Political Paradoxes

1
Political Paradoxes

• Florin Fesnic, FSPU Bucharest, Spring 2008

2
Purpose of the course

• Politics is interesting and fun!
• Political science courses are often boring and
  dull
• Methodology courses are even worse
• It shouldnt be this way
• What is to be done?

3
A new kind of course

• A mix of
• (a) Mainstream political science literature
  (e.g., retrospective voting theory)
• (b) A few basic methodological tools (e.g.,
  linear regression)
• (c) A paradox (shark attacks having an impact
  on voting in US presidential elections)
• ? rigorous analysis using (b) to reconcile (a c)

4
More parties ? higher turnout?

• Why would that be the case?
• Yes more parties, more choices
• ? higher turnout
• No more parties, more headaches (higher
  cognitive skills required)
• ? lower turnout
• Whos right?

5
(No Transcript)
6
Whos right, then?

• Neither side is right, right?
• Higher number of parties neither encourage, nor
  discourage turnout
• Wrong!
• Both sides are right!

7

• Figure 1 results at the aggregate level
• The effects are at the individual level
• Two types of individuals - two kinds of effects
• High-information voters
• More parties ? higher probability (vote)
• Low-information voters
• More parties ? lower probability (vote)

8
(No Transcript)
9

• Both sides are right!
• Increase in the number of parties? Two effects
• ? Increase the chance that high-information
  voters will vote
• ? Decrease the chance that low-information voters
  will vote
• In the aggregate, the two effects cancel each
  other we observe no effect

10
Other examples

• Shark attacks voting behavior
• Achen Bartels, Blind Retrospection
• a string of shark attacks in the summer of 1916
• ? impact on voting behavior in the subsequent
  presidential election

11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
Vote for left across time and space

• Gender ? vote for left?
• Mid-20th century women significantly more
  conservative than men
• Late 20th-early 21st century women significantly
  less conservative than men
• Education ? vote for left?

15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
Clientelism voting behavior

• Are clientelistic messages more effective than
  policy-oriented messages?
• Can we test this empirically? (i.e., a real
  test/experiment?)
• Take a real country (Benin), a real election (the
  2001 presidential election) and do an experiment

19

• If politics is fun, teaching learning about it
  should be, and can be, equally fun
• This can apply even to methodology
• Making Methodology Work a tall order, but
• If we pick appropriate examples, not an
  impossible task

20
Suggestions?

• Additional topics?
• Additional readings?
• Structure?
• Sequence?
• ? complement of classic methods courses
• ? should it be taught before or after such
  courses?