Join us in the essential fight against survey fraud, tackling each response with precision and integrity. By validating the quality of every submission, we not only elevate the reliability of data but also ensure that every individual’s voice is heard and respected.
- Click Browse Files to use a CSV file of your open-ended responses. The upload size limit is 200MB per file. Find your local file and upload it for analysis.
- Select the column containing unique user IDs you want to analyze.
- Select the column containing open-ended responses you want to analyze.
- Initiate the analysis by clicking the Run Quality Checks button to identify responses that fall below your quality criteria. The data quality agent runs the following checks for each defined open-end:
- Gibberish—Identifies responses that lack coherent content.
- No Response/NA—Responses that serve as fillers, such as "None," "N/A," "No," or blank spaces.
- Profanity—Profane responses are those that include profane, vulgar, or explicit language.
- AI Detection—Assesses whether a response is likely generated by a large language model (LLM), such as GPT by analyzing specific linguistic characteristics.
- Off Topic- Responses that are unrelated to the question.
- Review the flagged responses to identify low-quality ones and decide on next steps, like re-surveying participants or excluding them. Document your analysis and the reasons for flagging specific responses for future reference.