Applications of R Shiny to Explore, Evaluate and Improve Total Survey Quality

Abstract

Maintaining and assessing total survey quality on a large scale and complex survey such as the National Resource Inventory (NRI) often involves iterative human interaction. Tools such as R Shiny that allow graphical displays with immediate user interactions turn out to be practical and useful. This paper introduces two web-based applications we developed, viscover and iNtr. viscover is an interactive tool to visualize soil survey geographic data, cropland data and their overlay. Its primary purpose is as a diagnostic tool to assess the accuracy of an overlay operation needed to define the covariates for an unit-level small area model. It is freely available through https://lyux.shinyapps.io/viscover. iNtr is an interactive NRI table review tool that has been helping the 2015 NRI make exhaustive comparisons among different estimation runs. To avoid violating the privacy policy of NRI, we are here presenting a modified version of the tool that allows a comparison of the released 2015 and 2012 NRI estimation results. The adapted version is available at https://lyux.shinyapps.io/table_review/.

Date
Jun 11, 2019 4:30 PM — 5:00 PM
Location
Bergamo, Italy
Annie Lyu
Annie Lyu
Principal Data Scientist

rstats, dataviz, she/her