Michaelene Cox
of simple correlation analyses to address the first research question, this study takes a qualitative approach to evaluate G2E ethics training programs. It draws primarily upon content analysis of sample training programs, federal and state legislation, and an interview. The theoretical framework within which my argument falls is consistent with the principle‐agent model to explain the contractual relation between government and the public, and rational choice theory to understand how principals and agents make decisions in terms of economic gains and tradeoffs.
3. Political corruption and G2E ethics training
This paper explores correlations between U. S. state‐level corruption and facets of e‐Government using bivariate analyses to suggest direction and strength of the relationships. Simple correlations cannot examine causal relationships, but can nevertheless demonstrate if there are grounds for further investigation of e‐ Government as an anti‐corruption strategy. While multivariate regressions suggest direction and strength of the relationships, current data limitations prevent examining causality at this moment. Among other things, tests for causality require data analyzed over a period of time, a shortcoming in assessment now since most states implementing G2E training have only done so recently.
Analyses here include all 50 states in the U. S. and draw upon the most recently available data. I consider two measures of political corruption: number of convictions per state by the DOJ and state scores of corruption risk produced by the Center for Public Integrity. Measures of e‐Government include e‐Government scores per state provided by the think tank Brookings Institute( West 2008), and e‐ethics training reported by the National Conference of State Legislatures( NCSL) Ethics Commission( NCSL 2011). The NCSL is a bipartisan organization that reports on state activities such as available ethics training for state executive and legislative officials and employees. I assign each state branch with a 0 if it offers no ethics training for its elected officials or employees, 1 if it offers classroom or paper ethics training; and 2 if it provides on‐line ethics training. For an aggregate score to measure availability of e‐ethics training in both branches, I tally those values. Finally, a measure for mandatory ethics training is constructed by assigning the state a 0 if it is not available or is optional, and 1 if ethics training is required by statute and / or rule.
Before assessing any links between corruption and ethics training, it is interesting to briefly consider the extent of those phenomena among states. There were a total of 930 federal public corruption convictions in the U. S. in 2011, according to the DOJ. Table 1 reports the five states with the most and least convictions for political malfeasance that year and five states with the highest and lowest rankings for corruption risk. As noted above, this latter ranking does not indicate actual corruption levels but reflects lack of an ethics commission and / or other institutions deemed effective for monitoring and curbing political corruption, inadequate public access to information, and few other provisions to foster transparency and probity in public administration. In short, a high corruption risk score suggests a fertile environment in which corruption might flourish, and may raise a red flag for more careful scrutiny in the future. The NCSL notes that currently nine states do not have an ethics board or commission: Arizona, Idaho, New Hampshire, New Mexico, North Dakota, South Dakota, Vermont, Virginia, and Wyoming. As seen in Table 1, a third of those states are judged by the Center for Public Integrity to provide the greatest risk for corruption. Although we can take into account changes in conviction rates from year to year that reflect a flurry of related and high‐profile prosecutions in various states, overall patterns remain fairly consistent over the past decade or so. This in itself suggests that recent developments in e‐ Government and e‐ethics training programs in some states have not yet dramatically affected federal public conviction rates. Time series analyses are required and might feasibly be conducted within a few years.
Table 2 provides a snapshot of e‐Government and ethics training programs, listing states with the most and least capabilities in these matters. The Brookings Institute conducts analyses of nearly 1,500 state government websites to construct an overall state e‐Government index score based on features such as publications, audio materials, user fees, disability access, privacy policies and the like, as well as the number of online executable services. Although the zero‐to‐100 point e‐Government index does not specifically take into account online ethical training programs, it serves to illustrate general e‐Government capabilities across state executive, legislative and judicial branches. For a closer view at states scoring the highest and lowest e‐Government scores, Table 2 also reports the e‐Ethics training score assigned to states, with the construction of this measure previously described as a combination of executive and legislative training formats. A score of 0 indicates no training, 1 indicates that in‐person training in one of the branches is offered, 2 indicates on‐line training offered in one branch, and 4 indicates that on‐line ethics training is provided to elected officials and
147