SCOPE CONDITIONS PODCAST

Scope Conditions is a podcast featuring cutting-edge research in comparative politics hosted by Alan Jacobs and me. In each episode, we talk with an emerging scholar about their new book or paper, with a particular focus on innovative empirical research covering a wide range of themes and regions of the world. The conversations on Scope Conditions speak to both substance and method: what these scholars have found and how they have found it. You can find it on Spotify, Apple, or wherever you listen to podcasts.

SW-ScopeConditions-Final2.jpg

DIVERSITY AND INCLUSION IN ACADEMIA INITIATIVES

Hackathon_logo.png

2018 APSA Hackathon Team for Graduate Student Diversity and Inclusion (co-led with Leah Rosenzweig)

What are some ways in which political science departments can create an inclusive and productive climate for all graduate students? The goals of this team are to brainstorm action items for improving graduate student life, to analyze our original climate survey data of 400 current and recent graduate students from political science departments across the country, and to create this website to house all of the resources discussed and visualize the data.

SOFTWARE

rr: Statistical Methods for the Randomized Response Technique: 

rr is an R package which enables researchers to conduct multivariate statistical analyses of survey data with randomized response technique items from several designs, including mirrored question, forced question, and unrelated question. This includes regression with the randomized response as the outcome and logistic regression with the randomized response item as a predictor. In addition, tools for conducting power analysis for designing randomized response items are included. The package implements methods described in Blair, Imai, and Zhou (2015) "Design and Analysis of the Randomized Response Technique.” Type install.packages(“rr”) to install in R.

Example using rrreg() to conduct multivariate regression analysis. 


data(nigeria)
set.seed(1)

## Define design parameters
p <- 2/3 # probability of answering honestly in Forced Response Design
p1 <- 1/6 # probability of forced 'yes'
p0 <- 1/6 # probability of forced 'no'

## Fit linear regression on the randomized response item of whether respondents had direct contact to armed groups
rr.q1.reg.obj <- rrreg(rr.q1 ~ cov.asset.index + cov.married +
        I(cov.age/10) + I((cov.age/10)^2) + cov.education + cov.female,
        data = nigeria, p = p, p1 = p1, p0 = p0,
        design = "forced-known")
summary(rr.q1.reg.obj)

## Replicates Table 3 in Blair, Imai, and Zhou (2015)