Lecture 21
11/05/2014
What happened yesterday? Is the US overflowing with the spirit of
democracy and vital, robust civil society?
Political Spectrum
Strong: republican, partisans
Weak: republican, partisans
Independents: who lean republican
swing voters
Tax & Education Policy Template
FL
TX
Tax
Is there a balanced
budget
requirement?
What is necessary to
raise taxes? (e.g.
voter approval,
legislative approval,
etc.)
Are certain taxes
prohibited? If so,
what are they?
How does a budget
get passed? (What
a
Health, Welfare & Social Policy Template
FL
What percentage of
the state budget is
spent on Medicare?
What percentage of
the state budget is
spent on Medicaid?
Did the state accept
expansion of health
care under ACA?
What percentage of
the state budget is
PS 3: Spatial Party Competition and Introduction to
Game Theory
Jason Wittenberg
University of California, Berkeley
Wittenberg (UC Berkeley)
PS 3: The Art of Theory-Building
1 / 29
Downsian Spatial Model: Assumptions
One-dimensional: left-right space with
PS 3: Causality
Jason Wittenberg
University of California, Berkeley
Wittenberg (UC Berkeley)
PS 3: Causality
1 / 18
The focus on causality
One goal of political science is to evaluate causal theories/make
causal inferences.
Establishing causal relationshi
Practice Regression Problem
You are working for an incumbents political campaign and are eager to show off what you
learned in PS3. So you run a simple OLS regression, with incumbent vote as the dependent
variable and two independent variables. One is goo
PS 3: Research Design
Jason Wittenberg
University of California, Berkeley
Wittenberg (UC Berkeley)
PS 3: Research Design
1 / 22
What is being compared to what?
Making good comparisons is one of the keys of doing social science.
The simple bivariate compar
PS 3: Multivariate Regression
Jason Wittenberg
University of California, Berkeley
Last Lecture Topic :(
Wittenberg (UC Berkeley)
PS 3: Multivariate Regression
Last Lecture Topic :(
1 / 33
Review of the research process
Start out with a question: what fact
PS 3: Bivariate Regression
Jason Wittenberg
University of California, Berkeley
Wittenberg (UC Berkeley)
PS 3: Bivariate Regression
1 / 34
Correlation and bivariate regression
Recall table 7.1: when we have a continuous dependent variable and
a continuous
PS 3: Conceptualization and Measurement
Jason Wittenberg
University of California, Berkeley
Wittenberg (UC Berkeley)
PS 3: Conceptualization and Measurement
1 / 16
How do we evaluate our theories?
How do we come to a conclusion about whether our theory is
Political Science 124 Midterm Study Sheet
Important Concepts
- Infantry: foot soldiers; used in both ancient and modern battles; modern day infantry uses
firearms
- Archers: soldiers wielding bows and arrows; today is incorporated in the modern-day
infant
Megan Perry
26 April 2017
Dean Spade examines in his piece, Mutilating Genders, that the creation of the subject
position transsexual by the medical establishment restricts individuals seeking body alteration
and promotes the creation of norm-abiding gend
1
25446999
PS 123s
Melanie Thompson
28 September 2016
Domestication of CEDAW in Nigeria
While the Universal Declaration of Human Rights is considered a universal articulation
of human rights, violations on the basis of sex continue to persist within its f
Ideas which are being refuted:
-Economic development occurs in a succession of capitalist stages and todays
underdeveloped countries remain in a stage which the now developed countries passed through
long ago. Pg 375
-Development of underdeveloped countri
Megan Perry
PS 139b
Thursday 11-12:30
Marxism Maoism and Utopia
Maurice Meisner makes the claim that both Marxism and Maoism are concerned with the
development of society and the resultant growing gap between town and country. He notes
however that though
Megan Perry
Terms for Section 10/28
Formal vs Substantive Equality
In the piece Sex Equality Under the Constitution of India, Catharine Mackinnon notes that one
of the hindrances to legislating a world in which women achieve equality is the very vision of
focus of this essay is the second area where multilevel contexts are
likely to exist. Multilevel contexts are likely to be present where the
treatment consists of something that can be communicated across social
ties, such as information. This type of ran
control, and we are able to look at the difference in terms of expected
outcomes. We define the ITT effect as ITT = E(Yi|t = 1)-E(Yi|t = 0).
SUTVA allows us to consider only the assignment of individuals. If we
assume that there may be spillovers, however
appears to calculate the causal effect of X on Y, but as we will see below,
closer examination shows that the estimator isolates the causal effect for
X on Y for those subjects affected by the initial intervention subjects
whose X outcome changed due to t
which we can measure potential spillovers and also design experiments
in order to be able to correct for their potential effects. We contend that
spillover effects could exist within households or within groups. We now
look at an example where, in the pre
voted if they graduated from high school but did not vote if they did not
graduate from high school (yi1=1, yi0=0); 3) Individuals who did not
vote if they graduated from high school but did vote if they did not
graduate from high school (yi1=0, yi0=1); a
easily come from: covariance adjustment done with multiple regression
without additional diagnostics poses a real problem for diagnosing
whether the imbalance is so severe as to provide no common support
in the distributions of the covariates. In such cas
adjusted for baseline turnout and number of candidates (just) excludes
zero from its 95% confidence interval. The two approaches differ when
the difference of means is adjusted for the Census variables. The most
notable difference here is for median house
an even more complicated problem where there could be communication
between many households within a group. The true ITT would then need
to be written based upon all the instances of communication. The
consequences of these spillovers are such that it is
test a test which assessess balance across all linear combinations of
the covariates in the table. Yet, the variable by variable display is useful
in the same way that graphs such as Figure 32-2 are useful in
suggesting (not proving) the sources of imbala
earlier, the assumption of parallel lines is not correct. Second, we begin
to notice another problem not mentioned in textbooks like Cox and Reid
(2000) or Cox (1958) random assignment will, in large samples,
ensure balance in the distributions of covaria
which experiments are often untenable for practical and ethical reasons.
Green and Gerber (2002) define these downstream benefits as
knowledge acquired when one examines the indirect effects of a
randomized experimental intervention (394). Analyzing the s