I am trying to build a function in R that will allow me to generate weighted tables of named variables within in a data frame using the R survey package by Thomas Lumley. I think I am running into errors because the svytable and svychisq functions in the survey package are not recognizing the arguments in my homemade function as the names of the desired variables from my data frame I want analyzed.
I had tried to utilize the example given here to develop a solution: lapply with anonymous function call to svytable results in object 'x' not found
I also took a look at this post when trying to troubleshoot svychisq: Error using dynamic variable specification in R survey function svychisq()
None of the suggestions from these posts have worked for me to date. I would very much appreciate if people could look through my code and give feedback for what I could try to generate a function that prints 1. a weighted table of totals 2. the results of running a chi square analysis on that weighted table.
Here is my data/code to date:
#data to make the code reproducible
independent <- c(2,1,1,1,1,1,2,2,1,1,2,2,2,2,1,1,1,2,2,2,2,2,2,1,1,1,2,1,2,1,1,1,2,2,1,1,2,1,1,2,1,2,1,1,1,1,1,1,1,2,1,2,1,2,2,2,2,1,1,1,1,2,2,1,1,2,2,1,1,1,1,2,1,2,1,1,1,1,2,1,2,1,2,2,1,1,1,2,2,1,2,2,2,1,2,1,2,2,2,2,2,1,1,2,2,1,2,1,2,2,2,1,1,2,1,1,2,1,1,1,2,1,2,1,1,2,1,2,2,2,1,1,1,1,1,1,2,2,1,1,1,2,2,1,1,2,2,2,2,1,2,1,2,2,2,2,2,2,2,2,2,2,2,2,2,1,1,2,2,1,2,2,1,1,2,2,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,1,2,2,1,2,1,1,2,1)
dependent <-c(1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,3,1,1,2,1,1,1,2,1,98,1,1,1,1,1,1,2,1,1,1,1,1,1,2,1,1,1,1,1,2,2,1,1,1,1,3,1,1,2,1,2,1,2,1,1,1,1,1,1,2,1,1,1,1,1,2,1,1,1,4,1,1,1,1,1,1,1,1,1,1,1,1,1,1,4,1,1,1,5,1,2,1,1,1,1,1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,1,2,1,1,5,1,2,1,1,1,2,1,1,1,1,2,1,1,1,2,1,1,1,2,1,4,3,1,3,1,1,2,1,1,1,1,1,1,1,2,1,1,1,1,1,1,2,1,1,2,1,2,2,1,2,1)'
weight <- c(2937,9389,7918,3851,8560,7433,6577,4507,5601,1358,7274,4507,5160,2429,618,4938,3734,5177,3683,3988,3534,2364,1979,5879,5131,6119,4321,4425,1302,3563,5133,10947,2183,3988,6607,5637,4507,5598,3086,6341,2553,6040,15050,809,5056,5709,2557,3734,1136,2596,7029,4507,2099,1296,4103,4030,4944,3940,4358,5598,1794,4636,6130,6119,431,4944,8361,5732,2602,6117,3719,2727,1406,6023,15050,3870,6808,3957,4507,3827,3658,7600,4244,4507,1850,4358,3770,4103,4927,4801,4012,2364,5547,4381,4103,4444,1427,5575,4656,3708,2865,4447,5709,477,2617,6232,3160,2745,3103,1604,2925,3950,1679,1341,7420,7600,1655,6307,4041,3882,5058,6412,4332,3415,7913,4330,5604,4027,305,4933,4635,1224,4464,8988,4664,1143,882,4061,6142,5763,9811,882,580,780,6973,1752,4263,3809,3440,4447,4740,443,3438,4076,3948,3438,2429,3582,4397,4283,4399,2028,758,15944,3593,4430,4159,4292,5976,3619,18834,2301,4485,942,5861,4284,2102,2102,8503,942,4747,8503,4358,4348,1794,4783,1899,373,5811,2183,5575,4702,4466,4653,3216,4553,7523,2606,2396,3216)
psu <- c(108,167,224,185,167,151,294,187,274,111,161,187,286,179,228,145,108,248,240,267,253,109,179,240,228,172,278,165,129,153,271,161,141,267,243,278,187,168,108,250,179,194,210,278,114,240,189,108,287,189,154,187,141,251,108,108,187,142,145,168,143,253,172,172,178,187,165,187,267,172,177,165,167,186,210,107,168,108,187,167,186,273,187,187,179,145,174,108,115,115,215,109,253,120,108,240,123,150,115,115,278,174,240,142,109,159,187,185,176,140,220,166,129,129,240,273,111,161,141,141,121,115,120,175,168,472,472,142,142,145,147,145,150,132,186,166,166,169,161,169,163,166,166,166,163,179,178,176,177,174,180,178,171,180,171,171,179,173,192,192,157,154,151,160,154,142,186,186,163,171,160,141,142,143,143,148,141,141,143,143,150,143,145,147,143,141,141,142,148,141,150,141,148,148,136,134,140,136,136,136)
#make the data frame for reproducible example
dat <- as.data.frame(cbind(independent,dependent,weight,psu))
dat$independent <- factor(dat$independent, levels=c("1", "2"))
dat$dependent <- factor(dat$dependent, levels = c("1", "2", "3", "4", "5", "98"))
str(dat)
#Function to print svytables + chi square statistic
library(survey) #opens the survey package in R
x <- independent
y <- dependent
cstbl <- function(x,y) {
dat_w <- svydesign(id=~psu, weights=~weight, data=dat)
tbl <- svytable(bquote(~.(as.name(y)+as.name(x))), dat_w) #should make table
chi <- svychisq(as.formula(paste("~", x, "+", y)), dat_w) #should give test stat
final <- list(tbl, chi)
return(final) #should return table + chi square test stat
}
cstbl(x,y)
This throws the error: "Error in as.name(y) + as.name(x) : non-numeric argument to binary operator"
I get the same error if I replace the third line of the function with
tbl <- svytable(~y+x, dat_w)
Your guidance and patience are much appreciated, as I am new to programming.
