Iefieldkit

Revision as of 19:33, 5 February 2019 by Bbdaniels (talk | contribs) (Created page with "== iefieldkit == ===Summary=== iefieldkit provides a set of commands that enable a reproducible primary data collection and cleaning workflow. The package is developed to fa...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

iefieldkit

Summary

iefieldkit provides a set of commands that enable a reproducible primary data collection and cleaning workflow. The package is developed to facilitate a workflow including (1) data collection (in particular using opendatakit.org, more specifically surveycto.com); (2) basic data cleaning, such as labeling and recoding; (3) reconciling survey rounds; (4) preparing codebooks to document data sets. iefieldkit was developed to standardize and simplify best practices for high-quality primary data collection across the World Bank's Development Research Group Impact Evaluations team (DIME). The commands can also be used independently, and are developed to be applicable to many other contexts as well. See https://github.com/worldbank/iefieldkit for more details, or read the DIME Wiki entries for:

- ietestform

- ieduplicates

- iecodebook

Details

The iefieldkit package is a set of commands designed to simplify a series of tedious and repetitive tasks for Stata users who are in the process of collecting primary survey data in the field. This package currently supports three major components of that workflow: survey design; survey completion; and data cleaning and survey harmonization.

One of the most important developments in economics research over the past two decades has been the rise of empirical data collection, especially with unique primary datasets collected by the researchers themselves. The authors of iefieldkit have supported the implementation of a wide range of primary data collection in fields including agriculture, health, energy and environment, edutainment, financial and private sector development, fragility, conflict, and violence, gender, governance, and transport. They have developed workflows to support general best practices for data collection, and as a rule develop new packages only when they fill an essential gap in Stata functionality. The packages here are a first attempt to provide Stata-based tools for managing the primary data collection process using native tools from start to finish.

Specifically, iefieldkit performs three essential tasks. Before data collection occurs, iefieldkit allows for rapid error-checking of ODK-based electronic surveys, including best practices for SurveyCTO-styled forms. This ensures that data, once collected, will import in Stata-friendly formats -- such as avoiding name conflicts and ensuring compliant variable naming and labelling. While data collection is ongoing, ieduplicates and iecompdup provide a workflow for detecting and resolving duplicate entries in the dataset, ensuring that the final survey dataset will be a correct record of the survey sample to merge onto the master sampling database. Finally, once data collection is complete, the iecodebook commands provide a workflow for rapidly cleaning, harmonizing, and documenting datasets.

All three commands utilize spreadsheet-based workflows so that their inputs and outputs are significantly more human-readable than Stata dofiles completing the same tasks would be, and these tasks can be supported and reviewed by personnel who specialize in field work rather than code tools. The increasing diversity and specialization of research teams has made accessibility to non-Stata-proficient personnel an essential component of data management workflows, and the iefieldkit package takes this development seriously. The code is also open-source and available for public contribution and comment on GitHub at https://github.com/worldbank/iefieldkit.