> For the complete documentation index, see [llms.txt](https://help.metaforms.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.metaforms.ai/open-end-cleaning-and-coding/open-end-response-quality-checks.md).

# Open-end response quality checks

Metaforms automatically reviews every open-ended response against a suite of quality checks before it enters the coding workflow. Each response is classified, scored for confidence, and surfaced to the reviewer for action: keep, junk, edit, or translate.

The checks performed today:

**1. Response Length Check**\
Flags responses too short to carry meaningful information.\
\&#xNAN;*Example:* "yes" in answer to "What do you like most about our service?"

**2. AI-Generated Response Check**\
Flags responses that appear written by a chatbot or large language model rather than a human respondent.\
\&#xNAN;*Example:* "There are several compelling reasons this product stands out in today's competitive marketplace, including but not limited to its premium build quality and thoughtful design choices…"

**3. Off-Topic Response Check**\
Flags responses unrelated to the question being asked and to the broader research subject.\
\&#xNAN;*Example:* "I had pasta for lunch" in answer to "How would you describe your bank?"

**4. Question Repetition Chec**k\
Flags responses that restate or paraphrase the question itself instead of answering it.\
\&#xNAN;*Example:* "What I like about your service" in answer to "What do you like about our service?"

**5. Cross-Question Duplicate Check**\
Flags when the same respondent gave an identical answer to multiple different open-ended questions in the same survey.\
\&#xNAN;*Example:* Answering "great service" to every open-ended question across a six-question feedback survey.

**6. Cross-Question Similarity Check**\
Flags when the same respondent gave near-identical (but not exactly matching) answers to different open-ended questions in the same survey.\
\&#xNAN;*Example:* "It was great" and "Everything was great overall" given to two unrelated questions by the same respondent.

**7. Non-Response Check**\
Flags explicit refusals or "don't know" answers that carry no substantive content.\
\&#xNAN;*Example:* "n/a", "no comment", "don't know"

**8. Profanity Check**\
Flags responses containing profane, abusive, offensive, or hateful language.\
\&#xNAN;*Example:* A response containing slurs or hateful language directed at the brand or named individuals.

**9. Copy-Paste Check**\
Flags responses that appear pasted from an external source such as marketing copy, web text, or formatted boilerplate, rather than written as a personal answer.\
\&#xNAN;*Example:* "Our mission is to deliver world-class customer experiences through innovation and excellence." in answer to a personal feedback question.

**10. Gibberish Check**\
Flags random keystrokes, character spam, emoji-only responses, or punctuation-only strings.\
\&#xNAN;*Example:* "asdfghjkl", "!!!!!!", ".........."

**11. Language Mismatch Check**\
Flags responses written in a language other than the expected survey language. Automatically routes to a translation flow for reviewer approval.\
\&#xNAN;*Example:* A response in Spanish appearing in a survey fielded in English.

***

#### On the roadmap

* **PII detection**\
  Flags and optionally redacts names, phone numbers, email addresses, and other identifying details that respondents include in open-ended answers.
* **Inconsistency across a respondent's answers**\
  Catches contradictions made by the same respondent across different open-ended questions in the same survey.
* **Cross-respondent duplicate detection**\
  Surfaces identical or near-identical verbatim text appearing across different respondents, a signal of coordinated responding and panel fraud.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://help.metaforms.ai/open-end-cleaning-and-coding/open-end-response-quality-checks.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.