Author(s): Davide BARBIERI
Publication name: Romanian Intelligence Studies Review
Publisher name: Mihai Viteazul National Intelligence Academy
Publication type: Journal article
Publication date: December 31, 2016
Pagination:
Issue/ Volume: 16/2016
DOI:
Abstract
Traditional data analysis and intelligence analysis are strongly focused on
testing hypotheses, which can be either corroborated or rejected by means of collected
observations. Still, we must acknowledge that many hypotheses are just assumptions or
prejudices, which are unlikely to be supported by evidence. Researchers may therefore
select the observations which comply with their ideas, in an attempt to prove they were
right in making assumptions. To avoid this pitfall, analysts should not neglect the
preliminary steps of exploratory data analysis, by means of which new – and possibly
more robust – hypotheses are explored, diminishing the constraints imposed on the
analysis by the investigators’ creativity. Several descriptive and graphical techniques can
be employed by intelligence analysts in order to succeed in this endeavor. This paper will
give a summary account of some of them.
Keywords: hypotheses finding, data exploration, descriptive techniques.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 2.0 Generic License.