Over the last decades, there have been a lot of studies about impact of information and communication technology (ICT) investment on productivity and efficiency of organizations. Most of the previous studies have mainly dealt with the problems of commercial sectors, but there have been few studies on ICT influence on the efficiency of non-commercial sectors such as public institutions. Therefore, this study aims to investigate the relationship between ICT investment and efficiency in public institutions. We selected the ward offices, local government providers, in Seoul, Incheon, Daegu, Gwangju and Daejeon at Korea as the research subjects. This research is composed of three parts. First, the efficiency of each ward office is measured by using data envelopment analysis (DEA). Second, main factors explaining characteristics of ICT in each ward office are extracted from ICT related data by using factor analysis. The state of ICT in each ward office is examined based on these factors. Third, after grouping the ward offices based on factors yielded in the previous stage, analysis of variance (ANOVA) was conducted to test difference of efficiency among groups. In addition, regression analysis is used to analyze the relationship between ICT and efficiency. The major findings of this study can be summarized as follows. Two factors explaining features of ICT in ward offices were extracted and these factors were labeled as infrastructure-level of ICT and usage-level of ICT. Based on the two factorized dimensions, ward offices were classified into four groups: Usage-oriented Group, Informatization Group, Non-informatization Group, and Infrastructure-oriented Group. The ANOVA results show that there are efficiency difference in three paired groups: Usage-oriented Group vs. Infrastructure-oriented Group, Informatization Group vs. Infrastructure-oriented Group, and Non-informatization Group vs. Infrastructure-oriented Group. It was founded that efficiency was directly affected by the input and output and also indirectly affected by infrastructurelevel of ICT. Statistical method of factor analysis, t-test, ANOVA, post hoc comparison, and regression analysis were utilized for analysis of data.