excess of votes for him. In Palm Beach County, Buchanan's proportion of the vote on election-day ballots is four times larger than his proportion on absentee (nonbutterfly) ballots, but Buchanan's proportion does not differ significantly between election-day and absentee ballots in any other Florida county. Unlike other Reform candidates in Palm Beach County, Buchanan tended to receive election-day votes in Democratic precincts and from individuals who voted for the Democratic U.S. Senate candidate. Robust estimation of overdispersed binomial regression models undernins much of the analysis.

On television's Law and Order, the police catch criminals and hand them over to the lawyers to get convictions. Part of the program's dramatic tension comes from the police operating under the scrutiny of a rigid and unforgiving legal system. The suspense increases as the lawyers try to do their job even though there is often a gap between justice and what the law requires.