Negative racial stereotypes routinely work together with myths about American equality to undermine public support for policies that would reduce inequality. But what happens when white Americans are confronted with information about structural racial inequality, which contradicts the myth of equal opportunity? Evidence from an original survey experiment conducted among approximately 4,000 white Americans demonstrates that emphasizing the systemic origins of racial inequalities in the COVID-19 pandemic makes respondents more accepting of policies aimed at reducing racial inequalities in a variety of domains. Qualitative insights from post-treatment reflections further show that facing the reality of structural inequality disrupts blame-based narratives and generates support for policies meant to confront inequality. The findings suggest that discussing structural inequality can disrupt individualistic understandings and increase approval for policies that promote equality across multiple domains; they also illuminate why opponents of equality see discussions of structural inequality as so threatening.

Unlike most other countries, in the United States, subnational governments (states) have substantial authority over collective bargaining and union organization laws. Because states compete for business investment and union (dis)organization likely has spillover effects beyond state borders, weak unions in one state may affect union organization in other states. We examine how union decline in one state is associated with union decline in neighboring states, and whether the presence of prounion (left-leaning) governments may limit the spread of union decline. Examining a period of major union decline (1983-2014), we find that union weakness in one state is associated with union weakness in nearby states. We observe that Democratic power in Congress is associated with higher unionization rates, but that liberal state governments have been relatively powerless to stop union decline in this period. These findings have important implications for understanding the historical and contemporary weakness of American unions and for the future of union strength in the United States.

The statistical distributions of cosmogenic nuclide measurements from moraine boulders contain previously unused information on moraine ages, and they help determine whether moraine degradation or inheritance is more important on individual moraines. Here, we present a method for extracting this information by fitting geomorphic process models to observed exposure ages from single moraines. We also apply this method to 94 10Be apparent exposure ages from 11 moraines reported in four published studies. Our models represent 10Be accumulation in boulders that are exhumed over time by slope processes (moraine degradation), and the delivery of boulders with preexisting 10Be inventories to moraines (inheritance). For now, we neglect boulder erosion and snow cover, which are likely second-order processes. Given a highly scattered data set, we establish which model yields the better fit to the data, and estimate the age of the moraine from the better model fit. The process represented by the better-fitting model is probably responsible for most of the scatter among the apparent ages. Our methods should help resolve controversies in exposure dating; we reexamine the conclusions from two published studies based on our model fits.

Can researchers draw consistent inferences about the U.S. public's issue attitudes when studying survey results from both the in-person and telephone interview modes of the 2000 National Election Studies (NES) survey? We address this question through an analysis contrasting the distribution of issue attitudes across modes in the dual sample design of the 2000 NES. We find clear differences across mode even when applying a method devised by the NES to improve comparability by recoding issue attitude scales from the in-person mode. We present an alternative method of recoding these scales, which substantially improves comparability between modes. Through an analysis of the covariance structure of the issues and simple models of vote choice, we discuss the implications of the results for the study of issue attitudes in the 2000 NES.

A lagged dependent variable in an OLS regression is often used as a means of capturing dynamic effects in political processes and as a method for ridding the model of autocorrelation. But recent work contends that the lagged dependent variable specification is too problematic for use in most situations. More specifically, if residual autocorrelation is present, the lagged dependent variable causes the coefficients for explanatory variables to be biased downward. We use a Monte Carlo analysis to assess empirically how much bias is present when a lagged dependent variable is used under a wide variety of circumstances. In our analysis, we compare the performance of the lagged dependent variable model to several other time series models. We show that while the lagged dependent variable is inappropriate in some circumstances, it remains an appropriate model for the dynamic theories often tested by applied analysts. From the analysis, we develop several practical suggestions on when and how to use lagged dependent variables on the right-hand side of a model.

The mass franchise led to more responsive government and a more equitable distribution of resources in the United States and other democracies. Recently in America, however, voter participation has been low and increasingly biased toward the wealthy. We investigate whether this electoral “class bias” shapes government ideology, the substance of economic policy, and distributional outcomes, thereby shedding light on both the old question of whether who votes matters and the newer question of how politics has contributed to growing income inequality. Because both lower and upper income groups try to use their resources to mobilize their supporters and demobilize their opponents, we argue that variation in class bias in turnout is a good indicator of the balance of power between upper and lower income groups. And because lower income voters favor more liberal governments and economic policies we expect that less class bias will be associated with these outcomes and a more equal income distribution. Our analysis of data from the U.S. states confirms that class bias matters for these outcomes.

Research shows that when the more liberal Democratic Party controls the national government, unemployment is lower, but whether liberal state governments are associated with lower unemployment has not been examined. We argue that more left-leaning governments in the U.S. states have the same preference for and willingness to use government to reduce unemployment, but that the greater resource and policymaking constraints that the states face during economic downturns limit their ability to shape unemployment to economic growth periods. We find evidence for these arguments in an analysis of the U.S. states for the period of 1975–2010. Specifically, when economic growth is low, liberal state governments are associated with increases in unemployment rates similar to or even somewhat higher than conservative governments, but when growth is moderate to high, liberal state governments are associated with greater-than-expected reductions in unemployment. We also provide some evidence that different state spending decisions between liberal and conservative state governments may explain these patterns.