Question about entering Ne data when it is only vague- and details not available

[shoban]

Sean: Ash and I have a question about entering Ne data in the Kobo.  I've been checking over some datasets and was discussing one with Ash Hamilton.  We have a species, the blunt nosed lizard, for which an Ne study has been done with DNA data.  We want to include this because relatively few USA species have actual Ne data.  However:

• The Ne study was done and not published- I think the agency contracted a scientist to do the genetic study but it wasn't a scientific publication.  So we cannot get the actual Ne estimate, marker info, etc.  There is not even an unofficial report, just a personal communication cited.
  
• The Ne of populations is only reported as either <50, 50-250, >250.  In other words populations are just put into bins of Ne.  So we don't know the actual Ne estimate or confidence intervals.  See Table 9 for the bins and Table 10 for each population.
  
• We wanted to put 50-250 for those populations for which its known but that doesn't fit our Kobo form.  For the Kobo form it needs a point estimate and confidence intervals

Alicia's response was:

Indeed there won't be a way to include this vague information into the form in its current format. To be honest, opening the Ne field to allow for text such as you would need it, would open so many data entry errors that no, just no. My suggestion would be to go for the mid point between 50 and 250, use that as point estimate with 50 as lower limit and 250 as upper, and in the comments explain what you did.

If we have more cases like this, we could add another field for "range Ne" as we did with the rangeNc but consider that in downstream analyses something similar to what I suggest above would be done to actually estimate the indicator.