Many different types of computer user interfaces are used to perform office tasks. Although some guidelines are proposed for selecting user interfaces appropriate to perform differnet tasks, they have not been supported by empirical studies. This study examined five user interface styles and their combinations which were used to perform representative office tasks. A variety of analysis techniques such as brainstorming, analytic hierarchy process, prototyping, and expert opinions were employed to evaluate the usability of the interfaces. Also, a human factors experiment was conducted to validate empirically the analysis results. The analysis procedures and results are described along with guidelines for selecting user interfaces in terms of subjective preferences and performance measures.