How to Identify Questions with Straight-Lining/Flat-Lining

This article describes how to identify questions with straight-lining (also known as flat-lining), where a respondent chooses the same option for all items in a question, in your project.

Requirements

• A Displayr document with a data set.

Method

1. With a data set selected, go to any of the following toolbar options:
   • Anything > Report > Straight-Lining/Flat-Lining
   • Table > Report > Straight-Lining/Flat-Lining
2. A QScript will run and identify the questions where straight-lining is present, and create a new folder in the Pages tree containing pages that summarize which questions have been identified where straight-lining is present.
   

NOTE: The Nominal - Multi questions in a project are examined, and new questions are created that show the extent of straight-lining. It assumes that there are no missing values (i.e., if a respondent has missing data, then they will not be classified as having flatlined, even if they gave the same answers to all the questions that they answered).

New filter variables are added to the project:

• Straight Lining by Question - This shows the proportion of straight-lining respondents in each question.
• Number of Questions with Straight Lining - Counts the number of questions that have been straight-lined by each respondent.
• Always Straight Lines - A filter to select those respondents who have straight-lined all considered questions.

These filter variables can be used to select respondents who have straight-lined all considered variable sets. The filter is available in the Data Editor where it can be used to identify and delete cases.

Next

How to View or Edit Raw Data in Displayr

How to Delete Cases From a Data Set