The value of publicly available data sets for social science research and evaluation

Lincoln J. Fry

This paper is concerned with large publically accessible data files and provides guidance as to how to access these available large data sets which social scientists can use for planning, research, as well as assessment of large social/governmental programs. Two types of big data are identified, one generated by the internet and the other by large scale survey research projects. The author’s experience is with the latter, and this paper will highlight work done through one of what are known as barometers, in this instance, the Afrobarometer Project. The purpose of this paper is to promote the use of large data by social scientists which have been neglected for those purposes. The objectives of this paper will be to assist social scientists to locate data sources, data warehouses or special projects like the various barometers, understand how to acquire data sets of interest, learn how to download them, and then how to transfer these data into a format that is consistent with the researcher’s available software programs, particularly those programs that can accommodate large data files. Examples of the types of study that can be produced are provided, In addition, the author gathered information about the experiences of both big data users, including those who are embedded in what can be called the big data industry, and software providers. The input of all of these professionals greatly contributed to the completion of this paper.

Download PDF: