Volume 66, Issue 3 pp. 354-369
Full Access

The Effects of E-Government on Trust and Confidence in Government

Caroline J. Tolbert

Corresponding Author

Caroline J. Tolbert

Kent State University

Caroline J. Tolbert is an associate professor of political science at Kent State University and coauthor of Virtual Inequality: Beyond the Digital Divide (Georgetown University Press, 2003). Her research examines voting, elections and public opinion, direct democracy, technology and electoral politics, and race and ethnicity. She is the coauthor of Educated by Initiative: The Effects of Direct Democracy on Citizens and Political Organizations (2004) and coeditor of Citizens as Legislators: Direct Democracy in the United States (1998). E-mail:[email protected].

Karen Mossberger is an associate professor of public administration at the University of Illinois at Chicago and coauthor of Virtual Inequality: Beyond the Digital Divide (Georgetown University Press, 2003). Her research interests include information technology, urban public policy, economic development, and policy learning and diffusion. She has published previously in the Public Administration Review. E-mail:[email protected].

Search for more papers by this author
Karen Mossberger

Corresponding Author

Karen Mossberger

University of Illinois at Chicago

Caroline J. Tolbert is an associate professor of political science at Kent State University and coauthor of Virtual Inequality: Beyond the Digital Divide (Georgetown University Press, 2003). Her research examines voting, elections and public opinion, direct democracy, technology and electoral politics, and race and ethnicity. She is the coauthor of Educated by Initiative: The Effects of Direct Democracy on Citizens and Political Organizations (2004) and coeditor of Citizens as Legislators: Direct Democracy in the United States (1998). E-mail:[email protected].

Karen Mossberger is an associate professor of public administration at the University of Illinois at Chicago and coauthor of Virtual Inequality: Beyond the Digital Divide (Georgetown University Press, 2003). Her research interests include information technology, urban public policy, economic development, and policy learning and diffusion. She has published previously in the Public Administration Review. E-mail:[email protected].

Search for more papers by this author
First published: 24 May 2006
Citations: 662

Abstract

Trust in government has been declining for more than three decades now. E-government has been proposed as a way to increase citizen trust in government and improve citizen evaluations of government generally. Using two-stage models to analyze recent Pew survey data, this research explores the relationship between e-government use, attitudes about e-government, and trust in government. There is a statistically significant relationship between trust and use of a local government Web site, as well as other positive assessments of federal and local governments. The evidence suggests that e-government can increase process-based trust by improving interactions with citizens and perceptions of responsiveness. The findings are theoretically important for reconciling the conflicting research on the effects of e-government and for understanding variations by level of government. Citizen attitudes toward government, including trust, are core concerns for democratic governance and public administration.

Notes

  • 1 Very few demographic or attitudinal factors are statistically related to use of federal, state, or local e-government Web sites. The more educated are statistically more likely to have visited a federal government Web site; government workers are more likely to have visited a state government Web site; and African Americans are more likely to have visited a local government Web site. Beyond these limited factors, gender, age, income, race, ethnicity, partisanship, and frequency of use provide no explanatory power in predicting e-government use. Logistic regression models predicting e-government use based on demographic, economic, and attitudinal factors offer little explanatory power, with very low explained variance (pseudo-R2 range from .02 to .04).
  • 2 As with any two-stage model, we made some identification assumptions in the structural models. We hypothesized that demographic factors such as race, ethnicity, age, education, and income would affect citizen satisfaction with e-government. Frequent users of e-government should be more likely to perceive improved government transparency, accessibility, and responsiveness; the same should be true for government workers, who presumably use e-government more frequently. Partisanship may also shape perceptions of government processes. Because Republicans controlled the presidency and Congress at the time of the survey, we expect Republican partisans to have more favorable views of government. To simplify the calculation of predicted probabilities, reported in tables 5–7, they were based on first-stage Poisson regression models rather than ordered logistic regression models, as reported in tables 1–3. This resulted in one overall prediction (or value) per respondent rather than predicted values for low, moderate, and high evaluations of government transparency and efficiency, accessibility, and responsiveness.
  • 3 Estimations were produced using Clarify: Software for Interpreting and Presenting Statistical Results by Michael Tomz, Jason Wittenberg, and Gary King.
    • The full text of this article hosted at iucr.org is unavailable due to technical difficulties.