Hi everyone,
I hope your paper is coming along. The deadline is tomorrow 4pm. As you
enter the reception area of CBRSS, you will see on your right a mailbox
labeled as "Gov2001." Please drop off one copy of the paper (one copy per
each group) there. The paper format should follow *all* of the guidelines
specified in http://gking.harvard.edu/papers/papers.html Please read this
document carefully one more time before you turn in your paper. Good luck!
Kosuke
Hi, all comments are welcome and appreciated.
Thanks, Olivia and Asif.
Title: Lighting the Powderkeg: Why Leaders Initiate International Crises
Do international crises occur because aggressors look for easy gains?
Gelpi and Grieco (2001) argue that potential challengers find democracies
to be easy prey because democratic leaders are more likely to be
inexperienced, and hence more likely to make concessions during
international crises. We correct for time-dependency in Gelpi and
Grieco's dependent variable and argue that they overestimate the
intervening effect of leadership tenure on the probability of crisis.
We argue that if the probability of crisis reflects the challenging
leader's estimation that the defending leader will make concessions,
then it should have a negative effect on the observed length of crises.
We find that increasing the predicted probability of crisis initiation
from 0 to 1 all else being equal increases the predicted length of
the crisis from 93 days to 160 days. Rather than supporting the
argument that challenging leaders make opportunistic judgments about
the defending leaders willingness to concede, our results support
the hypothesis that international crises occur because states
actually have substantive areas of disagreement.
---------- Forwarded message ----------
Date: Tue, 13 May 2003 09:49:35 -0400
From: Matthew Cox via RT <king_support(a)backup.hmdc.harvard.edu>
To: king(a)harvard.edu
Subject: [hmdc.harvard.edu #3039] re: gov 2001 (from Anna Nelson)
Including the text of ticket 3042:
To: fascs-l(a)fas.harvard.edu
From: Noah Selsby <nselsby(a)fas.harvard.edu>
Subject: Sporadic Downtime Today
Date: Mon, 12 May 2003 22:44:58 -0400
FAS UNIX system services were unavailable or sporadically available between
11:15 AM and 6:30 PM today (05/12/2003). Affected services included:
- Home Directory Access
- Mail Delivery
- POP/IMAP/Webmail mail access
- Web Services
These interruptions resulted from a series of software and hardware
failures. We are continuing to investigate the causes of these failures and
apologize for any inconvenience these interruptions caused.
Thank you for your patience,
Noah Selsby
--
Matthew P. Cox
Systems Administrator
Harvard-MIT Data Center
Be aware of some possible interruptions althogh I'm not sure if ice
servers will be affected.
Kosuke
---------- Forwarded message ----------
>FAS UNIX system services were unavailable or sporadically available
>between 11:15 AM and 6:30 PM today (05/12/2003). Affected services included:
>
>- Home Directory Access
>- Mail Delivery
>- POP/IMAP/Webmail mail access
>- Web Services
>
>These interruptions resulted from a series of software and hardware
>failures. We are continuing to investigate the causes of these failures
>and apologize for any inconvenience these interruptions caused.
>
>Thank you for your patience,
>Noah Selsby
>
>_______________________________________________
>fascs-l mailing list
>fascs-l(a)fas.harvard.edu
>http://www.fas.harvard.edu/mailman/listinfo/fascs-l
This is a multi-part message in MIME format.
------=_NextPart_000_0020_01C318F3.D4959C00
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MessageIf you want real table notes, you have to use the astronomy =
package. I'm not sure how you call it exactly, but it's something like =
\documentclass[...]{article} where [...] is the astronomy package. =20
If you want to fake it, you insert the symbol you want (^\symbol) where =
you want it, and then after \end{tabular} \end{center}
\\
\footnotesize
then you indicate your symbol and type your notes=20
\normalsize=20
\end{table}
Other people fake it by doing a big \multicolumn{number of =
columns}{l}{footnote text}. =20
----- Original Message -----=20
From: Stanislav Markus=20
To: 'Olivia Lau'=20
Sent: Tuesday, May 13, 2003 1:56 AM
Subject: table footnotes..?
do you remember by any chance how to do table footnotes in latex..?
****************************
Stanislav Markus
Ph.D. Candidate
Harvard University
Department of Government
e: smarkus(a)fas.harvard.edu
t: 617.513.5407=20
------=_NextPart_000_0020_01C318F3.D4959C00
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD><TITLE>Message</TITLE>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2800.1170" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>If you want real table notes, you have to use the =
astronomy=20
package. I'm not sure how you call it exactly, but it's something =
like=20
\documentclass[...]{article} where [...] is the astronomy package. =
</FONT></DIV>
<DIV><FONT size=3D2></FONT> </DIV>
<DIV><FONT size=3D2>If you want to fake it, you insert the symbol you =
want=20
(^\symbol) where you want it, and then after \end{tabular}=20
\end{center}</FONT></DIV>
<DIV><FONT size=3D2>\\</FONT></DIV>
<DIV><FONT size=3D2>\footnotesize</FONT></DIV>
<DIV><FONT size=3D2>then you indicate your symbol and type your notes=20
</FONT></DIV>
<DIV><FONT size=3D2>\normalsize </FONT></DIV>
<DIV><FONT size=3D2>\end{table}</FONT></DIV>
<DIV><FONT size=3D2></FONT> </DIV>
<DIV><FONT size=3D2>Other people fake it by doing a big =
\multicolumn{number of=20
columns}{l}{footnote text}. </FONT></DIV>
<DIV><FONT size=3D2></FONT> </DIV>
<DIV><FONT size=3D2></FONT> </DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV style=3D"FONT: 10pt arial">----- Original Message ----- </DIV>
<DIV=20
style=3D"BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: =
black"><B>From:</B>=20
<A title=3Dsmarkus(a)fas.harvard.edu=20
href=3D"mailto:smarkus@fas.harvard.edu">Stanislav Markus</A> </DIV>
<DIV style=3D"FONT: 10pt arial"><B>To:</B> <A =
title=3Dolau(a)fas.harvard.edu=20
href=3D"mailto:olau@fas.harvard.edu">'Olivia Lau'</A> </DIV>
<DIV style=3D"FONT: 10pt arial"><B>Sent:</B> Tuesday, May 13, 2003 =
1:56 AM</DIV>
<DIV style=3D"FONT: 10pt arial"><B>Subject:</B> table =
footnotes..?</DIV>
<DIV><BR></DIV>
<DIV><SPAN class=3D996355505-13052003><FONT face=3DArial size=3D2>do =
you remember by=20
any chance how to do table footnotes in latex..?</FONT></SPAN></DIV>
<DIV> </DIV><!-- Converted from text/plain format -->
<P align=3Dleft><FONT =
size=3D2>****************************<BR><BR>Stanislav=20
Markus<BR>Ph.D. Candidate<BR><BR>Harvard University<BR>Department of=20
Government<BR><BR>e: <A=20
=
href=3D"mailto:smarkus@fas.harvard.edu">smarkus(a)fas.harvard.edu</A><BR>t:=
=20
617.513.5407</FONT> </P>
<DIV> </DIV></BLOCKQUOTE></BODY></HTML>
------=_NextPart_000_0020_01C318F3.D4959C00--
I think I understand what ROC curves indicate for binary choice models:
whether, for many values of C, one model always, sometimes, or never
predicts 0/1's correctly.
Now, can ROC curves compare an imputed dataset to a non-imputed one? The
idea would be this: for the same model, more information allows you to
better predict the 0/1's than less information. Does this make sense? Is
it trivial?
Thanks,
Ryan
--
------------------------------------------
Ryan T. Moore ~ Government & Social Policy
Ph.D. Candidate ~ Harvard University
If you're using \caption{} in latex, put a \\ where you want to
wrap.
It may be because the graphic you're placing is itself wider
than your text width. Remember that R graphics by default have
a half-inch margin of white-space all around. You can reduce
this with par(mai = ...) in the plot command.
----- Original Message -----
From: "Stanislav Markus" <smarkus(a)fas.harvard.edu>
To: "'Olivia Lau'" <olau(a)fas.harvard.edu>
Sent: Tuesday, May 13, 2003 1:53 AM
Subject: RE: [gov2001-l] text wrapping in latex tables
> LATEX, not R ;)
>
> ****************************
>
> Stanislav Markus
> Ph.D. Candidate
>
> Harvard University
> Department of Government
>
> e: smarkus(a)fas.harvard.edu
> t: 617.513.5407
>
>
>
> -----Original Message-----
> From: Olivia Lau [mailto:olau@fas.harvard.edu]
> Sent: Tuesday, May 13, 2003 1:53 AM
> To: Stanislav Markus
> Subject: Re: [gov2001-l] text wrapping in latex tables
>
>
> If you're producing the title using plot(... main = " title ")
> then you just put a \n where you want the break in your title.
>
>
> ----- Original Message -----
> From: "Stanislav Markus" <smarkus(a)fas.harvard.edu>
> To: <gov2001-l(a)fas.harvard.edu>
> Sent: Tuesday, May 13, 2003 12:54 AM
> Subject: [gov2001-l] text wrapping in latex tables
>
>
> > for some reason my text (e.g. long title) in tables is not
> automatically
> > wrapped but extends beyond the margins.. how do i fix it?
> >
> > thanks!
> >
> > ****************************
> >
> > Stanislav Markus
> > Ph.D. Candidate
> >
> > Harvard University
> > Department of Government
> >
> > e: smarkus(a)fas.harvard.edu
> > t: 617.513.5407
> >
> >
> >
> > -----Original Message-----
> > From: gov2001-l-admin(a)fas.harvard.edu
> > [mailto:gov2001-l-admin@fas.harvard.edu] On Behalf Of Ryan
> Thomas Moore
> > Sent: Tuesday, May 13, 2003 12:49 AM
> > To: gov2001-l(a)fas.harvard.edu
> > Cc: Gary King; Kosuke Imai
> > Subject: [gov2001-l] ROC "models"
> >
> >
> >
> > I think I understand what ROC curves indicate for binary
> choice models:
> >
> > whether, for many values of C, one model always, sometimes,
or
> never
> > predicts 0/1's correctly.
> >
> > Now, can ROC curves compare an imputed dataset to a
> non-imputed one?
> > The
> > idea would be this: for the same model, more information
> allows you to
> > better predict the 0/1's than less information. Does this
> make sense?
> > Is
> > it trivial?
> >
> > Thanks,
> > Ryan
> >
> >
> > --
> > ------------------------------------------
> > Ryan T. Moore ~ Government & Social Policy
> > Ph.D. Candidate ~ Harvard University
> >
> > _______________________________________________
> > gov2001-l mailing list
> > gov2001-l(a)fas.harvard.edu
> > http://www.fas.harvard.edu/mailman/listinfo/gov2001-l
> >
> > _______________________________________________
> > gov2001-l mailing list
> > gov2001-l(a)fas.harvard.edu
> > http://www.fas.harvard.edu/mailman/listinfo/gov2001-l
> >
>
This is a multi-part message in MIME format.
------=_NextPart_000_0001_01C3187D.D1DBFDE0
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
charset="US-ASCII"
Thanking you in advance for comments and criticism, Beau.
An Argument for Excluding Low-Income Countries from Cross-Country Gun
Control Studies
Miron (2001a) argues that differences in drug enforcement explain
differences in homicide rates and that omission of the former explains
why some studies find a negative relationship between gun control and
violence. Miron's paper, however, does not report that he imputes zero
for missing drug seizure values and I illustrate the bias of this
approach. After employing multiple imputation, I find that his measures
of drug enforcement and gun control play virtually no role in explaining
homicide rates. Further, when controlling for these variables in the
full datasets, simulations reveal that countries with high per capita
GNP have tremendously lower homicide rates than countries with low per
capita GNP. I conclude that those wanting to understand the
relationship between guns and homicide in high-income countries should
not include low-income countries in their analyses.
Beau Kilmer
Pre-doctoral Candidate
John F. Kennedy School of Government
PhD Mailboxes
79 JFK Street
Cambridge, MA 02138
------=_NextPart_000_0001_01C3187D.D1DBFDE0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
charset="US-ASCII"
<html>
<head>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<meta name=3DGenerator content=3D"Microsoft Word 10 (filtered)">
<style>
<!--
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Times New Roman";}
a:link, span.MsoHyperlink
{color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{color:purple;
text-decoration:underline;}
p.MsoAutoSig, li.MsoAutoSig, div.MsoAutoSig
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman";}
span.EmailStyle17
{font-family:Arial;
color:windowtext;}
@page Section1
{size:8.5in 11.0in;
margin:1.0in 1.25in 1.0in 1.25in;}
div.Section1
{page:Section1;}
-->
</style>
</head>
<body lang=3DEN-US link=3Dblue vlink=3Dpurple>
<div class=3DSection1>
<p class=3DMsoNormal><font size=3D2 face=3D"Times New Roman"><span =
style=3D'font-size:
11.0pt'>Thanking you in advance for comments and criticism, =
Beau.</span></font></p>
<p class=3DMsoNormal><b><font size=3D2 face=3D"Times New Roman"><span
style=3D'font-size:11.0pt;font-weight:bold'> </span></font></b></p>
<p class=3DMsoNormal><b><font size=3D2 face=3D"Times New Roman"><span
style=3D'font-size:11.0pt;font-weight:bold'>An Argument for Excluding =
Low-Income Countries
from Cross-Country Gun Control Studies</span></font></b></p>
<p class=3DMsoNormal><font size=3D2 face=3D"Times New Roman"><span =
style=3D'font-size:
11.0pt'> </span></font></p>
<p class=3DMsoNormal style=3D'text-align:justify'><i><font size=3D2
face=3D"Times New Roman"><span =
style=3D'font-size:11.0pt;font-style:italic'> Miron
(2001a) argues that differences in drug enforcement explain differences =
in
homicide rates and that omission of the former explains why some studies =
find a
negative relationship between gun control and violence. =
Miron’s paper,
however, does not report that he imputes zero for missing drug seizure =
values
and I illustrate the bias of this approach. After employing =
multiple imputation,
I find that his measures of drug enforcement and gun control play =
virtually no
role in explaining homicide rates. Further, when controlling for =
these
variables in the full datasets, simulations reveal that countries with =
high per
capita GNP have tremendously lower homicide rates than countries with =
low per
capita GNP. I conclude that those wanting to understand the =
relationship
between guns and homicide in high-income countries should not include
low-income countries in their analyses.</span></font></i></p>
<p class=3DMsoNormal><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;
font-family:Arial'> </span></font></p>
<p class=3DMsoNormal><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;
font-family:Arial'> </span></font></p>
<p class=3DMsoNormal><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;
font-family:Arial'> </span></font></p>
<p class=3DMsoNormal><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;
font-family:Arial'> </span></font></p>
<p class=3DMsoNormal><font size=3D2 face=3DArial><span =
style=3D'font-size:10.0pt;
font-family:Arial'> </span></font></p>
<p class=3DMsoAutoSig><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'>Beau Kilmer</span></font></p>
<p class=3DMsoAutoSig><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'>Pre-doctoral Candidate</span></font></p>
<p class=3DMsoAutoSig><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'>John</span></font> F. Kennedy School of Government</p>
<p class=3DMsoAutoSig><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'>PhD Mailboxes</span></font></p>
<p class=3DMsoAutoSig><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'>79 JFK Street</span></font></p>
<p class=3DMsoAutoSig><font size=3D3 face=3D"Times New Roman"><span =
style=3D'font-size:
12.0pt'>Cambridge</span></font>, MA 02138</p>
<p class=3DMsoNormal><font size=3D2 face=3D"Times New Roman"><span =
style=3D'font-size:
11.0pt'> </span></font></p>
</div>
</body>
</html>
------=_NextPart_000_0001_01C3187D.D1DBFDE0--
Hi all,
We have consistent patterns in our data that show ballot fatigue.
However, due to our very small data set sizes (115 to 138 pooled
time-series cross-sectional observations in each data subset) and the need
to calculate panel corrected standard errors, we are having difficulty
finding results that are statistically significant with 95% confidence.
Is it legitimate to report results with, say, 80% confidence? For
instance, when we drop to 80% confidence, we find that low (but
non-zero) number of initiatives on the ballot increases turnout an average
of 6.4% over elections where there are a fairly high number of initiatives
on the ballot. This seems like decent evidence of "ballot fatigue," and
the point estimate of a 6.4% difference in turnout seems quite politically
significant.
So, is this sort of result -- with 80% confidence -- legitimate to report?
Thanks,
Anna
I have a problem. I have a dependent variable Y, and two independent
variables X1 and X2, such that
X1 -> Y
X1 -> X2
X2 -> Y
Is this multicollinearlity? I'm at a loss for words, someone please help!
Thanks,
Olivia.