# Bridge an open-text survey codebook revision

Trace reclassification when a survey theme definition changes instead of presenting it as a new respondent trend.

If nothing was unclear or difficult, say so. Skip questions about steps you did not experience; use not applicable where appropriate.

## 1. What does the timing label suggest this comment concerns?

Answer: ____________________

## 2. Which part of this fabricated comment distinguishes unclear information from a changed date?

Answer: ____________________

## 3. When would both proposed themes apply to the same comment?

Answer: ____________________

## 4. What case would you leave uncertain under these definitions?

Answer: ____________________

## Author notes / illustrative keys where applicable

1. Probe the broad label’s meaning before choosing a split.
2. Identify the actual inclusion evidence for the narrower themes.
3. Test multi-coding rather than forcing exclusive bins.
4. Expose an unresolved boundary that should remain reviewable.

Interpretation caution: Changed classification of unchanged text is not attitude change. Ambiguous or mixed snippets need their stated review and multi-coding rules.


## Illustrative finding and follow-up

Observation: Three unchanged fabricated snippets are redistributed by a definition change.

Action to test: Define clarity as missing or unclear date information and changed dates as explicit movement of an announced date; keep mixed and uncertain cases reviewable.

Follow-up: Use fresh fabricated snippets including both issues and neither; check inclusion, exclusion and uncertain treatment.


## Collection protocol

When and whom to ask: Before applying a revised codebook, ask reviewers to explain the proposed theme boundaries using synthetic snippets.

Decision owner: The coding owner versions definitions; a second reviewer checks ambiguous and multi-theme cases.

Follow-up: Use fresh fabricated snippets including both issues and neither; check inclusion, exclusion and uncertain treatment.

Response handling: Use invented text and avoid extracting identifiable comments into a public worksheet.

## Decision workshop

Synthetic examples only. Manual worksheet; no automatic import or product writes.

### Synthetic pilot and reporting records

Synthetic snippet | Text | Codebook v1 | Codebook v2
--- | --- | --- | ---
s1 | Date was not explained | Timing | Clarity
s2 | Date moved twice | Timing | Changed dates
s3 | No date provided | Timing | Clarity

### Version changes

Boundary | Draft | Revised
--- | --- | ---
Theme rule | Broad timing label | Separate inclusion rules
Trend | Reclassified records called a decline | Same responses, changed codebook

### Draft or earlier version (review its limits)

Date-change concerns fell from three to one.

### Revised example

The same three synthetic snippets move from one broad timing theme to two clarity cases and one changed-date case under revised definitions. This is reclassification, not respondent change.

### Decision situations

- Original text and both definitions are retained: Write a snippet-level bridge and disclose the coding-version break.
- Earlier text or definitions are unavailable: Hold exact reclassification claims.
- A later respondent wave is compared: Review both instrument and coding comparability before a trend claim.

### Method references

- AAPOR survey best practices: https://aapor.org/standards-and-ethics/best-practices/ — Pretesting and transparent sample context.