Record Details

Replication data for: There's Gotta be a Better Way! Crowdsourcing the Measurement of Observational Data

Harvard Dataverse (Africa Rice Center, Bioversity International, CCAFS, CIAT, IFPRI, IRRI and WorldFish)

View Archive Info
 
 
Field Value
 
Title Replication data for: There's Gotta be a Better Way! Crowdsourcing the Measurement of Observational Data
 
Identifier https://doi.org/10.7910/DVN/28600
 
Creator D'Orazio, Vito
Kenwick, Michael
Lane, Matthew
Palmer, Glenn
Reitter, David
 
Publisher Harvard Dataverse
 
Description Much of the data used in Political Science is extracted from news reports. This is typically accomplished using a resource-intensive method in which expert coders read news stories and quantify the relevant information. As a result of this method's inefficiencies, many databases are confined in terms of time and space or updated infrequently. Such situations restrict the data's utility, and place limits on the questions researchers may address. This paper introduces a method for analyzing news documents that combines crowdsourcing and computational approaches to produce reliable data efficiently. The new method is tested on documents about Militarized Interstate Disputes, and its accuracy ranges between about 68 and 76 percent. This is a considerable improvement over the accuracy of automated coding, another resource-efficient method, which ranges between about 30 and 45 percent. These results show crowdsourcing to be an accurate and efficient method for collecting observational data from news reports.