William G. Jacoby

Loess:: a nonparametric, graphical tool for depicting relationships between variables☆

Abstract Loess is a powerful but simple strategy for fitting smooth curves to empirical data. The term “loess” is an acronym for “local regression” and the entire procedure is a fairly direct generalization of traditional least-squares methods for data analysis. Loess is nonparametric in the sense that the fitting technique does not require an a priori specification of the relationship between the dependent and independent variables. Although it is used most frequently as a scatterplot smoother, loess can be generalized very easily to multivariate data; there are also inferential procedures for confidence intervals and other statistical tests. For all of these reasons, loess is a useful tool for data exploration and analysis in the social sciences. And, loess should be particularly helpful in the field of elections and voting behavior because theories often lead to expectations of nonlinear empirical relationships even though prior substantive considerations provide very little guidance about precise functional forms.

Variability in State Policy Priorities: An Empirical Analysis

This article examines variability in policy priorities across the American states; that is, the ways that state governments allocate resources to meet societal needs. Specifically, our analysis uses 1992 data on state program expenditures to produce a comprehensive geometric representation- or model-of state policy priorities for that year. This model is parsimonious, powerful, and substantively meaningful. The structure of state policy priorities is manifested as a sharp contrast between programs that deliver particularized benefits and those that supply collective goods. Furthermore, we show that policy priorities are largely determined by public opinion and interest group activity within the respective states. Therefore, our analysis not only operationalizes successfully a critical aspect of the policy process; it also makes a useful contribution to the study of state politics.

Data theory and dimensional analysis

Introduction Measurement Data Theory Dimensionality Data Theory and Scaling Methods Alternating Least Squares, Optimal Scaling Conclusions

Levels of Conceptualization and Reliance on the Liberal-Conservative Continuum

This article examines the degree to which people at different levels of conceptualization use the liberal-conservative continuum to organize their thinking about political candidates. The analysis uses weighted multidimensional scaling, and the results show that 1980 candidate evaluations are based upon two evaluative dimensions. However, the levels do vary systematically: The higher levels--ideologues and near ideologues--rely more heavily on the liberal-conservative dimension and also display a more systematic structure in their candidate perceptions than the other levels. These empirical results are fully consistent with the substantive theory underlying the levels of conceptualization.

Is There a Culture War? Conflicting Value Structures in American Public Opinion

This article examines the “culture war” hypothesis by focusing on American citizens’ choices among a set of core values. A geometric model is developed to represent differences in the ways that individuals rank-order seven important values: freedom, equality, economic security, social order, morality, individualism, and patriotism. The model is fitted to data on value choices from the 2006 Cooperative Congressional Election Study. The empirical results show that there is an enormous amount of heterogeneity among individual value choices; the model estimates contradict any notion that there is a consensus on fundamental principles within the mass public. Further, the differences break down along political lines, providing strong evidence that there is a culture war generating fundamental divisions within twenty-first century American society.

The sources of liberal-conservative thinking: Education and conceptualization

In this study, I examine two sources of liberal-conservative thinking within the American electorate: Education and level of conceptualization. This analysis differs from previous work in at least two important ways. First, I test the impacts of education and conceptualization simultaneously. Second, I focus on the degree to which people actively use the liberal-conservative continuum to organize their perceptions of the parties and candidates. Empirical analysis of data from the 1984 CPS National Election Study confirms that education and conceptualization have roughly equivalent effects on ideological structuring in political perceptions. I discuss the implications of this finding for the study of mass belief systems and information processing within the American public.

The Structure of Bureaucratic Decisions in the American States

Bureaucrats have a significant impact on the ways that governmental policies impinge on American citizens. This is particularly true at the state level. Administrators in state agencies have wide latitude to make vital decisions in important public programs. And yet, surprisingly little is known about this process. The present analysis will address this topic by focusing on the structure of bureaucratic initiatives in state Medicaid policies. We believe that doing so has several important advantages. Substantively, our focus on the structure inherent in bureaucratic decisions reveals a great deal about the nature of administrative policy making in social program developments at the state level. Methodologically our analysis produces an empirical measure of bureaucratic policy outputs, which can be used as an analytic variable in other research efforts. Altogether the information obtained in this analysis provides new insights about the role of administrators in the American policy process. Background In this study, we will examine state adoptions of Medicaid optional health care services. The federal government and state legislators establish the general parameters for the Medicaid program (Coughlin, Ku, and Holahan, 1994). But, state-level bureaucrats make important decisions that determine the exact configuration of services to be offered within each state (Bovbjerg and Holahan, 1982; Congressional Research Service, 1993; Schneider and Jacoby, 1996). Hence, many of the differences that exist across state Medicaid programs are due to administrative initiatives rather than to state legislative statutes or federal regulations (Davison, 1980). The objective of this article is to determine whether there is an underlying structure among these interstate differences. If such a structure does exist, then it should be useful for discerning the criteria that Medicaid bureaucrats have in mind when they make administrative decisions. Given the importance of state bureaucracies in the Medicaid policy making process, it is somewhat surprising that so little is known about how they operate. There has been virtually no research conducted on this topic so we do not know whether there is any pattern or structure to bureaucratic decisions. Previous empirical analyses have focused on other aspects of the Medicaid program, such as state expenditures and assessments of program intention, content, and/or scope (Hanson, 1984; Holahan and Cohen, 1986; Barrilleaux and Miller, 1988; Schneider, 1988; Buchanan, Cappelleri, and Ohsfeldt, 1991). Those few studies that have focused directly on administrative initiatives in the Medicaid program have tended to be speculative and interpretive (Lemov, 1991; Hovrath, 1992). They generally conclude that there is no discernible pattern among state bureaucratic policy adoptions. Instead, they argue that the range of available services is too wide to be analyzed or summarized in any comprehensible form (Davison, 1980; McDonough, 1992). We emphatically disagree with the preceding conclusion and believe that there is an underlying structure to bureaucratic decisions in state Medicaid programs. We hypothesize a pattern of bureaucratic activity based upon the degree of difficulty involved in providing various health care services. State administrators would begin by adopting easy options; only after doing so would they move on to implement more difficult services. The exact nature of the difficulty is an empirical matter, to be determined as pa-rt of the analysis. It could be based upon the costs of the services, the size of the clientele groups for the various services, or the level of political controversy surrounding different health care options. In a-nv case, a cumulative pattern should arise because states adopt easier services before proceeding to more difficult options. Most states are willing to provide a set of minimal health care services while others are willing to bear greater burdens. …

A Culture of Dependence? The Relationship Between Public Assistance and Public Opinion

In the United States it is widely believed that public assistance may have harmful effects on the social and political orientations of those who receive it. Certain kinds of government support – particularly welfare – may foster a ‘culture of dependence’ comprising values and beliefs that are different from, and perhaps contrary to, the predominant American political culture. We examine the relationship between government assistance and public opinion using survey data taken from the 1992 CPS National Election Study. Our empirical results show that welfare benefits do have some effect on issue attitudes. But recipients of public aid are virtually identical to non-recipients in terms of their core values, reactions to the political system and general beliefs about American society. There is no evidence that a distinctive ‘culture of dependence’ has developed among people who rely on financial support from the federal government.

Is It Really Ambivalence

Ambivalence within citizens’ political orientations is currently an important topic for scholars interested in American public opinion. The very existence of the volume within which this chapter appears is telling evidence of that fact. The usual conception of ambivalence stresses that individuals feel pulled in conflicting directions by the beliefs that enter into the development of their own attitudes. Empirically, ambivalence is manifested by the expression of contradictory opinions. However, I believe that a note of caution would be useful. We should not be too eager and always assume that contradictory opinions signal the presence of attitudinal ambivalence. There are, of course, many other potential sources for conflicting feelings about government, politics, and policy.

Comment: The Dimensionality of Public Attitudes toward Government Spending

In “The Two Faces of Government Spending,” Paul Goren applies a confirmatory factor model to data from the Center for Political Studies (CPS) National Election Studies (NES) to show that public attitudes toward policy spending are two dimensional. Here, I show that a unidimensional model, derived from nonparametric item response theory, can be used to represent the same data. This latter model provides additional insights about specific spending responses and also produces results that are equivalent to the two-dimensional model with respect to the impact of racial stereotypes. Therefore, by scientific standards, the unidimensional representation of public opinion toward government spending seems to be preferable.