PLAY Survey Data

Author

Rick Gilmore

Modified

September 9, 2024

Purpose

PLAY project researchers use KoBoToolbox to collect data from participant families. Data from all sessions at all PLAY data collection sites are pushed to a common set of files on the KoBoToolbox server.

This document describes the process of i) downloading, ii) parsing, iii) cleaning, and iv) exporting these files into forms more suitable for data analysis.

For details about the PLAY Project data collection protocol, see https://PLAY-behaviorome.github.io/protocol/.

Overview of data sets

There are three sets of data collected using KoBoToolbox.

One set is collected during the recruiting and screening calls for the study. This is called the screening or demographic data.

The second set is collected during the home visit. This is called the home visit data.

The third set consists of notes generated by the experimenter after the visit. This data is called the post-visit data.

Data access and processing

All data are downloaded and manipulated locally but are not synched to GitHub. This is accomplished by adding data to .gitignore.

There are separate processes for detecting and deleting personal identifiers from specific data files and exporting ‘cleaned’ versions that may be shared on Databrary.

Home visit data

There are three (3) age groups (12-, 18-, and 24-month-olds) and four (4) different language environment groups (English-only, Spanish-only, Bilingual-English-dominant, Bilingual-Spanish-dominant). Since the language of the questions and some of the specific questions vary by age group and by language environment group, there multiple KoBoToolbox survey form files that must be processed.