13th European Conference on eGovernment – ECEG 2013 1 | Page 527

Virgil Stoica and Andrei Ilas
page is filled with some information to make it more attractive(“ contents”), but without a concern for e‐ services(“ services”), for the feedback that citizens could provide(“ digital democracy”) or for protecting the data on citizens gathered while they are using the webpage(“ security”).
Figure 3: The five dimensions of e‐Government score, for rural and urban areas
In order to find out if the villages, the small, the medium and big cities are scoring differently as groups, we have applied an ANOVA test. P‐value was almost zero which indicated that there are differences between the average scores obtained by each group.
Table 3: Tukey Kramer outputs Tukey Kramer Multiple Comparisons
Group
Sample Mean
Sample Size
Rural
11.51788
198
Very
small
cities
14.17525
101
Small
cities
16.60186
59
Medium
cities
20.895
22
Large
cities
28.70323
31
Other Data
Level of significance 0.05 Numerator d. f. 5 Denominator d. f. 406
Absolute Difference
Std. Error of Difference
Critical Range
Comparison
Group 1 to
Group 2
2.6573687
0.456614702
1.7625
Group 1 to Group 3
5.0839856
0.553879816
2.138
Group 1 to
Group 4
9.3771212
0.839218373
3.2394
Group 1 to
Group 5
17.185347
0.721293266
2.7842
Group 2 to
Group 3
2.4266169
0.611900597
2.3619
Group 2 to
Group 4
6.7197525
0.878593745
3.3914
Group 2 to
Group 5
14.527978
0.766748633
2.9596
Group 3 to
Group 4
4.2931356
0.932851989
3.6008
Group 3 to
Group 5
12.101361
0.828365457
3.1975
Group 4 to
Group 5
7.8082258
1.041006224
4.0183
Results Means are different Means are different
Means are different Means are different Means are different Means are different Means are different Means are different Means are different Means are different
MSW 27.88978 Q Statistic 3.86
Furthermore, a Tukey Kramer test confirmed that these differences are significant statistics. These results suggest that size matters when it comes to e‐Government performance in Romania.
505