Decipher Expanded Variables Writeback - Naming Convention

Introduction

When writing data back to Decipher using the “Expanded Variables” writeback type, each code applied is written to a separate single variable in Decipher.  This is in contrast to the “Dichotomous Variables” writeback type which creates a multiple question containing a separate binary value (i.e. column) for each code in the codeframe.
Some users prefer the “Expanded Variables” approach because it minimises the number of columns in exports and eliminates the “empty space” involved with the “Dichotmous Variables” approach.

However, the “Expanded Variables” approach will result in a variable number of questions being created in Decipher to store the coded data. The exact number of questions will vary depending on the number of iterations in the source data and the number of codes applied to each verbatim.

When Codeit creates these questions in Decipher, it follows a specific naming convention.
This page explains this naming convention.

Naming Convention
In short, the naming convention follows this pattern:
 
 <Target Variable Name>[_oVerbatim Ordinal + 1][_cCode Index + 1]

To help understand what that means, let’s look at an example.

Simple Example
Suppose we have a set of verbatim data as follows:

Suppose we code Q1 into a target variable called Q1_coded
If we just apply just one code to each verbatim then the naming of our output questions/columns in Decipher is straightforward:

Multi-coded Example
But if we apply more than one code, then we need to generate a name for each of the additional codes, e.g:

Multi-response Example
In addition to multi coding, we sometimes need to handle multi-response verbatims.
For example, suppose Q1 contains brand mentions where the respondent mentions up to 3 brand names, like so:

Since each respondent can give multiple answers to Q1, we again need a way of labelling the multiple outputs. Following the naming convention above, we would label the questions written to Decipher as follows:

Multi-coded, Multi-response Example
Lastly, sometimes we see a combination of multiple response data that needs to be multi-coded.
For example, suppose a respondent mentions two brands in one cell, like so:

 In this example, we would label the output data as follows