Survey data
===========

| In addition to the wearable sensor data, daily survey about infant heart rate
  sensor and infant leg movement sensors was collected. The survey was sent to
  primary caregiver once per day for 3 days during the 72 hour data collection
  of infant hear rate sensor data and infant leg movement sensor data. 
| Surveys were automatically sent via email or text message (per caregiver
  preference) at 10 am eastern / 9 am central / 8 am mountain / 7 am pacific.
| The surveys were sent after the sensors had been recording for 24, 48, and 72
  hours. The survey asked about the preceding 24-hour period. It asked whether
  the sensors were worn for most of the time and about the daily activities of
  the infant (whether or not it was a fairly typical 24 hours for the infant,
  estimates of how much they napped and slept, how much they were in a car seat,
  swing or other restraining device, how much they were in a moving car or stroller
  or being carried, and whether or not their legs are swaddled during naps and
  sleep).
  This survey was created by the workgroup.

A research assistant (RA) marked preferred contact and preferred method of contact
(email or text message) at visit, then surveys were sent automatically. Surveys were
provided in Spanish or English per family preference.

Survey data collection occurred at **V02 (0-1 months of age)** and at **V03
(3-8 months of age)**.

Key references
--------------

None


Quality Control (QC) Processes
------------------------------

* **QC Procedures**: Surveys were spot-checked during the data collection time frame.
  Only a small percentage of surveys were randomly checked each week as the process
  was manual. When checked, surveys were checked for the presence of response. When
  no response were provided, RA's were contacted to confirm whether or not surveys had
  gone out to the participants as intended.

* **Common Issues Identified**: Common issues identified during the QC proceeses
  included surveys that had no responses. Sometimes this was human error (preferred
  contact or prefereed method of communication not selected by RA) or potentially
  technological malfunction, and sometimes it was caregiver non-response. Caregiver
  non-response was the reason for the majority of the surveys without responses.
  Caregiver non-response was common. Please consider that occurrence of caregiver
  non-response may not be randomly distributed across the sample.


Potential Issues Flagged by Subject Matter Experts
--------------------------------------------------

No issues were found.