ABSTRACT
As interest in the study of non-governmental organizations (NGOs) has grown in recent years, the paucity of quality, generalizable data has become an increasingly clear barrier to research advancement. Many NGO scholars have created excellent primary data sources for their particular research questions (see for example Mitchell and Schmitz 2014 and the TNGO research project at Syracuse or Bush 2016 and optical character recognition (OCR) of NGO primary documents). Other scholars have made use of institutional datasets on NGOs from the Union of International Associations (e.g. Murdie and Davis 2012a; Smith and Wiest 2005) or technological tools to create data from the text of news articles (IDEAS as used by Murdie and Bhasin 2010) or links on the internet (IssueCrawler as used by Carpenter 2007). Each of these projects has made important contributions toward developing our understanding of NGOs by opening new empirical windows and testing new theories. But these datasets are also each bounded in time and by specific research questions, and thus it is very difficult to combine or reuse this data again. In order for NGO researchers to develop larger datasets that are easily integrated, we need to collaborate on a grander scale. Aggregated and shared data on NGOs will help to place the study of NGOs on similar empirical footing as democracy studies, conflict processes, electoral studies, or humanitarian development, each with their own high-profile datasets (e.g. Polity, Correlates of War, and V-Dem) that have enabled a related research community to flourish internationally and advance community research agendas.
