The Effect of Outliers on Regression Analysis: Regime Type and Foreign Direct Investment

Seung-Whan Choi

Quarterly Journal of Political Science, 2009, vol. 4, issue 2, 153-165

Abstract: The presence of outliers and influential cases can dramatically change the magnitude of regression coefficients and even the direction of coefficient signs (i.e., from positive to negative or vice versa). When researchers ignore abnormal observations, especially with respect to dependent variables, their empirical results can be misleading. Unfortunately, the fact remains underappreciated in studies of political science. To expound upon the outlier issue, I reexamine an empirical study that reports on two opposing effects of democratic institutions on inflows of foreign direct investment (FDI). In doing so, I illustrate the way influential outliers can drastically affect the substantive results of regression analysis. After properly reanalyzing outlying countries, I conclude that democratic countries attract more FDI than authoritarian countries.

Date: 2009
References: Add references at CitEc
Citations: View citations in EconPapers (9)

Downloads: (external link)
http://dx.doi.org/10.1561/100.00008021 (application/xml)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:now:jlqjps:100.00008021

Access Statistics for this article

More articles in Quarterly Journal of Political Science from now publishers
Bibliographic data for series maintained by Lucy Wiseman (lucy.wiseman@nowpublishers.com).