Who can use this feature?
Available on:
All Culture Amp subscriptions that include Engagement or Effectiveness.
When conducting a survey, Culture Amp offers two types of raw data extracts: de-identified data extracts and full raw data extracts. Understanding which one to use and how to access them is important for preserving confidentiality and meeting your data analysis needs.
Types of data extracts
De-identified data extract:
This type of data extract removes any personally identifiable information to ensure confidentiality. It’s the standard option for surveys.
Confidentiality: No identifiable information like Email or Employee ID is included.
One extract limit: Only one de-identified data extract is available per survey.
Excel format: Delivered as a password-protected Excel file in a .ZIP format.
Turnaround time: Delivered within a week of survey closure upon request.
Restriction: Multiple files with different demographics are not provided.
Access permissions: The file can only be shared with a Survey data analyst
What’s included:
Survey responses (e.g., a 4 out of 5 rating will display as “4”).
One demographic of your choice (e.g., department), provided that it meets the minimum group reporting threshold.
Skipped questions are marked as ‘-1’.
Multi-select question choices appear in a single cell.
Exclusions: No comments or submission timestamps are included.
Full raw data extract:
The full raw data extract includes identifiable information such as Email or Employee ID, along with all submitted survey responses. This option is useful for teams that require more detailed data for advanced analysis
Identifiable data: Includes all enabled demographics, submitted responses and identifiable information.
Advanced analysis: Useful for data professionals who need to integrate the data with other datasets.
Export formats: If enabled, data is available in .csv format through the survey Operations page or via the Survey API in JSON format.
Key consideration: Full transparency with participants is required when collecting this data, as it may impact participation rates. Warnings for participants are integrated into the survey's reporting rules when this option is enabled.
What’s included:
Submitted survey responses with identifiers (e.g., Email or Employee ID).
All demographics (e.g., department, location) enabled on the survey.
Participant comments and their associated responses.
Guidelines for using full raw data extracts
While a full raw data extract can provide comprehensive insights, it’s important to carefully consider confidentiality and data quality to maintain the integrity of your survey.
Best practices:
Enable raw data before survey launch: Ensure the full raw data extract is enabled before launching your survey. This option cannot be enabled after launch, and Culture Amp's support team cannot provide raw data extracts if participants weren’t informed.
Survey transparency: Clearly communicate to participants that identifiable data will be collected. Warnings are integrated into the survey's reporting rules, visible on the survey Welcome screen. We also recommend outlining your data analysis approach in advance and explaining why certain individuals may have access to raw data. Transparency is crucial.
Build trust: Explain to your participants how their data will be used to foster trust and encourage honest, meaningful feedback.
Data security: Only authorized personnel, such as Survey data analysts, should have access to the raw data extract.
FAQs
Why would I need a full raw data extract?
Why would I need a full raw data extract?
Full raw data extracts are ideal for detailed analysis, such as integrating data into a warehouse or combining it with other datasets. Data professionals like data scientists often find it valuable for research purposes.
Is a full raw data extract right for my organization?
Is a full raw data extract right for my organization?
Consider the level of transparency needed for your survey. Running a fully transparent survey depends heavily on your workforce's trust. If respondents are concerned about confidentiality, this could lead to low participation or even inaccurate data due to fake answers. High data quality is essential, especially when making decisions based on survey responses.
Related pages:
💬 Need help? Just reply with "Ask a Person" in a support conversation to speak with a Product Support Specialist.