Last active
September 6, 2015 18:30
-
-
Save mjrich/871fd20b7f3831fbb8ba to your computer and use it in GitHub Desktop.
This is an example script for how to compute weights in base R using demographic variables like age & gender and then how to calculate frequencies while applying those weights
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# | |
#You have a dataframe "df", with variables gender and age, and other variables of interest about radio listenership | |
# | |
#Add weights below. The population parameter over the prop.table(..) portion calculating sample statistics. | |
weight_male = .49500764/(prop.table(table(df$Gender))["Male"]) | |
weight_male | |
weight_female = .50499236/(prop.table(table(df$Gender))["Female"]) | |
weight_female | |
weight_15to24 = .31/prop.table(table(df$agecat))["15 to 24"] | |
weight_15to24 | |
weight_25to34 = .25/prop.table(table(df$agecat))["25 to 34"] | |
weight_25to34 | |
weight_35plus = .45/prop.table(table(df$agecat))["35+"] | |
weight_35plus | |
#Multiply weights to achieve total weight for each record in your dataset | |
df$weight = 1.0 | |
df$weight[df$Gender == "Male"] = sapply( | |
df$weight[df$Gender == "Male"], function(x) x*weight_male) | |
df$weight[df$Gender == "Female"] = sapply( | |
df$weight[df$Gender == "Female"], function(y) y*weight_female) | |
df$weight[df$agecat == "15 to 24"] = sapply( | |
df$weight[df$agecat == "15 to 24"], function(x) x*weight_15to24) | |
df$weight[df$agecat == "25 to 35"] = sapply( | |
df$weight[df$agecat == "25 to 35"], function(x) x*weight_25to34) | |
df$weight[df$agecat == "35+"] = sapply( | |
df$weight[df$agecat == "35+"], function(x) x*weight_35plus) | |
#turn back into vector | |
df$weight = unlist(df$weight) | |
first_weight = sum(df$weight) | |
table(df$weight) | |
first_weight | |
#plus one last rebalancing to bring the N back to the total N, if needed | |
df$weight = sapply(df$weight, function(x) x*nrow(df)/first_weight) | |
#turn back into vector | |
df$weight = unlist(df$weight) | |
#these two should now be the same | |
sum(df$weight) | |
table(df$weight) | |
#Weighted frequency function | |
table_wght<- function (x){ | |
z=df$weight | |
N<-xtabs(z~x) | |
return(N) | |
} | |
#Radio listenership analysis | |
table(df$RadioHr2) | |
table(df$RadioHr4) | |
table_wght(df$RadioHr2) | |
table_wght(df$RadioHr4) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment