What is Validity?
Validity is how well a survey or question measures what it is intended to measure. A valid survey produces answers you can trust to represent the concept you care about.
Validity describes whether a question, scale or whole survey accurately captures the idea it aims to measure (for example, satisfaction, need, or understanding). It is not about whether the survey is consistent (that is reliability) but about whether the survey items genuinely reflect the underlying concept. Common forms of validity are: face or content validity (items look appropriate and cover the topic), construct validity (items behave as expected in relation to other measures), and criterion validity (survey results match an external standard). Threats to validity include unclear wording, leading questions, cultural differences in interpretation, translation errors and sampling problems.
Usage example
A council asks residents to rate 'trust in local services.' If the translated question uses a word that implies 'satisfaction' rather than 'trust' in another language, responses will not reflect trust but a different concept — the question has low validity in that language. Testing the wording and reviewing participant feedback helps restore validity.
Practical application
Validity matters because decisions (policy, service changes, funding, referrals) rely on what respondents actually mean. In multilingual surveys, validity can break down when translations change nuance or when a concept is unfamiliar across cultures. Improving validity saves time and reduces the risk of acting on misleading results. Practical steps to protect validity include writing clear, simple questions; piloting and cognitive testing in each language; using consistent, reviewed translations; monitoring response patterns; and incorporating participant feedback to refine wording. Hearo’s built-in translation, participant flagging and admin review workflow make these steps easier by allowing one source survey to be tested and improved across languages without maintaining separate forms.
FAQ
How is validity different from reliability?
Reliability is about consistency — whether the same question gives similar results if repeated. Validity is about accuracy — whether the question measures what you intend. A measure can be reliable but not valid (consistent but wrong), and both are needed for trustworthy results.
Can automatic translation harm validity?
Automatic translation can introduce changes in meaning that reduce validity, especially for nuanced or culturally specific language. That risk is managed by reviewing translations, piloting with target communities, allowing participants to flag problematic wording, and making iterative edits — workflows that Hearo is designed to support.
How can I check validity for a multilingual survey?
Test questions with small groups in each language (cognitive interviews), compare response patterns across languages, review open-text answers translated back into your language, use back-translation or expert review for critical items (consent, eligibility), and collect participant feedback on wording. Fix and re-test rather than maintaining separate untranslated versions.
Do I need professional translators to ensure validity?
Professional translators help, particularly for legally sensitive or highly technical text, but they are not always required. A practical approach is to use machine translation to reach many languages quickly, then involve bilingual community reviewers or subject experts to check and refine wording, plus participant feedback to catch remaining issues.