Michaelene Cox
employees of both branches. It is noteworthy that the e‐Ethics training scores are overall lower in states with the lowest e‐Government scores, and that three of the five top performers in e‐Government provide online ethics training to individuals in both state executive and legislative branches.
Table 1: U. S. corruption rates and risk for corruption by state
State |
# Convictions |
State |
Corruption risk |
|
Most |
|
Highest ranking |
Texas |
88 |
Georgia |
50 |
Maryland |
58 |
South Dakota |
49 |
Virginia |
57 |
Wyoming |
48 |
California |
52 |
Virginia |
47 |
New York |
52 |
Maine |
46 |
|
Least |
|
Lowest ranking |
Nebraska |
2 |
Nebraska |
5 |
North Dakota |
2 |
California |
4 |
Utah |
2 |
Washington |
3 |
Connecticut |
0 |
Connecticut |
2 |
New Hampshire |
0 |
New Jersey |
1 |
Table 2: U. S. e‐Government and ethics training scores by state
State |
e‐Gov score |
e‐Ethics score |
|
Highest score |
|
Delaware |
83.7 |
2 |
Georgia |
78.3 |
4 |
Florida |
77.9 |
4 |
California |
70.9 |
2 |
Massachusetts |
69.5 |
4 |
|
Lowest score |
|
Hawaii |
35.8 |
2 |
Wyoming |
35.7 |
0 |
Maryland |
32.9 |
4 |
New Mexico |
32.5 |
0 |
Mississippi |
31.1 |
1 |
Bivariate correlation coefficients( Pearson’ s r) can determine the relationship between various measures of state corruption levels and on‐line ethics training, and indicate the magnitude and direction of the association. The closer the correlation is to either + 1 or ‐1, the stronger the correlation. Table 3 reports coefficients and level of statistical significance for two‐tailed tests conducted on relevant variables for the 50 U. S. states. Not reported here are the statistically significant coefficient correlations between mandatory training and corruption variables. Mandatory training is positively associated with number of convictions, and negatively associated with corruption risk. The table does illustrate a significant relationship between the number of federal public corruption convictions last year and both state e‐Government scores and availability of ethics training online to government officials and employees in the executive and legislative branches. The relationship is positive, meaning that greater numbers of convictions are associated with greater e‐ Government capabilities and online ethics training. This is not the expected direction. The hypothesis generated by supporters of e‐Government is that greater G2E should lower corruption rates. As a reminder, however, while bivariate correlation coefficients presume a linear relationship, we cannot extract a causal statement from the tests. There may be a myriad of explanations for the positive association. For instance, state ethics commissions may develop online training as a response to preexisting high levels of corruption, and without time yet to cultivate a more ethical environment through newly developed training programs, corruption rates remain unaffected. Or, we might posit that ethics training increases employee awareness and willingness to report illicit behavior committed by other government employees, and thus leads to more arrests and convictions. On the other hand, there is a statistical and negative relationship between e‐ Government and online ethics training with corruption risk. That is, greater e‐capability is associated with less corruption risk. This latter correlation is expected.
The results of these preliminary analyses indicate several avenues for future research. Time‐series analyses will be helpful for discerning causal relationships between G2E capability and public corruption.
148