Location of Survey Point is way off

[JimW]

I have a problem with the location that a survey is recorded being way off.  The majority of my points are collected where the user is standing.  However, I always find a bunch of points that are not correctly located.  The points get collected off the coast of Africa and these points all get collected at the exact same location. 

This happens on different IPads and I do not have a geopoint question as part of my survey.  Has anyone else experienced this?  Thanks!

[XanderBakker]

The point that you marked is the X,Y = 0,0 location and could have been caused be a erroneous coordinate system. As you know when a location is collected as Latitude Longitude, the range will be -90 to +90 for Latitude and -180 to +180 for longitude. If for some strange reason the coordinates that are collected are interpreted as for instance Web Mercator Auxiliary Sphere (projected coordinates with the 0,0 at the same location) your latitude and longitude values may have been interpreted as meters and that will place them near the 0,0 location. 

How many different devices were used and how many presented this behavior? Any differences in settings on the devices?  

[JimW]

This happens across a number of devices (all IPads).  They were all set up the exact same and restrictions are enabled on each IPad so that changes to each IPad by the end user are minimized.  When it happens it appears to be for 2 or 3 days in a row but not every point has this issue.  For example, on 3/19/18, every point collected by one user was collected at 0,0.  On 3/20 only 9 out of 23 were collected at 0,0 (see image below).  This behavior is similar for other users as well.

[XanderBakker]

That sounds more like the device could not define a location using the GPS, since when you use the same project in Collector it should not be possible to record the points in two different projections. Have you contacted Esri Support? They could work with you to see what is the reason for this behavior.