The study has an aim to analyze the multiple-choice questions in the final exam of Civic Education subject for the elementary school in Banjarnegara District. It applied a quantitative-descriptive approach. The sampling methods used here are random, stratified, and proportional sampling. From the sampling, it gave 10 schools, 4 core and 6 additional schools. The qualitative analysis applied a review and the quantitative one used the computer software of ITEMAN 3.00 and the manual method. The result of qualitative analysis shows the questions have agreed with the standard, but need some improvements of materials, construction, and language/culture in some parts. The quantitative analysis proves that difference impact of the questions is poor (3%), satisfactory (11%), and good (86%). The questions difficulty level is too difficult (8,5%), moderate (68,5%), and too easy (23%). The effectiveness of the distractor in the question is effective (57%), ineffective (43%). The reliability of the items in a whole reaches 0,839, meaning the questions have a high consistence. From those results, it can be concluded that the questions are matched to the standard, but they need some improvements of materials, construction and language/culture. The questions have a good difference impact, a moderate difficulty, and an effective distractor. In addition, the questions are also highly reliable. Using item validity calculation, it is known that there is one invalid question, i.e. the item 8. Keywords: an Item Analysis; ITEMAN; Difficulty Level and Discrimination Power