Revisiting the Evidence for a Cardinal Treatment of Ordinal Variables

27 Pages Posted: 15 Jul 2015

See all articles by Carsten Schröder

Carsten Schröder

Free University of Berlin (FUB) - Department of Business and Economics

Shlomo Yitzhaki

Hebrew University of Jerusalem - Department of Economics; National Bureau of Economic Research (NBER)

Date Written: July 2015

Abstract

Well‐being (i.e., satisfaction, happiness) is a latent variable, impossible to observe directly. Hence, questionnaires ask people to grade their well‐being in different life domains. The most common practice - comparing well‐being by means of descriptive analysis or linear regressions - ignores that the underlying collected well‐being information is ordinal. If the well‐being function is ordinal, then monotonic transformations are allowed. We demonstrate that treating ordinal data by methods intended to be used for cardinal data may give an incorrect impression of a robust result. Particularly, we derive the conditions under which the use of cardinal method to an ordinal variable gives an illusionary sense of robustness, while in fact one can reverse the conclusion reached by using an alternative cardinal assumption. The paper provides empirical applications.

Suggested Citation

Schröder, Carsten and Yitzhaki, Shlomo, Revisiting the Evidence for a Cardinal Treatment of Ordinal Variables (July 2015). SOEPpaper No. 772, Available at SSRN: https://ssrn.com/abstract=2630181 or http://dx.doi.org/10.2139/ssrn.2630181

Carsten Schröder (Contact Author)

Free University of Berlin (FUB) - Department of Business and Economics ( email )

Boltzmannstrasse 20
D-14195 Berlin, 14195
Germany
+49 030 838-52259 (Phone)
+49 030 838-52560 (Fax)

Shlomo Yitzhaki

Hebrew University of Jerusalem - Department of Economics ( email )

Mount Scopus
Jerusalem, 91905
Israel
+972 2 659 2201 (Phone)
+972 2 652 2319 (Fax)

National Bureau of Economic Research (NBER) ( email )

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
60
Abstract Views
502
Rank
640,387
PlumX Metrics