Christopher Z. Mooney

Pulling cost-effectiveness analysis up by its bootstraps: A non-parametric approach to confidence interval estimation

The statistic of interest in the economic evaluation of health care interventions is the incremental cost effectiveness ratio (ICER), which is defined as the difference in cost between two treatment interventions over the difference in their effect. Where patient-specific data on costs and health outcomes are available, it is natural to attempt to quantify uncertainty in the estimated ICER using confidence intervals. Recent articles have focused on parametric methods for constructing confidence intervals. In this paper, we describe the construction of non-parametric bootstrap confidence intervals. The advantage of such intervals is that they do not depend on parametric assumptions of the sampling distribution of the ICER. We present a detailed description of the non-parametric bootstrap applied to data from a clinical trial, in order to demonstrate the strengths and weaknesses of the approach. By examining the bootstrap confidence limits successively as the number of bootstrap replications increases, we conclude that percentile bootstrap confidence interval methods provide a promising approach to estimating the uncertainty of ICER point estimates. However, successive bootstrap estimates of bias and standard error suggests that these may be unstable; accordingly, we strongly recommend a cautious interpretation of such estimates.

Modeling Regional Effects on State Policy Diffusion

Generations of state politics scholars have believed that a U.S. state is more likely to adopt a law if its neighboring states have already done so, that is, that there is a positive regional effect on policy diffusion. But rarely has the social learning process, the theoretical underpinning of this effect, been examined critically Furthermore, the statistical models used to assess this effect have been fundamentally flawed. In this article, I consider more fully the potential impacts of social learning on policy diffusion, and develop a simple approach to modeling these impacts more flexibly. Using this approach, I take a fresh look at data from two classic studies of state policy diffusion and demonstrate that the regional effect is more complex than previously believed.

Constructing confidence intervals for cost‐effectiveness ratios: an evaluation of parametric and non‐parametric techniques using Monte Carlo simulation

The statistic of interest in most health economic evaluations is the incremental cost-effectiveness ratio. Since the variance of a ratio estimator is intractable, the health economics literature has suggested a number of alternative approaches to estimating confidence intervals for the cost-effectiveness ratio. In this paper, Monte Carlo simulation techniques are employed to address the question of which of the proposed methods is most appropriate. By repeatedly sampling from a known distribution and applying the different methods of confidence interval estimation, it is possible to calculate the coverage properties of each method to see if these correspond to the chosen confidence level. As the results of a single Monte Carlo experiment would be valid only for that particular set of circumstances, a series of experiments was conducted in order to examine the performance of the different methods under a variety of conditions relating to the sample size, the coefficient of variation of the numerator and denominator of the ratio, and the covariance between costs and effects in the underlying data. Response surface analysis was used to analyse the results and substantial differences between the different methods of confidence interval estimation were identified. The methods, both parametric and non-parametric, which assume a normal sampling distribution performed poorly, as did the approach based on simply combining the separate intervals on costs and effects. The choice of method for confidence interval estimation can lead to large differences in the estimated confidence limits for cost-effectiveness ratios. The importance of such differences is an empirical question and will depend to a large extent on the role of hypothesis testing in economic appraisal. However, where it is suspected that the sampling distribution is skewed, normal approximation methods produce particularly poor results and should be avoided.

Citizens, Structures, and Sister States: Influences on State Legislative Professionalism

U.S. state legislatures vary substantially in the degree to which they are professional institutions. Drawing on the state policy literature, this study identifies three general categories of potential influence on this variation: the characteristics of a states population, its governmental structures, and the level of legislative professionalism in its peer group states. Using four measures of professionalism from different periods, the author tests hypotheses about several such variables in cross-sectional and panel analyses. He finds that each of these categories of influence has significant, independent effects on legislative professionalism, and that taken together these influences account for most of the observed variation in professionalism across the states.

Morality Policy Reinvention: State Death Penalties

How does morality policy change as it diffuses? Social learning theory holds that later adopters learn from earlier adoptions to modify, or reinvent, a policy to fit their needs better. But because of its technical simplicity, saliency, and conflicts of basic values, morality policy may not be amenable to policy learning. We develop and test three reinvention hypotheses reflecting distinct roles for learning. Our analysis of U.S. state death penalty policy supports each hypothesis but under different political conditions. We conclude that, when possible, policymakers make morality policy in their usual way, by incremental steps and learning from previous adoptions. But when basic moral conflicts surface, considerations other than policy learning drive reinvention.

Term Limits as a Boon to Legislative Scholarship: A Review

After an exceptionally swift reform movement in the 1990s, 15 states are now adjusting to the most significant institutional change to their governments in a generation—legislative term limits. Beyond the basic task of identifying and cataloguing their various substantive impacts, term limits have presented scholars with an exceptional opportunity to test legislative theory for two reasons. First, most legislative theory is based on behavioral assumptions that term limits appear to affect systematically, such as the reelection motivation. Second, the distribution of term limits across the states gives scholars exceptional methodological advantages for testing theory. In this article, I review the scholarly literature on the impacts of state legislative term limits, describe their theoretically relevant implications and methodological advantages, and advocate using the reform to develop an important research agenda.

State Politics and Policy Quarterly and the Study of State Politics: The Editor's Introduction

The American states provide, arguably, the world’s most advantageous venue in which to test general propositions about political behavior and policymaking (Jewell 1982; Brace and Jewett 1995). One would be hardpressed to design a better system of polities for conducting political research. The states represent fifty units of analysis with broadly similar political structures, cultures, and populations, but with significant variation across a range of social, policy, and institutional characteristics important to many political theories. Unlike the examination of single units of government, such as the U.S. federal government, the variation in these characteristics allows scholars to test hypotheses of cause-and-effect relationships of political behavior and policymaking in a valid way. But unlike the comparison of different countries, the comparison of states does not produce so much variation that it overwhelms scholars’ ability to identify relationships that might exist. Just as the states are a natural laboratory for public policy (New State Ice Co. v. Liebmann 1932), they are also a natural laboratory for testing hypotheses derived from political theory. Furthermore, the accessibility of state government officials makes quite feasible many research strategies that would be unthinkable in the study of federal officials and institutions. Beyond these clear advantages to the empirical study of political phenomena, the states are also of central and increasing importance in the U.S. political system. While the past forty years have seen an unprecedented resurgence in the states’ political power, policy responsibility, and institutional capacity (Bowman and Kearney 1986; Hedge 1998; Weber and Brace 1999), the states have been critical to politics and policymaking in the U.S. from the

Explaining Legislative Leadership Influence Simple Collective Action or Conditional Explanations

Why are the leaders in some U.S.-style legislatures more influential than others? This study uses individual-level data on lawmakers’ perceptions of their leaders’ influence to test three general theories of legislative power delegation: legislative leaders have no real power, simple collective action theory, and Conditional Party Government theory. These perceptions of speakers’ legislative influence are modeled with varying intercept, multilevel, ordered probit models. The analyses strongly support the simple collective action problem explanation of legislative leadership influence, in particular suggesting that collective problems caused by the internal dynamics of the legislative process drive the delegation of influence to leaders.

Putting It on Paper The Content of Written Information Used in State Lawmaking

What kinds of information do state legislators consider in their legislative deliberations? This article examines four dimensions of the content of the written information state representatives in Indiana, Massachusetts, and Oregon used in 1989: whether it was policy or political information, one-sided or multisided, in agreement or disagreement with the position of the legislator using it, and whether it had any hard or soft scientific content. Legislators are found to use information heavily dosed with political preferences, and they tend to look only at one side of an issue - the one with which they agree. However, they also use a substantial amount of scientific information.

Measuring State House Speakers’ Formal Powers, 1981–2010

The cross-sectional variation among state legislative leaders’ formal powers provides a unique opportunity to test a variety of hypotheses of the American legislative process that involve the causes and effects of these powers. But to date, measures of these powers have been developed primarily for single years or biennia, limiting their applicability, especially for drawing causal conclusions. In this note, I develop an index of state House speakers’ formal powers for each biennia from 1981 to 2010 (for even-year-election states) and from 1982 to 2009 (for odd-year-election states).