GerberGreen

<p>This data set contains the most recent corrected data from the field
experiment analyzed in Gerber and Green (2000).</p>

datasets

The heterogeneous treatment effect estimation procedure
proposed by Imai and Ratkovic (2013)<DOI: 10.1214/12-AOAS593>.
The proposed method is applicable, for
example, when selecting a small number of most (or least)
efficacious treatments from a large number of alternative
treatments as well as when identifying subsets of the
population who benefit (or are harmed by) a treatment of
interest. The method adapts the Support Vector Machine
classifier by placing separate LASSO constraints over the
pre-treatment parameters and causal heterogeneity parameters of
interest. This allows for the qualitative distinction between
causal and other parameters, thereby making the variable
selection suitable for the exploration of causal heterogeneity.
The package also contains a class of functions, CausalANOVA,
which estimates the average marginal interaction effects (AMIEs)
by a regularized ANOVA as proposed by Egami and Imai (2019)<DOI:10.1080/01621459.2018.1476246>.
It contains a variety of regularization techniques to facilitate
analysis of large factorial experiments.

Naoki Egami

FindIt

Finding Heterogeneous Treatment Effects

GerberGreen function

<p>A data frame consisting of 9 columns and 29,380 observations.</p><table><tr><td> voted98 </td>
<td> integer </td>
<td> voted in 1998 </td>
<td> 0,1 </td>
</tr><tr><td>
persngrp </td>
<td> factor </td>
<td> personal contact attempted </td>
<td> 0,1 </td>
</tr><tr><td> phnscrpt
</td>
<td> factor </td>
<td> script read to phone respondents </td>
<td> 7 levels </td>
</tr><tr><td> mailings
</td>
<td> factor </td>
<td> number of mailings sent </td>
<td> 0 - 3 </td>
</tr><tr><td> appeal </td>
<td> factor
</td>
<td> content of message </td>
<td> 3 levels </td>
</tr><tr><td> age </td>
<td> integer </td>
<td> age of
respondent </td>
<td> </td>
</tr><tr><td> majorpty </td>
<td> factor </td>
<td> Democratic or Republican </td>
<td>
</td>
</tr><tr><td> voted96.1 </td>
<td> factor </td>
<td> voted in 1996 </td>
<td> 0,1 </td>
</tr><tr><td> voted96.0 </td>
<td>
factor </td>
<td> abstained in 1996 </td>
<td> 0,1 </td>
</tr></table><p></p><p></p><p>Note: The levels of phnscrpt and appeal are follows.</p>
<p>phnscrpt: Script read to phone respondents</p><table><tr><td> 0 </td>
<td> No phone </td>
</tr><tr><td>
1 </td>
<td> Civic-Blood </td>
</tr><tr><td> 2 </td>
<td> Civic </td>
</tr><tr><td> 3 </td>
<td> Civic or Blood-Civic </td>
</tr><tr><td> 4
</td>
<td> Neighbor </td>
</tr><tr><td> 5 </td>
<td> Neighbor or Civic-Neighbor </td>
</tr></table><p></p><p></p><p>appeal: Content of message</p><table><tr><td> 1 </td>
<td> Civic Duty </td>
</tr><tr><td> 2 </td>
<td>
Neighborhood Solidarity </td>
</tr></table>

GerberGreen: Data from the 1998 New Haven Get-Out-the-Vote Experiment

Description

Arguments

Format

References

voted98	integer	voted in 1998	0,1
persngrp	factor	personal contact attempted	0,1
phnscrpt	factor	script read to phone respondents	7 levels
mailings	factor	number of mailings sent	0 - 3
appeal	factor	content of message	3 levels
age	integer	age of respondent
majorpty	factor	Democratic or Republican
voted96.1	factor	voted in 1996	0,1
voted96.0	factor	abstained in 1996	0,1

0	No phone
1	Civic-Blood
2	Civic
3	Civic or Blood-Civic
4	Neighbor
5	Neighbor or Civic-Neighbor