How To Structure Nested Loops Surveys for Successful Data Export

 

Overview

In some cases, SurveyToGo Studio may display a warning that the survey contains an extremely large number of variables. This number is usually much larger than expected for a standard project and may cause issues when exporting or analyzing the data.

The most common cause is the use of nested loops. Nested loops can generate a very large number of export variables because the number of variables is multiplied by the maximum number of iterations in each loop.

How Loop Variables Are Created

Each question variable inside a Loop Chapter is exported with the iteration index added to the beginning of the variable name.

For example, if a loop can run up to two iterations and contains a Single Choice question with the variable name Q_1, the exported variables will be:

I_1_Q_1
I_2_Q_1

With nested loops, additional iteration indexes are added. For example:

I_1_I_1_Q_1
I_1_I_2_Q_1
I_2_I_1_Q_1
I_2_I_2_Q_1

This means that each variable is multiplied by the maximum number of iterations in each loop. As a result, the number of export variables can become very large very quickly.

Example

Assume you have one Multiple Selection question with 25 answers placed inside two nested loops:

• Outer loop: 30 beverage categories

• Inner loop: 150 possible beverage brands

• Multiple Selection question: 25 answers

This creates:

30 x 150 x 25 = 112,500 variables

This is only for one question.

In practice, most of these variables will always remain empty. For example, a combination such as Still Water as the category and Coca-Cola as the brand may never be displayed or answered.

However, the export structure still needs to account for all possible combinations, which can make the dataset very large, difficult to analyze, slow to export, and in some cases impossible to export.

Recommended Scripting Method

In most cases, the survey can be scripted differently to create a more efficient export structure.

The recommended approach is to use the nested loop structure for display and data collection only, while preventing it from being exported. Then, create a separate export structure that stores only the relevant selected data.

The attached samples demonstrate this concept:

1. A simple nested loop structure that does not account for the export variable impact.

2. An optimized structure that looks the same during the interview and collects the same type of data, but creates a much more manageable export structure.

Both samples use an Excel file that defines the relationship between categories and brands, which is a common setup for nested loop scenarios.

How to Review the Samples

To compare the two surveys:

1. Import both surveys into your organization. See How to Import a Survey.

2. Create the Subject Store using the attached Excel file. See How to Use an Excel File as a Data List in a Survey.

3. Name the Subject Store BrandsCategories, matching the Excel file name.

4. Review the structure of both surveys.

In the optimized survey, note the following:

• The nested chapter is configured not to export in the Advanced tab.

• The Export Variables chapter is not displayed during the interview, but it is configured for export.

• The exportable questions are populated by an Expression question placed after the nested loops and before the Export Variables chapter.

Basic Concept

The nested loops are used to collect product-related data, where each product is made up of several attributes.

Instead of exporting every possible category-brand-answer combination, the survey stores only the products and attributes that were actually selected. Each iteration in the Export Variables chapter represents one selected product with its relevant attributes.

This creates a smaller, cleaner, and more usable export structure.

If the Script Cannot Be Restructured

In some cases, it may not be possible to restructure the survey due to time, resource, or scripting constraints.

If you cannot adjust the script structure, refer to the guide that explains how to reduce the number of variables during export.

For additional assistance, contact Dooblo Support from the Lumivero Support Hub.