Method

A quantitative content analysis of press articles is used in this paper to analyze and describe the historical development of clean technology industries. This method has been used in different contexts and disciplines lately (Migendt et al., 2014; Phillips et al., 2004; Ventresca & Mohr, 2002; Wuthnow, 1989). The approach used allows for a deep, but still unbiased assessment of large quantities of data. The quality of the information depends not only on the textual sources, but also on the process of the software guided analysis (Berelson, 1952; Elo & Kyngas, 2007; R. P. Weber, 1990). A lesser quantitative approach then in this paper has been used to assess the system functions in innovation systems in the German biomass digestion industry (Negro & Hekkert, 2008). This paper uses WordStat 6.0 by Provalis Research to analyze the article database. Wordstat is a commercial software package used for content analysis and text mining in unstructured textual documents. WordStat analyzes texts and structures results according to dictionaries. These dictionaries consist of words grouped in thematic clusters, which are then processed through the software. The software is able to recognize the occurrence of single words or word clusters as well as relations (Krippendorff, 2012; Neuendorf, 2002). To describe the development of the different industries two separate, but still related dictionaries were developed. The main structure of the word clusters was equal, just the technology frame differs between the sectors (see dictionaries in the appendix). The dictionary was built according to the actors finance, innovation, and policy, with additional sub-actors in entrepreneurship and venture capital. Industry specific technology themes have been added as additional actors, as well. The words and word groups creating the dictionaries were developed in a deductive way. The deductive approach, also called top-down or directed approach is based on existing frameworks or theory to derive categories and included topics. An already existing reference is used to analyze textual information. In this paper the dictionaries include words and word groups based on the actors and actors’ synonyms as well as key topics associated with the main actors. For example, the sub-actor “ENTREPRENEURSHIP” consists of words based on the root word “ENTREPR*” (the * signals a flexible ending of a word) and the words “FOUNDER”, “STARTUP” and “START-UP” Additionally to the actors and subactors, the technology actor consists in the solar energy sector dictionary of solar energy technology, photovoltaic and thermal. Respectively, in the fuel cell sector dictionary consists of the different basic materials for fuel cells: alkaline, molten carbonate, phosphoric acid, proton exchange membrane, solid oxide and zinc air (Elo & Kyngas, 2007; Hsieh & Shannon, 2005; R. P. Weber, 1990). The approach used can best be described as summative content analysis. The goal is to explore usage of words and interpret it with help of further information. In the course of the analysis the occurrence of these actors and sub-actors can be tracked. In the following paragraph, the relative importance of the actors is analyzed over time and compared to the others. Mirroring the press articles on the technologies with the corresponding investment data enriches the results (Boykoff, 2008; Hsieh & Shannon, 2005).