Problem using srvyr package

General
1

Help everyone. I am currently trying to analyse a complex survey dataset, and I need to create a survey object in order to carry out weighted statistics. I am trying to use the srvyr package since it is easier to subsequently make plots with ggplot, but I am facing an error which I don't really know how to solve. Below Is my code:

if (!require(c("devtools","srvyr","tidyr","xtable","survey","readr","dplyr","plyr","ggplot2","readxl","tidyverse","svyPVpack","furrr","ggthemes","wesanderson"))) install.packages(c("devtools","srvyr","tidyr","xtable","survey","readr","dplyr","plyr","ggplot2","readxl","tidyverse","svyPVpack","furrr","ggthemes","wesanderson"))
library(c("devtools","srvyr","tidyr","xtable","survey","readr","dplyr","plyr","ggplot2","readxl","tidyverse","svyPVpack","furrr","ggthemes","wesanderson"))
packages <- c("devtools","srvyr","tidyr","xtable","survey","readr","dplyr","plyr","ggplot2","readxl","tidyverse","svyPVpack","furrr","ggthemes","wesanderson")
lapply(packages, library, character.only = TRUE)

setwd("~/Documents/Tesi/RECS/2015")
recs2015_public_v4 <- read_excel("recs2015_public_v4.xlsx")
recs=as.data.frame(recs2015_public_v4)
rm(recs2015_public_v4)

recs_design1=recs %>%
as_survey_rep(type="Fay",rho=0.5,repweights=starts_with("brrwt"), weights=nweight,combined.weights=FALSE)

p1<- recs_design1 %>%
group_by(hhsex) %>%
summarise(race = survey_mean(householder_race, na.rm=TRUE))

Error: Survey context not set
Call rlang::last_error() to see a backtrace.

Could you please give me an hint on how to solve this problem? Thanks you all!

2

Hi, welcome!

We don't really have enough info to help you out. Could you ask this with a minimal REPRoducible EXample (reprex)? A reprex makes it much easier for others to understand your issue and figure out how to help. In particular, we don’t have access to the data on your local computer.

1 Like
3

Hey! Thank you for replying. Below is the code I hope will help enough:

setwd("~/Documents/Tesi/RECS/2015")
recs2015_public_v4 <- read_excel("recs2015_public_v4.xlsx")
recs=as.data.frame(recs2015_public_v4)
rm(recs2015_public_v4)
attach(recs)
names(recs)=tolower(names(recs)) # All variables in lowercase
recs_design1=recs %>%

  • as_survey_rep(type="Fay",rho=0.5,repweights=starts_with("brrwt"), weights=nweight,combined.weights=FALSE)

recs_design1
Call: Called via srvyr
Fay's variance method (rho= 0.5 ) with 96 replicates.
Sampling variables:

  • repweights: brrwt1 + brrwt2 + brrwt3 + brrwt4 + brrwt5 + brrwt6 + brrwt7 + brrwt8 + brrwt9 + brrwt10 + brrwt11 + brrwt12 + brrwt13 + brrwt14 + brrwt15 + brrwt16 + brrwt17 + brrwt18 + brrwt19 + brrwt20 + brrwt21 + brrwt22 + brrwt23 + brrwt24 + brrwt25 + brrwt26 + brrwt27 + brrwt28 + brrwt29 + brrwt30 + brrwt31 + brrwt32 + brrwt33 + brrwt34 + brrwt35 + brrwt36 + brrwt37 + brrwt38 + brrwt39 + brrwt40 + brrwt41 + brrwt42 + brrwt43 + brrwt44 + brrwt45 + brrwt46 + brrwt47 + brrwt48 + brrwt49 + brrwt50 + brrwt51 + \n brrwt52 + brrwt53 + brrwt54 + brrwt55 + brrwt56 + brrwt57 + brrwt58 + brrwt59 + brrwt60 + brrwt61 + brrwt62 + brrwt63 + brrwt64 + brrwt65 + brrwt66 + brrwt67 + brrwt68 + brrwt69 + brrwt70 + brrwt71 + brrwt72 + brrwt73 + brrwt74 + brrwt75 + brrwt76 + brrwt77 + brrwt78 + brrwt79 + brrwt80 + brrwt81 + brrwt82 + brrwt83 + brrwt84 + brrwt85 + brrwt86 + brrwt87 + brrwt88 + brrwt89 + brrwt90 + brrwt91 + brrwt92 + brrwt93 + brrwt94 + brrwt95 + brrwt96
  • weights: nweight

svytotal(~hhsex,recs_design1)
total SE
hhsexFemale 66697844 895315
hhsexMale 51510406 895315

svytotal(~householder_race,recs_design1)
total SE
householder_raceWhite 95409173 1147801
householder_raceBlack or African/American 13606140 1095251
householder_raceAmerican Indian or Alaska Native 1297035 264046
householder_raceAsian 4945446 410667
householder_raceHawaiian or Pacific Island 435490 108994
householder_raceOther Race 0 0
householder_raceOr more race 2514967 252289

p1<- recs_design1 %>%

  • group_by(hhsex) %>%
  • summarise(race = survey_mean(householder_race, na.rm=TRUE))
    Error: Survey context not set
    Call rlang::last_error() to see a backtrace.

I don't understand why when I use commands from the "survey" package, I do not get errors; this suggests me that the survey design object has been correctly created. However, when I try to make statistics using the srvyr package, I get this error!

4

@mfherman is dead on about reprex but I can offer some insight.

When you see Error, look for the function immediately above it, in this case

summarise(race = survey_mean(householder_race, na.rm=TRUE))

and then open help(summarise) and run the example

library(srvyr)
#> 
#> Attaching package: 'srvyr'
#> The following object is masked from 'package:stats':
#> 
#>     filter
library(survey)
#> Loading required package: grid
#> Loading required package: Matrix
#> Loading required package: survival
#> 
#> Attaching package: 'survey'
#> The following object is masked from 'package:graphics':
#> 
#>     dotchart
data(api)

dstrata <- apistrat %>%
  as_survey_design(strata = stype, weights = pw)

dstrata %>%
  summarise(api99 = survey_mean(api99),
            api00 = survey_mean(api00),
            api_diff = survey_mean(api00 - api99))
#> # A tibble: 1 x 6
#>   api99 api99_se api00 api00_se api_diff api_diff_se
#>   <dbl>    <dbl> <dbl>    <dbl>    <dbl>       <dbl>
#> 1  629.     10.1  662.     9.54     32.9        2.08

Created on 2019-11-10 by the reprex package (v0.3.0)

The next question is why the difference? For that, look at the top of help(summarise) and help(survey_mean`) for the required arguments to those functions.

Finally, you have namespace conflicts between dplyr and srvyr for at least summarise, so use srvyr::sumarise as needed.

5

@technocrat Thank you! This is what I am getting after running the example you suggested me:

library(srvyr)
library(survey)
help(summarise)
data(api)
dstrata <- apistrat %>%
as_survey_design(strata = stype, weights = pw)
dstrata %>%
summarise(api99 = survey_mean(api99),
api00 = survey_mean(api00),
api_diff = survey_mean(api00 - api99))
Error: Survey context not set
Call rlang::last_error() to see a backtrace.

I still get the error even if using the example. Am I missing some packages or what? I don't manage to figure it out.

1 Like
6

Try restarting your R session and cut and pasting just the example code . If it still throws the error please post your sessionInfo()

sessionInfo()
#> R version 3.6.1 (2019-07-05)
#> Platform: x86_64-apple-darwin15.6.0 (64-bit)
#> Running under: macOS Mojave 10.14.6
#> 
#> Matrix products: default
#> BLAS:   /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRblas.0.dylib
#> LAPACK: /Library/Frameworks/R.framework/Versions/3.6/Resources/lib/libRlapack.dylib
#> 
#> locale:
#> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> loaded via a namespace (and not attached):
#>  [1] compiler_3.6.1  magrittr_1.5    tools_3.6.1     htmltools_0.4.0
#>  [5] yaml_2.2.0      Rcpp_1.0.3      stringi_1.4.3   rmarkdown_1.16 
#>  [9] highr_0.8       knitr_1.25      stringr_1.4.0   xfun_0.10      
#> [13] digest_0.6.22   rlang_0.4.1     evaluate_0.14

Created on 2019-11-10 by the reprex package (v0.3.0)

7

I restarted the session and only loaded those two packages along with the example, It worked.
I then tried to load packages one by one and I found that the problem is with the package plyr, which hides some functions, among which "summarise". That's why it wasn't working! Thank you a lot for suggesting me restarting R, that gave me the hint for trying that other way.

1 Like
8

Great. Please mark as solution for the benefit of anyone else who has the problem. Whenever there's a problem with namespace conflict among packages that are loaded, you can always call the one you want with srvyr::summarize

9

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.