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).

Though the overreporting of voter turnout in the National Election Study (NES) is widely known, this article shows that the problem has become increasingly severe. The gap between NES and official estimates of presidential election turnout has more than doubled in a nearly linear fashion, from 11 points in 1952 to 24 points in 1996. This occurred because official voter turnout fell steadily from 1960 onward, while NES turnout did not. In contrast, the bias in House election turnout is always smaller and has increased only marginally. Using simple bivariate statistics, I find that worsening presidential turnout estimates are the result mostly of declining response rates rather than instrumentation, question wording changes, or other factors. As more peripheral voters have eluded interviewers in recent years, the sample became more saturated with self-reported voters, thus inflating reported turnout.

Testing theories about political change requires analysts to make assumptions about the memory of their time series. Applied analyses are often based on inferences that time series are integrated and cointegrated. Typically analyses rest on Dickey—Fuller pretests for unit roots and a test for cointegration based on the Engle—Granger two-step method. We argue that this approach is not a good one and use Monte Carlo analysis to show that these tests can lead analysts to conclude falsely that the data are cointegrated (or nearly cointegrated) when the data are near-integrated and not cointegrating. Further, analysts are likely to conclude falsely that the relationship is not cointegrated when it is. We show how inferences are highly sensitive to sample size and the signal-to-noise ratio in the data. We suggest three things. First, analysts should use the single equation error correction test for cointegrating relationships; second, caution is in order in all cases where near-integration is a reasonable alternative to unit roots; and third, analysts should drop the language of cointegration in many cases and adopt single-equation error correction models when the theory of error correction is relevant.

This note extends Melissa P. Collie's “Universalism and the Parties in the U.S. House of Representatives, 1921–80,” American Journal of Political Science 32, 4 (November 1988): 865–883. Detecting a strongly negative correlation between the time series of universalism and partisanship in roll call votes for the 67th through 96th U.S. Houses, Collie concluded that consensus and partisanship are alternative, rival means of organizing legislative activity. If robust, this finding ought not to be time- or chamber-specific: it should be in evidence over the whole (partisan) histories of both House and Senate, session by session. Moreover, the inverse relationship should persist under alternative operationalizations of both partisanship and universalism. Using several measures of partisanship and universalism, mostly based on roll call votes tabulated for sessions of Congress, we reassess this relationship for the 43rd through 105th Congresses. Collie's core finding persists for both chambers over the longer time span provided that one uses her measures. But results are weaker when sessions of Congress rather than Congresses are used as units of observation, and alternative operationalizations of partisanship and universalism do not strongly replicate the original finding.