Coherence and Comparability as Criteria of Quality Assessment in Business Statistics

The problems of coherence and comparability exceed the classical notion of analysis of survey errors, because they do not concern single surveys or variables but the question of how results of two or more surveys can be used together and how relevant data can effectively be compared to obtain a better picture of social and economic phenomena over various aspects, e.g. space or time. This paper discusses characteristics of the main concepts of coherence and comparability as well as a description of differences and similarities between these two notions. Types of coherence and various aspects of perception of these notions in business statistics are analysed. Main sources of lack of coherence and comparability, factors affecting them (e.g. methodology, time, region, etc.) and methods of their measurement in context of information obtained from businesses will be also presented.