I am starting to do research on some veterans who served in the Civil War. Several of the collections available online that describe service records are compiled databases or indexes collecting all the records about the veterans by name.
Some compiled databases found via online providers like Ancestry display the information for each name as if it were all about one person. This makes it tempting to fall into the trap of "it's the right geographical area and the right name, it must be the same person".
The problems are like the scenarios in the related question How can I establish and describe my confidence that the child and the adult are the same person? But my question today is the larger one of how to establish the context, and determine the quality, of the records I find.
For example: a transcription of the notes made in a family Bible (itself a compiled source) records a Daniel Burns. Some of his family can be located in Vermont. The Bible says he died in the Civil War in Louisiana. Is this Daniel Burns the same person as the one in this memorial on Find A Grave? The person who created this memorial knows the person his Daniel Burns married, but not his parents -- the Bible lists the parents of the younger Daniel Burns, but not the person he married.
Daniel Burns from the family Bible (the 'young' Daniel Burns):
Daniel is listed as one of the eight children of Samuel Burns (Sept 20, 1795 - ?) and Lois Stevens (Feb 22, 1800 - Nov 15, 1856), m. March 15 1812. Daniel's nephew Edson Wellington Burns was born in 1849, in Jericho, Vermont (Chittenden County). Here is Daniel's information, with the transcriber's notes in parentheses:
Daniel Burns Jan 14, 1823 - Aug 22, 1862 died in the Civil War Louisiana
Daniel Burns married someone July 1, 1848 (no name was written down) Children unknown
The publication date of the Bible is a clue to when the entries were made. Any events that took place dates before the pub date of the Bible were obviously recorded after those events occurred. (They could not have been recorded at the time the events happened, because the Bible hadn't been printed yet.) The Bible was printed and published by Case, Lockwood, and Brainard in 1872. This means we know that all of Daniel's events took place before this Bible was printed, and they could not have been recorded in this Bible when the events took place. Even assuming the earliest-case scenario that this Bible was the first printing, and that it was purchased not long after the Bible was printed, the Bible is too new; all this information was added after Daniel's death.
So far, this Bible transcription is the only source I have for the young Daniel which lists him by name.
These are the census records for a family in Jericho which match the information found in the Bible. The 1830 US Federal Census and the 1840 US Federal Census only list the head of household, Samuel Burns, by name. The ages of the children are consistent with what I would expect to find, given the information from the Bible. There is a male of the right age in both of these census households that would correspond to Daniel Burns.
In the collection U.S., Civil War Draft Registrations Records, 1863-1865 on Ancestry.com, this entry made in May 1863 for the 38-year-old Vermont-born farmer, Geo. W. Burns, a resident of Jericho, is the right age to be Daniel's brother George Wallace Burns (Bible: born June 29 1825), the father of Edson Wellington Burns. Among the other men from Jericho, there is a Daniel of the right age, but he is Daniel B. Bishop, also a farmer. More investigation is needed to rule out an accident of copying.
In the 1850 Census, this record in Jericho, which I believe is the correct one, has Samuel and Lois, but only one of the eight siblings. I have not found any other record for Daniel Burns in that census year. Note that if the dates in the Bible are accurate, Daniel is already married by the time this census was taken.
Corporal Daniel G. Burns:
There is a photo of a headstone on this Find-a-Grave memorial, but it is very difficult to read. As far as I can see, the information in the memorial did not come from anything on the stone itself. The burial information associated with this stone is:
Chalmette National Cemetery, Chalmette, St. Bernard Parish, Louisiana, USA Plot: Section 69 Grave 5557
The creator of the memorial is related to Daniel's wife's family, so I'm assuming some of the information in the memorial, like the statement that Corporal Burns was re-interred, may have come down through her family.
Looking for historical records consistent with those assertions, I found this family in the 1860 Census, and a result from Ancestry's online database U.S., Civil War Soldier Records and Profiles, 1861-1865. (In this case, we know the source-of-the-source material: Revised Roster of Vermont Volunteers During the War of the Rebellion 1861-66 published Montpelier: Watchman, 1892, viewable via the Internet Archive, which is itself a compiled source.
This history of the Vermont Volunteers has tables which list residence information, the date of enlistment, the date of muster, the death date, and a column for remarks.
On Page 271:
SEVENTH REGIMENT Company A. Corporals
Daniel G. Burns Jericho, (enlistment) Dec. 7, '61 (muster) Feb. 12, '62 (Remarks) Died Aug. 22, '62 of disease.
The date of death (22 Aug 1862), the hospital at which he died, cause of death, and other information is recorded in the U.S., Registers of Deaths of Volunteers, 1861-1865 (images) on Ancestry.com.
Other records and information to consider:
Corporal Burns' widow Ann is listed in the United States 1890 Census of Union Veterans and Widows of the Civil War. Ancestry's index reads as Daniel Q. Burns. She is enumerated in Fairfax, Franklin County, Vermont. Note that the enumerator or someone else in the office has written "Killed" with the date of death across the columns for the date of discharge and the length of service; the length of service was not recorded.
Ann Burns, widow of Daniel G. Burns, Coporal, (Company) A, 7 V(ermon)t Inf(antry), (enlisted) 7 Dec 1861 Killed Aug. 22, 1862.
An index card from the collection, U.S., Civil War Pension Index: General Index to Pension Files, 1861-1934 (index and images) on Ancestry.com. The soldier's name is Daniel G. Burns. He served in Company A of the 7th Vermont Infantry, and his widow Ann Burns' application was filed on October 9, 1862. There is a application number and a certificate number; the field 'state from which filed' is blank.
An index card from a Vermont, Death Records, 1909-2008 collection on Ancestry.com (index and images) lists in the Jericho, Vermont town records, a Dan'l G. Burns, Laborer, who died of fever in New Orleans, La. on Sept 10, 1862. (Informant on card: E. H. Lane, Town Clerk)
Notes about the sources:
We have the source-of-the-source for this specific entry in the U.S., Civil War Soldier Records and Profiles, 1861-1865, but there are other compiled databases about Civil War soldiers on Ancestry which do not have associated bibliographies, so it is more difficult to find out where the information in the indices or other compiled woks might have come from.
A research guide to the Compiled Military Service Records (CMSR) Compiled by Bryan L. Mulcahy, Reference Librarian, Fort Myers-Lee County Library, is available for download from Rootsweb.
The section about Union Veterans in the guide says:
Sometimes the “volunteer” enlistment papers show age, residence at time of enlistment, occupation, and physical description. If the soldier was hospitalized, a bed care card > will show the age, birthplace, marital status, residence, and date/place where soldier was wounded. If the soldier died, a casualty sheet will show the date and place of death and next of kin (if known).
Ordering the CMSR for Corporal Daniel G. Burns is likely to show his wife as next of kin (if any is shown) and the best I can hope for him or any of the other Daniel Burns' CMSRs is (maybe) the residence at the time of enlistment, which might confirm the residence and other information given in the Vermont Volunteers' roster. (It probably won't have any information about his parents, unless he listed either of them as next-of-kin instead of his wife, which doesn't seem likely.)
The historical context:
It is tempting to say "these must be the same person!" because the marriage and death dates match, and the place of death matches. Corporal Daniel G. Burns apparently comes from the same part of Vermont as the family from the Bible. But BURNS is a common surname, and I dislike hanging an identification on so few pieces of evidence. As Elizabeth Shown Mills says, one source is not proof.
How might I puzzle out how many men named Daniel Burns might have been in the population of Jericho, Chittenden County, Vermont around this time period? How many might be in the surrounding counties and states?
What techniques can I use in searching to estimate how complete the coverage in a record collection might be?
What other questions should I be asking to ensure I am doing a reasonably exhaustive search?
I'm putting some material here which was originally in the comments, and some which was edited out of the initial version of this question.
Since I did not make the extraction, I do not have any notes about changes in handwriting that would provide clues about when the other entries were made. That information might have some bearing on how long we can establish some of the residence of the family in Vermont after Daniel died, but it's not relevant to determining the data bounds for Daniel's data, given the date of the Bible. (From the publication date, it seems likely that one of Daniel's siblings purchased the Bible. The parents' generation is the first one listed.)
The topic of the weekly chat, Genealogy Chat 24 May 2014 | What's in a name? prompted me to get the question written before I forgot to post it.
I'll review my records and provide more information about the specific individuals when time allows, but my question here is about the larger question: how I can properly evaluate the quality of the data in each of the items I discover, which includes finding out data about the collections.