KOF Working Papers, KOF Swiss Economic Institute, ETH Zurich 326
This paper describes how to use the R package gateveys to establish a transparent and reproducible aggregation work flow for longitudinal data stemming from business tendency surveys (BTS). Business tendency survey researchers are addressed in particular though the suggested work flow could also be applied to other processes that generate categorical data. The package has two main features: First, it provides functions to build an aggregation process that re-calculates all periods when a new survey wave is added and hence can be fully reproduced at any later stage. Second, the package can be used to dynamically add localized meta information to the resulting time series object during the aggregation process. Besides, the paper suggests a software architecture for use of the package in a scenario with regular, periodical survey waves.
R aggregation survey data qualitative survey categorical data business tendency surveys time series meta data meta information software architecture reproducible research