The problem of imperfect knowledge became crucial issue for many applications including economic applications. Rough set theory, proposed by Z. Pawlak (1982) is one of the existing approaches to understanding and manipulation of imperfect knowledge. This theory has attracted many researchers and practitioners all over the world, who contributed essentially to its numerous developments and applications. The aim of this chapter is to present preliminaries of rough set theory (including information systems and decision tables, indiscernibility, lower and upper approximations of sets, reducts, patterns) and application of rough set methods to the construction industry data analysis with respect to the economic conditions. We present some methods based on rough set approach for solution of the following problems related to the analysis of such data: - Extracting dependencies between queries in questionnaires. - Searching for relationships between respondent answers and classification. - Extraction of (minimal) relevant subsets of queries predicting with high quality answers for other questions. Characterization of respondent situation changes in the consecutive year quarters. We discuss the results of Computer experiments including also visual representation of the received results.