<!-- README.md is generated from README.Rmd. Please edit that file -->

srvyr <img src="man/figures/logo.png" align="right" height="149" width="149"/>

<!-- badges: start -->

<!-- badges: end -->

srvyr brings parts of dplyr’s syntax to survey analysis, using the survey package.

srvyr focuses on calculating summary statistics from survey data, such as the mean, total or quantile. It allows for the use of many dplyr verbs, such as summarize, group_by, and mutate, the convenience of pipe-able functions, rlang’s style of non-standard evaluation and more consistent return types than the survey package.

You can try it out:

install.packages("srvyr")
# or for development version
# remotes::install_github("gergness/srvyr")

Example usage

First, describe the variables that define the survey’s structure with the function as_survey()with the bare column names of the names that you would use in functions from the survey package like survey::svydesign(), survey::svrepdesign() or survey::twophase().

library(srvyr, warn.conflicts = FALSE)
data(api, package = "survey")

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

Now many of the dplyr verbs are available.

• mutate() adds or modifies a variable.

dstrata <- dstrata %>%
  mutate(api_diff = api00 - api99)

• summarise() calculates summary statistics such as mean, total, quantile or ratio.

dstrata %>% 
  summarise(api_diff = survey_mean(api_diff, vartype = "ci"))
#> # A tibble: 1 × 3
#>   api_diff api_diff_low api_diff_upp
#>      <dbl>        <dbl>        <dbl>
#> 1     32.9         28.8         37.0

• group_by() and then summarise() creates summaries by groups.

dstrata %>% 
  group_by(stype) %>%
  summarise(api_diff = survey_mean(api_diff, vartype = "ci"))
#> # A tibble: 3 × 4
#>   stype api_diff api_diff_low api_diff_upp
#>   <fct>    <dbl>        <dbl>        <dbl>
#> 1 E        38.6         33.1          44.0
#> 2 H         8.46         1.74         15.2
#> 3 M        26.4         20.4          32.4

• Functions from the survey package are still available:

my_model <- survey::svyglm(api99 ~ stype, dstrata)
summary(my_model)
#> 
#> Call:
#> svyglm(formula = api99 ~ stype, design = dstrata)
#> 
#> Survey design:
#> Called via srvyr
#> 
#> Coefficients:
#>             Estimate Std. Error t value Pr(>|t|)    
#> (Intercept)   635.87      13.34  47.669   <2e-16 ***
#> stypeH        -18.51      20.68  -0.895    0.372    
#> stypeM        -25.67      21.42  -1.198    0.232    
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> (Dispersion parameter for gaussian family taken to be 16409.56)
#> 
#> Number of Fisher Scoring iterations: 2

Cheat Sheet

<a href="https://github.com/gergness/srvyr/blob/main/cheatsheet/srvyr.pdf"><img src="man/figures/srvyr-cheatsheet-preview.png" width="630" height="252"/></a><br>

Learning more

Here are some free resources put together by the community about srvyr:

• “How-to”s & examples of using srvyr
  • srvyr’s included vignette “srvyr vs survey” and the rest of the pkgdown website
  • Stephanie Zimmer, Rebecca Powell and Isabella Velásquez’s book Exploring Complex Survey Data Analysis Using R (releasing in November 2024). See also their 2021 AAPOR Workshop “Tidy Survey Analysis in R using the srvyr Package”
  • “The Epidemiologist R Handbook”, by Neale Batra et al. has a chapter on survey analysis with srvyr and survey package examples
  • Kieran Healy’s book “Data Visualization: A Practical Introduction” has a section on using srvyr to visualize the ESS.
  • The IPUMS PMA team’s blog had a series showing examples of using the PMA COVID survey panel with weights
  • “Open Case Studies: Vaping Behaviors in American Youth” by Carrie Wright, Michael Ontiveros, Leah Jager, Margaret Taub, and Stephanie Hicks is a detailed case study that includes using srvyr to analyze the National Youth Tobacco Survey.
  • The tidycensus package vignette “Working with Census microdata” includes information about using the weights from the ACS retrieved from the census API.
  • “The Joy of Calculating the Direct Standard Error for PUMS Estimates” by GitHub user @ldaly
• About survey statistics
  • Thomas Lumley’s book “Complex Surveys: a guide to analysis using R”
  • Chris Skinner. Jon Wakefield. “Introduction to the Design and Analysis of Complex Survey Data.” Statist. Sci. 32 (2) 165 - 175, May 2017. 10.1214/17-STS614
  • Sharon Lohr’s textbook “Sampling: Design and Analysis”. Second or Third Editions
  • “Survey weighting is a mess” is the opening to Andrew Gelman’s “Struggles with Survey Weighting and Regression Modeling”
  • Anthony Damico’s website “Analyze Survey Data for Free” has the weight specifications for a wide variety of public use survey datasets.
• Working programmatically and/or on multiple columns at once (eg dplyr::across and rlang’s “curly curly” {{}})
  • dplyr’s included package vignettes “Column-wise operations” & “Programming with dplyr”
• Non-English resources
  • Em português: “Análise de Dados Amostrais Complexos” by Djalma Pessoa and Pedro Nascimento Silva
  • En español: “Usando R para jugar con los microdatos del INEGI” by Claudio Daniel Pacheco Castro
  • Chapter 26 of the The Epidemiologist R Handbook, translated:
    • En français: Analyse d’enquête
    • Tiếng Việt: Phân tích khảo sát
    • En español: Análisis de encuestas
    • 日本語で: 標本調査データ分析
    • Em português: Analises de pesquisa de questionários (survey)
    • Türkçe: Anket analizi
    • На русском языке: Анализ опросов
  • På norsk: Data med vekter i R by Øyvind Bugge Solheim
• Other cool stuff that uses srvyr
  • A (free) graphical interface allowing exploratory data analysis of survey data without writing code: iNZight (and survey data instructions)
  • “serosurvey: Serological Survey Analysis For Prevalence Estimation Under Misclassification” by Andree Valle Campos
  • Several packages on CRAN depend on srvyr, you can see them by looking at the reverse Imports/Suggestions on CRAN.

Still need help?

I think the best way to get help is to form a specific question and ask it in some place like posit’s community website (known for it’s friendly community) or stackoverflow.com (maybe not known for being quite as friendly, but probably has more people). If you think you’ve found a bug in srvyr’s code, please file an issue on GitHub, but note that I’m not a great resource for helping specific issue, both because I have limited capacity but also because I do not consider myself an expert in the statistical methods behind survey analysis.

Have something to add?

These resources were mostly found via vanity searches on twitter & github. If you know of anything I missed, or have written something yourself, please let me know in this GitHub issue!

What people are saying about srvyr

1. Yay!

–<cite>Thomas Lumley, in the Biased and Inefficient blog</cite>

Contributi