Research projects
LES consists of two interlinked research projects titled Language Effects in Surveys and Studying Opinions and Populations in Online Text Data, which span across the disciplines of political science, computational linguistics and computer science. Both projects draw on recent advances in natural language processing and use online text data – sometimes referred to as Big Data – to address the changing landscape of comparative survey research.
Language Effects in Surveys - Project Summary
How do language and culture affect the ways in which people's understanding of concepts that are central to the social sciences from a comparative perspective? Over the past decades, researchers have wrestled with the challenge of establishing equivalence in survey items measuring attitudes to political concepts across culturally and linguistically varied nations. The research project Language Effects in Surveys sets out to develop and apply novel methods to study the impact of translational discrepancies in country comparative surveys.
The project leverages on recent developments in distributional semantics – a field within natural language processing – that enable quantification of similarities and differences in conceptual structures across and within languages. More specifically, the project aims at increasing our understanding of the relationship between survey translations, language use and conceptual similarities, which in turn will generate new knowledge about cross-national and cross-cultural variations in concepts employed in widely used survey items.
Studying Opinions and Populations in Online Text Data - Project Summary
The application of Big Data technology in various fields of humanities and social sciences in recent years has provided scholars with innovative ways and means to address the changing landscape of public opinion and political behavior. Although computer science has made, and continues to make, great progress both in terms of data management and in the development of analytical tools for handling vast amounts of data, problems of selection, measurement errors and other types of biases are still unsolved.
The research project Studying Opinions and Populations in Online Text Data seeks to validate the use of online text data as a complement to traditional surveys and polls. More specifically, we seek to develop computational methods for answering the following critical questions for scholars using online text data in conjunction with comparative survey data: 1) Is this text data relevant for my purposes?; 2) Is this text data reliable?; and 3) Is this text data representative of a population? The project contributes with important means for using online text data as a complement to traditional surveys and measures of public opinion as well as political attitudes and behaviours.