Surveys come in many guises – public opinion polls, household panels, population censuses etc. The general purpose of a survey is to learn something about the distribution of certain attributes in a population, usually based on self-reports about these attributes collected for a sample of individuals drawn from the population. Such inferences are prone to various errors – coverage, sampling, nonresponse, adjustment, and measurement errors. Survey methodology, in turn, is the study of the sources of error in surveys and ways of improving the quality of survey data.
This course offers an introduction to survey methodology. Starting from an exposition of the total survey error paradigm as the dominant conceptual foundation of the field, we will discuss sampling strategies, coverage and nonresponse problems, and measurement issues from a data quality perspective. We will also look at current developments in survey practice such as online access panels or computerized measurement techniques.
The aim of this course is to equip students with the knowledge and skills necessary to critically evaluate and analyze existing surveys, and to design and conduct high-quality surveys on their own. The data quality issues we discuss are by no means limited to surveys. They are also inherent in many other data sources and data collection strategies used in the social sciences. Therefore, the methodological principles and tools introduced in this class are highly relevant and useful beyond the immediate survey context.
The course takes place on Thursdays 11.45-13.15 in D-432. For further information please go to the ILIAS site of the course (password: SM1718).