Description
Data Management of Large Hierarchical Data.
Description
Import 'SPSS' data, handle and change 'SPSS' meta data, store and access large hierarchical data in 'SQLite' data bases.
README.md
eatGADS
Overview
eatGADS
(educational assessment tools: Greatest Assumable Data Set) is the data management and data handling tool used by the Institute for Educational Quality Improvement in Germany (IQB). It has a strong focus on (automated) handling and processing of meta data.
eatGADS
uses SQLite3 as a back end and is especially suited for importing data stored as SPSS
files. eatDB is used for data base creation and use, haven is used for importing SPSS files.
Installation
# Install stable version from CRAN via
install.packages("eatGADS")
# Install development version from GitHub via
remotes::install_github("beckerbenj/eatGADS")
Usage
The functionality of eatGADS
is extensively documented in its various vignettes. If you have questions regarding existing functionality or requests for new features, contact the package author.
# import SPSS data
sav_path <- system.file("extdata", "pisa.zsav", package = "eatGADS")
gads_obj <- import_spss(sav_path)
# see names of variables in the data set
namesGADS(gads_obj)
# extract meta data for specific variables
extractMeta(gads_obj, vars = c("schtype", "idschool"))
# extract data for data analysis while applying missing tags and value labels
dat1 <- extractData(gads_obj, convertLabels = "character")