An error grave of consequence. Empirical models to study propensities to vote for parties
Nathalie Giger1,
Simon Hug2
Département de science politique et relations internationales
Université de Genève
Paper prepared for presentation at the General Conference of the
European Political Science Association (Brussels June, 2016)
First version January 2016, this version Jan 20, 2016
Abstract
Vote propensities for political parties have become an important
component of many voter surveys. They are supposed to give more
accurate and broad-ranging assessment of voters' positions with
respect to the parties they may choose in an election. While this more
detailed information on voters is certainly an advantage, many
scholars analyze these data with empirical models that lead to biased
estimates of the quantities of interest. We discuss the most commonly
used models and demonstrate the conditions under which biased
estimates result. We also propose an alternative way to analyze vote
propensities (based on a SURE-model) which, under most assumptions,
does not lead to biased estimates. We illustrate the performance of
this model in empirical applications, where we compare it with the
more commonly used models.
Footnotes:
1 D ' e p a r t e m e n t d e s c i e n c e p o l i t i q u e e t
r e l a t i o n s i n t e r n a t i o n a l e s , F a c u l t ' e d e s s c i e n c e s d e l a s o c i ' e t ' e ; U n i v e r s i t ' e d e G e n ` e v e ; 4 0 , B d d u
P o n t d ' A r v e ; 1 2 1 1 G e n ` e v e 4 ; S w i t z e r l a n d ;
e m a i l : n a t h a l i e . g i g e r @ u n i g e . c h
2 D ' e p a r t e m e n t d e s c i e n c e p o l i t i q u e e t r e l a t i o n s i n t e r n a t i o n a l e s , F a c u l t ' e d e s s c i e n c e s d e l a s o c i ' e t ' e
; U n i v e r s i t ' e d e G e n ` e v e ; 4 0 B d d u P o n t
d ' A r v e ; 1 2 1 1 G e n ` e v e 4 ; S w i t z e r l a n d ; p h o n e + + 4 1 2 2 3 7 9 8 3 7 8 ; e m a i l :
s i m o n . h u g @ u n i g e . c h
File translated from
TEX
by
TTH,
version 4.03.
On 20 Jan 2016, 09:21.