What is Response Quality Score?
A Response Quality Score is a numeric indicator that summarises how useful, complete and trustworthy an individual survey response is. It helps teams prioritise, filter and weight responses for analysis and follow-up.
Response Quality Score (RQS) combines several objective and inferred signals about a single survey submission into one score. Common signals include completeness (are required questions answered), timeliness (reasonable time spent on the survey), relevance (answers fit expected formats or choices), length and substance of open-text replies, sentiment or coherence of free-text, and automated checks for bot-like behaviour or duplicates. In Hearo, RQS can also include translation-related signals — for example, machine-translation confidence, whether a participant flagged wording, or whether an admin has reviewed and adjusted translated text. Scores are typically normalised to a range (for example 0–1 or 0–100) and can be calculated with simple rules or with a trained model that weights signals according to their importance.
Usage example
After a community consultation, the engagement officer filters incoming responses to show only those with a Response Quality Score above 60 so the team can quickly review substantive feedback. They also sort by average RQS by language to identify languages where translation confidence is low and needs review.
Practical application
RQS matters because it speeds up analysis and improves data reliability. Instead of reading every response, teams can prioritise high-quality submissions for reporting and flag low-scoring entries for follow-up or exclusion. In multilingual projects, RQS helps surface responses where translation confidence is low or participants flagged wording, so administrators can decide whether to re-translate, ask for clarification, or train the translation model. It also supports sampling and weighting strategies: you can weight or exclude low-quality responses to reduce noise, allocate human review time where it will do the most good, and monitor whether changes to a survey (question wording, length, language options) affect the quality of answers.
FAQ
How is the Response Quality Score calculated?
There isn’t a single industry standard. RQS is usually built from multiple signals — completeness, response length, time taken, answer relevance, translation confidence, duplicate detection, and participant flags — combined by weighted rules or a machine-learning model. Hearo’s implementation can include translation-specific metrics (e.g. confidence score from the translator and whether a participant flagged wording) and allows teams to adjust weights for their context.
Should I remove low-scoring responses from my analysis?
Not automatically. Low RQS often indicates noise (bots, accidental submissions, or one-word answers) and can be excluded or down-weighted, but sometimes low scores reflect underserved respondents who gave brief but important feedback. It’s best to review a sample of low-scoring entries and set exclusion or weighting rules that match your project goals.
Can people game the Response Quality Score?
Some signals (like time spent or answer length) can be manipulated, so robust RQS systems combine multiple checks — format validation, duplicate detection, semantic relevance, and translation confidence — to reduce gaming. Periodic audits and manual reviews help catch patterns of manipulation and refine scoring rules.