addcategorytopolls: Adds Category assignment to Poll Data This function returns a...

In balexanderstats/bayesurvey: Bayesian Conjugate Prior Analysis of Survey Data

Description

Adds Category assignment to Poll Data This function returns a new data frame with a priorcat column to be accessed in other functions.

Usage

addcategorytopolls(
  poll_data,
  proploc,
  stateloc,
  election_data,
  cutoffs = c(-0.2, -0.1, -0.025, 0.025, 0.1, 0.2),
  groupnames = c("Strong Red", "Red", "Lean Red", "Competitive", "Lean Blue", "Blue",
    "Strong Blue"),
  logit = F
)

Arguments

poll_data	the data to fit the prior distribution , the first column must be the state name
proploc	the column the adjusted poll proportion is located
stateloc	the column the state is located in
election_data	the election data in df form. The first column must contain the name of that state.
cutoffs	the cutoffs used to split the data into the categories
groupnames	labels for the categories
logit	if a logit transformation for the prior means is desired

Value

poll_data with the prior category appended at the end

Examples

require(politicaldata)
elect2008  = subset(pres_results , year == 2008)
elect2008$margin = elect2008$dem - elect2008$rep
elect2012 = subset(pres_results , year == 2012)
elect2012$margin = elect2012$dem - elect2012$rep
electdata = data.frame("state" = elect2008$state, "2008" =  elect2008$margin, "2012" = elect2012$margin)
set.seed(16)
data(polls2016)
poll1 = polls2016[sample(1:nrow(polls2016), 500) ,]
poll2 = polls2016[sample(1:nrow(polls2016), 500), ]
propnormpolls1 = propnormdf(poll1, c(2,3))
propnormpolls2 = propnormdf(poll2, c(2,3))
addcategorytopolls(propnormpolls1, 30, 18, electdata)
newcutoff = c(-.15, -.1, -0.05, 0.05, .1, .15)
addcategorytopolls(propnormpolls2, 30, 18, electdata, cutoffs = newcutoff)

balexanderstats/bayesurvey documentation built on Sept. 20, 2020, 11:40 a.m.