Re: st: question on ivreg2, ivendog and "good" instruments

[Clive Nicholas <clivelists@googlemail.com>]

I was hoping to send this message a lot sooner than I was able to,
which some minor technical problems prevented me from doing. Following
some offlist advice from Marcello Pagano, here's my latest attempt.

Masha Zakharova wrote (at the beginning of this thread):

"My model looks like this: My dependent variable is party valence -
i.e., how much one likes the party/leader; my instrumented variable is
one's subjective distance to that party, which is an absolute distance
between respondent's self-placement on the left-right scale and
his/her party placement. The instrument is objective distance, which
is an absolute distance between one's corrected self-placement and
non-subjective party position in that election that are both
calculated using a scaling procedure in R (Aldrich-Mcalvey scaling).
So theoretically objective distance (instrument) should not be related
to valence (main DV)."

I don't really have any comments to make about your use of IV, or of
'valence' as a response variable, but I do have something to say about
your use of the term 'valence' in political science.

I'm afraid that it simply doesn't mean what you think it means. "How
much one likes the party/leader" connects to the concept of
_partisanship_, not valence. Essentially, valence has two components
which necessitates the putting of two survey questions:

(1) As far as you are concerned, what is the single most important
issue facing the country at this present time?

(There may then be a follow-up question asking which other national
issues the respondent considers to be important at present.)

(2) Which party is best able to handle this issue? (For those able to
answer (1).)

As you can see, the concept of political valence has nothing to do
with liking or disliking and everything to do with which party or
leader the respondent thinks would deliver best on the issue(s) they
care about the most.

These questions are taken from the 2005 British Election Study,
analysed in detail by Clarke, Sanders, Stewart and Whiteley (2009: fn.
6, 340). You would do well to read it, as I did recently (throughout,
they found that the decisions made by British voters are heavily
valenced) as well as the original paper by Stokes (1963), who
developed the concept of valence, in order to sharpen your
understanding of what has now become a very hot topic in the analysis
of voting.

It may also mean having to re-think your entire model.

--
Clive Nicholas

Clarke, H.D., Sanders, D., Stewart, M.C. and Whiteley, P.F. (2009)
Performance Politics and the British Voter, Cambridge: Cambridge
University Press.

Stokes, D.E. (1963) "Spatial Models of Party Competition", American
Political Science Review 57(2), 368-77.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/