Repeat Group Questions and Data in a Survey

Sometimes, a survey needs to collect multiple sets of the same information from one respondent.

For example, in a household survey, you may want to record details of every household member — such as their name, age, and gender.

Because households have different numbers of members, this block of questions cannot appear as fixed columns in the main dataset.

Instead, tools like ODK, KoboToolbox, or SurveyCTO capture these repeating questions as a repeat group.

How Does the Data Come In?

Data from a repeat group is exported differently from the main survey data:

The main survey appears as a table with one row per respondent.

The repeat group appears in a separate table, with:

One row per repeated entry (e.g., one row per household member)

A column (usually parent_index or PARENT_KEY) that links each repeated entry to its corresponding main record.

This link allows analysts to join the repeated data back to the main dataset.

The two tables below demonstrate this.