xergm (version 1.5.3)

chemnet: German Toxic Chemicals Policy Network in the 1980s (Volker Schneider)

Description

The chemnet dataset contains network and attribute data and for the 30 most influential political actors with regard to toxic chemicals regulation in Germany in 1983/1984. While the original dataset contains up to 47 actors, this dataset contains the "complete influence core" of mutually relevant actors. The data are cross-sectional. There are no missing data; the response rate was 100 percent. Volker Schneider (University of Konstanz) collected this dataset for his dissertation (Schneider 1988). The dataset was later re-used for a journal publication on information exchange in policy networks (Leifeld and Schneider 2012). The chemnet dataset contains network relations on political/strategic and technical/scientific information exchange, influence attribution, and membership in policy committees/forums, as well as nodal attributes on the actor type and opinions about the six most salient issues related to the political process that was leading to a new chemicals law at the time being.

Usage

data(chemnet)

source

The data were collected using paper-based questionnaires. The questionnaires were administered in personal interviews (PAPI). Further information, including the actual survey, data on additional actors, the full names of the policy committees/forums, and the full list of unabbreviated actor names can be found online at http://hdl.handle.net/1902.1/17004 in the replication archive of Leifeld and Schneider (2012).

• Replication archive:

{http://hdl.handle.net/1902.1/17004} AJPS publication:{http://dx.doi.org/10.1111/j.1540-5907.2011.00580.x}

References

Leifeld, Philip and Volker Schneider (2012): Information Exchange in Policy Networks. American Journal of Political Science 53(3): 731--744. http://dx.doi.org/10.1111/j.1540-5907.2011.00580.x. Schneider, Volker (1988): Politiknetzwerke der Chemikalienkontrolle. Eine Analyse einer transnationalen Politikentwicklung. Walter de Gruyter: Berlin/New York. Schneider, Volker and Philip Leifeld (2009): Ueberzeugungssysteme, Diskursnetzwerke und politische Kommunikation: Ein zweiter Blick auf die deutsche Chemikalienkontrolle der 1980er Jahre. In: Volker Schneider, Frank Janning, Philip Leifeld and Thomas Malang (editors): Politiknetzwerke. Modelle, Anwendungen und Visualisierungen. Pages 139--158. Wiesbaden: VS Verlag fuer Sozialwissenschaften. http://dx.doi.org/10.1007%2F978-3-531-91883-9_6.

Examples

# Replication code for Leifeld and Schneider (2012), AJPS.
# Note that the estimates can only be reproduced approximately 
# due to internal changes in the statnet package.

# preparatory steps
library("statnet")
library("xergm")
library("texreg")
seed <- 12345
set.seed(seed)
data("chemnet")

# create confirmed network relation
sci <- scito * t(scifrom)  # equation 1 in the AJPS paper
prefsim <- dist(intpos, method = "euclidean")  # equation 2
prefsim <- max(prefsim) - prefsim  # equation 3
prefsim <- as.matrix(prefsim)
committee <- committeediag(committee) <- 0 # the diagonal has no meaning
types <- types[, 1]  # convert to vector

# create network objects and store attributes
nw.pol <- network(pol) # political/stratgic information exchange
set.vertex.attribute(nw.pol, "orgtype", types)
set.vertex.attribute(nw.pol, "betweenness", 
    betweenness(nw.pol)) # centrality

nw.sci <- network(sci) # technical/scientific information exchange
set.vertex.attribute(nw.sci, "orgtype", types)
set.vertex.attribute(nw.sci, "betweenness", 
    betweenness(nw.sci)) # centrality

# ERGM: model 1 in the AJPS paper; only preference similarity
model1 <- ergm(nw.pol ~ edges + edgecov(prefsim), 
    control = control.ergm(seed = seed))
summary(model1)

# ERGM: model 2 in the AJPS paper; complete model
model2 <- ergm(nw.pol ~ 
    edges + 
    edgecov(prefsim) + 
    mutual + 
    nodemix("orgtype", base = -7) + 
    nodeifactor("orgtype", base = -1) + 
    nodeofactor("orgtype", base = -5) + 
    edgecov(committee) + 
    edgecov(nw.sci) + 
    edgecov(infrep) + 
    gwesp(0.1, fixed = TRUE) + 
    gwdsp(0.1, fixed = TRUE), 
    control = control.ergm(seed = seed)
)
summary(model2)

# ERGM: model 3 in the AJPS paper; only preference similarity
model3 <- ergm(nw.sci ~ edges + edgecov(prefsim), 
    control = control.ergm(seed = seed))
summary(model3)

# ERGM: model 4 in the AJPS paper; complete model
model4 <- ergm(nw.sci ~ 
    edges + 
    edgecov(prefsim) + 
    mutual + 
    nodemix("orgtype", base = -7) + 
    nodeifactor("orgtype", base = -1) + 
    nodeofactor("orgtype", base = -5) + 
    edgecov(committee) + 
    edgecov(nw.pol) + 
    edgecov(infrep) + 
    gwesp(0.1, fixed = TRUE) + 
    gwdsp(0.1, fixed = TRUE), 
    control = control.ergm(seed = seed)
)
summary(model4)

# regression table using the texreg package
screenreg(list(model1, model2, model3, model4))

# goodness of fit using the xergm package
gof2 <- gof(model2, roc = FALSE, pr = FALSE)
gof2  # print gof output
plot(gof2)  # visual inspection of GOF

gof4 <- gof(model4, roc = FALSE, pr = FALSE)
gof4
plot(gof4)

# MCMC diagnostics
pdf("diagnostics2.pdf")
mcmc.diagnostics(model2)
dev.off()

pdf("diagnostics4.pdf")
mcmc.diagnostics(model4)
dev.off()