High Frequency Checks
Learn about IPA’s High Frequency Checks—systematic data quality checks performed during data collection to identify and address issues early. This how-to guide covers the types of checks, implementation using the ipacheck Stata package, and best practices for ensuring reliable survey data.
- High Frequency Checks ensure data quality by identifying and addressing issues during data collection.
- The
ipacheck
Stata package provides a comprehensive set of tools to streamline HFC implementation.
What are High Frequency Checks?
High Frequency Checks are systematic checks performed on survey data at regular intervals, such as daily or weekly, during data collection. These checks help identify and correct issues or mistakes in the survey process, ensuring the data collected is of high quality. By catching errors early, teams can make necessary adjustments to the survey or data collection methods, leading to more reliable and accurate data.
Why it is important to implement High Frequency Checks
High-frequency checks evaluate different aspects of the data collection process and run at regular intervals as teams collect new data. At IPA/J-PAL, HFCs are typically implemented in Stata after the data flow is complete. These checks provide insights into:
- The quality of the data
- Enumerator performance
- Errors in the electronic survey program
- Systematic flaws in the data flow
Given their importance, high-frequency checks are one of the major benefits of Computer-Assisted Interviewing (CAI). Unlike CAI logic checks, which are pre-programmed into the survey tool, HFCs run post-data collection to identify trends across surveys.
Types of High Frequency Checks
Daily Logic Checks
- Verify the survey form version
- Detect duplicate observations and unique variable duplicates—for example, GPS, phone number
- Ensure critical variables are not missing
- Identify variables with all missing values
- Review “Other specify” values
- Detect outliers in numeric variables
- Review field comments
- Track survey progress
Enumerator Performance Checks
- Percentage of “Don’t know” and “Refuse to answer”
- “Yes” percentage for filter questions
- Enumerator productivity
- Average interview duration
- Active hours
- Statistics for numeric variables
Survey Dashboard Checks
- Survey consent rate
- Percentage of missing survey values
- Percentage of “Don’t know” and “Refuse to Answer”
- Number and percentage of “Other specify” values
- Variables with all missing values
- Survey productivity
Introduction to IPA’s HFCs Package
The Stata program ipacheck
is a set of user-written commands developed and maintained by IPA. It consists of three types of files: an input file, a do-file, and output files. Teams can modify these files to fit their project’s needs. Each check produces either a standalone workbook or a dedicated sheet in the HFC output workbook, displaying respondent IDs, enumerator details, and relevant flagged variables.
How are High Frequency Checks different from the Data Management System?
The Data Management System provides a structured approach to managing research data, assisting with:
- Creating project folder structures
- Managing data flow
- Running high-frequency checks
- De-duplicating and replacing data
- Tracking survey progress
- Running back checks
- Generating logs for project closure
High-frequency checks are just one component of the broader capabilities of the DMS, running alongside other project management tools.
Installation
ipacheck
is IPA’s Stata package for running high-frequency checks. The package includes various programs categorized as main or ancillary.
Main Programs
ipacheckcorrections
- Make corrections to data.ipacheckspecifyrecode
- Recode “Other specify” values.ipacheckversions
- Export survey form version statistics and flag outdated submissions.ipacheckids
- Identify duplicate survey IDs.ipacheckdups
- Detect duplicates in non-ID variables.ipacheckmissing
- Generate missingness and distinctness statistics.ipacheckoutliers
- Identify outliers in numeric variables.ipacheckspecify
- Extract “Other specify” values.ipacheckcomments
- Export field comments from SurveyCTO.ipachecktextaudit
- Analyze survey duration using SurveyCTO text audit files.ipachecktimeuse
- Export engagement statistics from SurveyCTO audit files.ipachecksurveydb
- Generate general dataset statistics.ipacheckenumdb
- Evaluate enumerator performance.ipatracksurvey
- Generate a dashboard for tracking survey progress.ipabcstats
- Compare survey data with back check data.
Ancillary Programs
ipacodebook
- Export a codebook to Excel with labeled variable notes.
ipacheck
comes with a structured project folder setup, including a master do-file, global variables file, preparation do-file, and Excel-based input sheets. The package generates outputs as well-formatted Excel spreadsheets for field team distribution. Since version four, ipacheck
incorporates programs from the ipahelper
package.
Installation Steps
* Install ipacheck from GitHuball replace from("https://raw.githubusercontent.com/PovertyAction/high-frequency-checks/master")
net install ipacheck,
* Update ipacheck anytime
ipacheck update
new project with folder structure and input files
* Create a new, surveys("SURVEY_NAME_1") folder("path/to/project")
ipacheck
using subfolders
* Create a project with multiple surveys new, surveys("SURVEY_NAME_1" "SURVEY_NAME_2") folder("path/to/project") subfolders
ipacheck
of the master do-file and Excel inputs
* Obtain fresh copies new, filesonly
ipacheck
sample data and instructions)
* Access IPA's training exercise (includes new, exercise
ipacheck
version
* Verify the installed version ipacheck
Check out the IPA High Frequency Check Exercise! These exercises help familiarize users with the setup and use of IPA’s Data Management System, which includes High-Frequency Checks, Survey Tracking, and Back Check comparison. Learn more
Conclusion
<<<<<<< HEAD <<<<<<< HEAD In conclusion, High Frequency Checks (HFCs) are an essential component of IPA’s data quality assurance framework. By implementing systematic and regular checks on survey data, HFCs help ensure the accuracy, reliability, and overall quality of the data collected. The ipacheck
Stata package provides a comprehensive set of tools to facilitate these checks, making it easier for research teams to identify and address issues promptly. By integrating HFCs into your data collection process, you can significantly enhance the integrity of your research findings and contribute to the overall success of your projects.
======= >>>>>>> c6c7bfe (edits to data quality pages) ======= >>>>>>> ab5cde4 (edits to data quality pages) High Frequency Checks are essential for maintaining data quality during collection. The ipacheck
Stata package provides comprehensive tools to implement these checks systematically, helping research teams identify issues early and ensure reliable results.