Welcome to City-Data.com Forum!
U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Science and Technology
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
 
Old 06-24-2012, 02:09 PM
 
Location: Philadelphia
1,342 posts, read 3,234,648 times
Reputation: 1528

Advertisements

Hello, I wasn't sure if this was the correct forum for this question, but here goes.

I am researching West Virginia Civil War history, and going through the Library of Virginia's Virginia Military Dead Database . The database lists a total of 31,777 men. I found that 1,814 casualties are attributed to West Virginia counties. These are "determined" casualties, meaning that they have assigned a county of origin. Complicating the finding of a definite number of West Virginia casualties is the "undetermined" category, which has 14,850 men of undetermined counties. The 1,814 men represent almost 11% of the "determined " total of 16,927.

My question is, would it be a reasonable assumption to count 10% of the "undetermined" as an approximation from West Virginia?
Reply With Quote Quick reply to this message

 
Old 06-24-2012, 06:54 PM
 
28,803 posts, read 47,552,525 times
Reputation: 37905
If the database is referring to Virginia then I fail to see why undetermined has bearing. If they are all from Virginia why does the county matter?

Wait! I misread that. Let me read it again...
Reply With Quote Quick reply to this message
 
Old 06-24-2012, 06:59 PM
 
28,803 posts, read 47,552,525 times
Reputation: 37905
I'd have a tough time doing it, but then I'm an unbeliever when it comes to statistical averaging. As soon as you make that assumption someone will find a reference that shows 98% of them were from Virginia.
Reply With Quote Quick reply to this message
 
Old 06-25-2012, 11:37 AM
 
23,557 posts, read 70,067,529 times
Reputation: 49056
Tek is on the right track, I think. You are attempting interpolation based on a questionable base of data points. You have no idea of bias that might be affecting the reporting. Did the counties in both areas have equal methods of reporting? Were there more undocumented fighters in various areas than others due to external factors? What is meant by "undetermined?"

What you are attempting is a guesstimate. In genealogy, it is not at all uncommon to run across people who had no idea where they were born, or who moved multiple times. Assigning them a time and place of birth without evidence is a fraud.

The axiom in photo editing is that you cannot add information that is not in the image. You can bring forward aspects of the image where the information is available, and guess at filling in a scratch or defect on the print, but not determine what great grandmother Winnie looked like based on a single faded grain of silver.

As a guesstimate, you are probably in the ball park, your percentage may be a little high, but it is likely somewhere around there.
Reply With Quote Quick reply to this message
 
Old 06-25-2012, 03:06 PM
 
Location: Philadelphia
1,342 posts, read 3,234,648 times
Reputation: 1528
Thanks Harry & Tek. Chances are all the soldiers might possibly be identified if some institution with the money and determination decided to do it, but the chances of that are slim. I was just trying to get a reasonable estimate of the total based on the analysis of the major portion of data. It would certainly not be anything more than a reasoned estimate, much like a poll projection. The data would have been collected over a 4 year period under widely varying circumstances, so I don't think there would be a determined bias to the infomation. Current estimates of Confederate soldiers from WV is about 20,000, and the casualty rate for the Confederate army was about 20%. The "guesstimate" I come up with using the Library of VA website data is about 3,400 men, which is in the range of that 20%. It is something that will probably never have a definitive answer.
Reply With Quote Quick reply to this message
 
Old 06-26-2012, 02:01 PM
 
28,803 posts, read 47,552,525 times
Reputation: 37905
I agree with your method and estimate considering the difficulty in collecting viable data. Sounds like you're going about it the right way.
Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply
Please update this thread with any new information or opinions. This open thread is still read by thousands of people, so we encourage all additional points of view.

Quick Reply
Message:

Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Science and Technology
Similar Threads

All times are GMT -6. The time now is 08:44 PM.

© 2005-2024, Advameg, Inc. · Please obey Forum Rules · Terms of Use and Privacy Policy · Bug Bounty

City-Data.com - Contact Us - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37 - Top