Conflicted Conservatives By Christopher Ellis factor analysis supp

Results of Factor Analysis

The table below displays the results of our factor analysis of the thirty-five policy preference questions in the 2004 American National Election Studies (NES). To be sure, there are more sophisticated ways to approach an understanding of the dimensionality of issue conflict. The purpose of this exercise is to combine statistical findings with intuition about the connections among various types of issues to get a general sense of the structure of issue conflict in American politics. Nevertheless, as our discussion of these results demonstrates, the findings are straightforward: They suggest the existence of two coherent dimensions along which the public arranges issues.

Issue

Dimension 1

Dimension 2

Right to government-provided health care

.43

.19

Gun control

.30

.17

Death Penalty

.23

.07

Government-mandated minimum standard of living

.50

.15

Spending on crime

.28

-.10

Partial privatization of Social Security

.23

.17

Spending on Social Security

.54

-.08

Spending on public schools

.56

.17

Spending on foreign aid

.22

.06

Spending on child care

.63

.10

Spending on aid to the poor

.61

-.04

Government’s role in improving the conditions of Hispanics

.34

.10

Government’s role in improving the conditions of blacks

.40

.12

Spending on welfare

.46

.08

Tradeoff: reduce government spending to cut taxes

.29

.21

Tradeoff: increase budget deficit to increase spending

.50

-.15

Tradeoff: reduce government spending to decrease the deficit

.40

.12

Tradeoff: increase taxes to increase government spending

.34

.13

Tradeoff: reduce government spending, or provide more services

.61

.16

Circumstances under which abortion should be legal

.01

.64

Government funding of abortion

.13

.50

Late-term abortion

.10

.37

Laws against discrimination against homosexuals

.12

.45

Adoption of children by homosexuals

.08

.63

Gay Marriage

.07

.68

Homosexuals in the military

.08

.46

*Spending on science and technology

.10

.14

*Spending on improving highways

.05

-.03

*Tradeoff: protecting the environment or preserving jobs

.17

.34

*Spending on terrorism

.03

.16

*Spending on illegal immigration

.04

.19

*School choice

.03

.18

*Spending on defense

.19

.22

*Tradeoff: should budget deficit be increased to cut taxes?

.04

.18

*Tradeoff: should taxes be increased to reduce budget deficit?

-.22

.31

Eigenvalues

5.32

2.35

Note: Entries are unstandardized principal factor loadings (varimax rotation).

*These issues either cross issue dimensions or load neither dimension. The fact that preferences on most of these issues are not structured coherently is not surprising, given that many (spending on highways, for example) are of relatively low salience in elite discourse or (in the case of school choice or terrorism spending) are issues where elite party positions on these issues were not yet clearly defined, and cleavages were “cross-cutting” across partisan, economic, and social lines.