demo/cjointdemo.R

In cjoint: AMCE Estimator for Conjoint Experiments

################ Demo for cjoint package - Strezhnev, Berwick, Hainmueller, Hopkins, Yamamoto ####################

# Load data from immigration conjoint data (from the replication data to "Causal Inference in Conjoint Analysis")
data("immigrationconjoint")
data("immigrationdesign")

##  You can construct the conjoint design manually in in R
attribute_list <- list()
attribute_list[["Education"]] <-c("no formal","4th grade",
                                  "8th grade","high school",
                                  "two-year college","college degree",
                                  "graduate degree")
attribute_list[["Gender"]] <- c("female","male")
attribute_list[["Country of Origin"]] <-  c("Germany","France","Mexico",
                                            "Philippines","Poland","India",
                                            "China","Sudan","Somalia","Iraq")
attribute_list[["Reason for Application"]] <- c("reunite with family",
                                                "seek better job",
                                                "escape persecution")
attribute_list[["Job"]] <- c("janitor","waiter","child care provider",
                             "gardener","financial analyst",
                             "construction worker","teacher",
                             "computer programmer","nurse",
                             "research scientist","doctor")
attribute_list[["Job Experience"]] <- c("none","1-2 years",
                                        "3-5 years","5+ years")
attribute_list[["Job Plans"]] <- c("contract with employer",
                                   "interviews with employer", "will look for work",
                                   "no plans to look for work")
attribute_list[["Prior Entry"]] <- c("never","once as tourist",
                                     "many times as tourist","six months with family",
                                     "once w/o authorization")
attribute_list[["Language Skills"]] <- c("fluent English",
                                         "broken English",
                                         "tried English but unable",
                                         "used interpreter")

# Randomization constraints in the conjoint design
constraint_list <- list()
constraint_list[[1]] <- list()
## constraint_list[[1]][["Education"]] <- c("no formal","4th grade",
##                                   "8th grade","high school",
##                                          "two-year college","college degree")
constraint_list[[1]][["Education"]] <- c("no formal")
constraint_list[[1]][["Job"]] <- c("waiter","child care provider",
                             "gardener","financial analyst",
                             "construction worker","teacher",
                             "computer programmer","nurse",
                             "research scientist","doctor")
constraint_list[[2]] <- list()
constraint_list[[2]][["Reason for Application"]] <- c("escape persecution")
constraint_list[[2]][["Country of Origin"]] <- c("Germany","France","Mexico",
                                            "Philippines","Poland","India",
                                            "China","Sudan","Somalia")

immigrationdesign2 <- makeDesign(type='constraints', attribute.levels=attribute_list, constraints=constraint_list)

# Run AMCE estimator - this calls the main amce() function in the cjoint package
results <- amce(Chosen_Immigrant ~  Gender + Education + `Language Skills` + `Country of Origin` + Job + `Job Experience` + `Job Plans` + `Reason for Application` + `Prior Entry`, data=immigrationconjoint, cluster=TRUE, respondent.id="CaseID", design=immigrationdesign2)

# Print summary
summary(results)
# Plot results
plot(results, xlab="Change in Pr(Immigrant Preferred for Admission to U.S.)", xlim=c(-.3,.3), breaks=c(-.2, 0, .2), labels=c("-.2","0",".2"), text.size=13)

# You can specify interactions in the formula using : or *
interaction_results <- amce(Chosen_Immigrant ~ Gender + Education + Job + Education*Job, data=immigrationconjoint, cluster=TRUE, respondent.id="CaseID", design=immigrationdesign) 
# Print summary
summary(interaction_results)
# Plot results
plot(interaction_results, xlab="Change in Pr(Immigrant Preferred for Admission to U.S.)", xlim=c(-.3,.3), breaks=c(-.2, 0, .2), labels=c("-.2","0",".2"), text.size=13)

# Alternative specification using ":"
interaction_results <- amce(Chosen_Immigrant ~ Gender + Education + Job + Education:Job, data=immigrationconjoint, cluster=TRUE, respondent.id="CaseID", design=immigrationdesign)
summary(interaction_results)

### Specify respondent-varying interactions the same way, but make sure to add the variable to the respondent.varying argument # Add respondent-varying interaction
# Subset out NA values prior to running - otherwise it will throw an error
interaction_results <- amce(Chosen_Immigrant ~ Education + Job + Education*ethnocentrism, data=subset(immigrationconjoint, !is.na(immigrationconjoint$ethnocentrism)), cluster=TRUE, respondent.id="CaseID",design=immigrationdesign, respondent.varying=c("ethnocentrism"))

# Summarize results, at quantiles of respondent-varying attribute
summary(interaction_results)

# Plot results by continuous variable, showing non-interacted or not
plot(interaction_results, facet.name=c("ethnocentrism"))
plot(interaction_results, facet.name=c("ethnocentrism"), show.all=T)

# You can manually set the baselines used in the function (in addition to just releveling the factors in your dataset) 
baselines <- list()
baselines$Education <- "graduate degree"

# Run function
interaction_results <- amce(Chosen_Immigrant ~ Gender + Education + Job + Education:Job, data=immigrationconjoint, cluster=TRUE, respondent.id="CaseID",design=immigrationdesign,baselines=baselines)

# Summarize results
summary(interaction_results)

# Plot results by factor variable
plot.amce(interaction_results,facet.name =c("Education"))

# Only plot for some levels of factor variable
facet.levels1 <- list()
facet.levels1[["Education"]] <-(c("college degree","two-year college"))
names(facet.levels1[["Education"]]) <- c("college degree","two-year college")
plot.amce(interaction_results,facet.name =c("Education"), facet.levels = facet.levels1)