Grid-Question-Score – Identifying response patterns in grid questions

Estimated reading: 2 minutes

This article explains all the important information about the Grid-Question-Score (GQS), formerly Item-Battery-Score (IBS) for the analysis of response patterns in grid questions.

The calculation of Grid-Question Scores

The GQS uses machine learning to recognise different answer patterns within grid questions.
The closer the response pattern is to a bad response pattern (e.g. straightlining, zigzag), the lower the respondent's score.
An individual score is generated for each grid question, which is then combined into an overall score.
It is important to note that patterns in the responses (with the exception of straightlining) can only be identified if the items are imported in the order in which they were viewed by the respondent.

In the interview detail view in the pop-up window, the response patterns of all grid questions are displayed for each interview.
You can contribute to the improvement of our services by reporting incorrect assignments to answer patterns - be it false positives or false negatives. Please use the yellow button in the top right-hand corner. Your feedback is valuable to us as it enables us to continuously optimize the machine learning model.

Tips for use

For the quality check, please only use grid questions where the occurrence of straightlining or zigzagging is considered unnatural behaviour to ensure that the GQS provides correct results.
Please note that at least 7 items per grid question are required for an appropriate quality check. A sufficiently accurate assessment cannot be guaranteed with a smaller number of items.
If several grid questions with the same scaling are placed directly one after the other in the questionnaire, these can be specified as one grid question.

Grid-Question-Score – Identifying response patterns in grid questions

The calculation of Grid-Question Scores

Tips for use

Similar articles

CONTENTS