Complexity of textual data in entrepreneurship and innovation research

Abstract

Innovation and Entrepreneurship are complex activities. They are also primarily language and relationship based. That is, it is largely through verbal communications (speech and text) that ideas are developed and business transacted. New methods are arising which are changing the way that we understand and can investigate innovation and entrepreneurship. Big Data Analytics allow researchers to uncover relationships and meaning in text documents, using a mix of quantitative and qualitative methods. This chapter shows that the complexity issues in innovation and entrepreneurship research with text comes from three sources. The first form of complexity is technical complexity. The second source of complexity is from language itself. The third source of complexity is in the concept itself. Each of these is discussed in detail. Complexity can either be addressed by simplifying the data or finding a mechanism for dealing with the complexity. A method of text data analytics using Corpus and Computational Linguistics deals with the complexity without eliminating data, allowing for a more nuanced investigation of innovation and entrepreneurship. The methodology is demonstrated by investigating how technological innovation and entrepreneurship are discussed in the United States Congress, using a corpus from 1981 to 2014.