There is a lot of junk on the Internet.
More experienced genealogists, both professionals and hobbyists, know this. We repeat it in our blogs, in our research plans, in our conversations with other genealogists. We stay away from the Public Family Trees on Ancestry.com and FamilySearch‘s International Genealogical Index. After all, these all just have junk put online by those “shaky leaf” clickers, right?
One should by no means trust an online family tree.
But neither should one trust a death certificate or a 19th-century county history or a federal census record or an obituary.
Just because it’s online does not make it more or less garbage than any other source. You still should evaluate the information the same way you would in any other record. Identify the informant. Determine their involvement in the reported event or the source of their information (if secondary).
Two cases are perfect examples of this philosophy:
Almost fifteen years ago, when “Internet genealogy” barely had an existence, I came across a family tree that contained my then-earliest known ancestor in my male line: Myron Grant Hait, my great-great-grandfather. I contacted the owner, who turned out to be my grandfather’s first cousin. My great-grandfather, who lived in New York, was one of six brothers, all of whom lived in different and distant states: California, Montana, North Carolina, Louisiana, etc. In those pre-Facebook days, distant relatives did not always maintain close contact. When my grandfather moved to Washington, D. C., to work for the federal government, he had even less contact with the extended family. He knew his uncles, but did not know any of his cousins.
This cousin, Linda, just so happened to have quite a number of family records in her possession, including letters to and from my great-grandparents from back in the 1970s when she started researching, and a family history written by my great-great-grandmother in the 1930s. She also put me in touch with another cousin who had in her possession a copy of a family bible, several old family photos, and a collection of Civil War letters!
Of course not all of her research was completely accurate, but much of it was, and of course the original records in the possession of these long-lost (to me) branches of the family were indispensible. Had I ignored this online family tree, I would have never obtained many of these records.
The second case involves a family that I was working on for a client. While searching for records on Ancestry, I discovered a public family tree. Though not a single offline source was cited, the information was extremely specific. I jotted down a few notes from the tree for confirmation, but then went on along my merry research way.
The next day at the Maryland State Archives I happened to run into a friend of mine: also a professional genealogist, member of my APG chapter, and a fellow Certified Genealogist. I knew that she did a lot of research in this particular county, so I asked her if she was familiar with the families I was researching. To make a long story short, the owner of the Ancestry public family tree was her client, who had uploaded the results of her research to the site without any source citations. In other words, though it looked like “junk” because it did not have any sources cited for any of the information, the tree actually reflected the work of a Certified professional genealogist. As I continued to research the family, I was able to confirm all of the information that was in the public tree.
As the first example shows, online family trees are often a great way to identify other descendants of the families you are researching. Some of these distant cousins may have family records passed down in their lines that you do not have access to: items like family bibles, old family photos, etc.
The first condition of the Genealogical Proof Standard is that we conduct a reasonably exhaustive search for all records relevant to our research problem. If you have ignored the search for family records in other lines, have you met this requirement?
Planting the Seeds 
I, too, think online trees should be consulted, but not taken gospel truth without some corroborating info. Having researched original birth/death certificates and found errors – based on other substantial data. These types of records are not necessarily more accuate than an online tree. I get irritated with those who claim that citing these original records is virtually infalible. All information: online trees, birth/death/cemetary records MUST be evaluated used only when confirmed by other sources or is the best data based on the context (location, family, age, etc).
Online trees can often provide clues to other avenues for research.
I put a tree online and have received much additional information that I would not have found anytime soon just because the tree was online.
If I’m stuck, I do sometimes look to family trees for clues. It was someone else’s tree that uncovered the fact that the woman I believed to be my great grandfather’s sister was actually his niece. Even my mom, who knew pretty much everyone involved wasn’t aware of the fact. The tree owner turned out to be one of my cousins and her information came from her mother, who was a step-sister of the woman in question. Not everyone cites their sources or the source may be an oral, rather than written one.
As far as written records, they’re only as accurate as the person giving the information. Even if the informant is giving accurate information, there’s still a potential for error because the person writing it down or transcribing the record may have goofed.
I try to look at all available information and draw my conclusions from the whole rather than just a piece.
This is a great example of why public trees shouldn’t be ignored, but they can’t be taken for ‘gospel’ either, without the appropriate due diligence. Thank you for writing on this topic.
Thank you for commenting!
I’m glad to see you write about this, and especially about the fact that nothing should be trusted without verification. Some of my research involves descendant studies, and when people are entering information about close family with whom they grew up, they may know some things that other researchers don’t (of course, there can still be mistakes). Also a good source for pictures – I had never seen a picture of my paternal grandfather until one of these cousins sent me two pictures.
That is exactly the point. Even though there is a lot of junk out there, there is also a lot of gold. Some of my greatest breakthroughs have come from locating information provided by descendants. For a more indirect example, see my article “Family Bibles as an underutilized genealogical resource.” I located a relevant family bible, not through direct communication with a living descendant, but through a family history book published by a living descendant. This published reproduced images of the pages of a family bible still in the possession of another living descendant.
Just last week, I found a family tree where a woman had given birth two years after her death. It seemed odd to me, but apparently not to the family tree creator. :o) But …. as I continued my search, it became apparent that the 1st wife had died and the 2nd wife had given birth (neither of these facts being on said family tree). I was not aware of wife number one and while I’m sure that I would have found her as my research progressed, it did give me a heads up about her existence.
So, yes, I read them, then read their citations and determine what parts of the information is useful.
Pingback: How to conduct a “reasonably exhaustive search” for relevant records « Planting the Seeds
I agree that it is worth checking online trees as part of a reasonably exhaustive search, but if the entries are all garbage, how much time is the genealogist expected to spend detailing them? Is it sufficient to say that you checked Ancestry.com and found unsourced records from generic trees, or are you supposed to document and comment on each garbage entry for the relevant people?
With experience you can usually determine which of the online trees were copied from other trees and which were compiled from the research of a living descendant. Most of the online family tree sites, like Ancestry, GenCircles, Geni, and MyHeritage, among others, have a way to contact the owner of the tree. You can then talk with them to discover important information such as how long they have been researching, and (more importantly) whether or not they might have sources important to your own research.
Understood. I can tell the useful ones from the bad ones with no problem. I had one of those good cases where I got a hit in a private tree and contacted the person. We were each able to share a lot of information that the other didn’t know.
Most of what I find, though, is going to be poorly sourced if at all, and often copied from the same source. If I’m doing client work, or even work for myself, though, I’m supposed to document the failures as well, to show that I was thorough. If I find one, or maybe 20 of these useless entries in tree, how far am I supposed to go in documenting them and explaining why I’m ignoring them?
Traditionally, it has been recommended that genealogists conduct a “literature search” for previously published research on the family of interest. Before there were online trees, this included published books and genealogical society journals. This now, in my opinion, includes online family trees.
Quite honestly, I have never seen a published case study that cited “junk” online family trees. If you contact the descendant and share information, then you would not be citing the family tree, you would be citing your correspondence with the researcher. If you use original records to verify (or refute) the information in the family tree, then you would be citing the original records.
When you are conducting research for a client, however, you should document how the research time was spent. If you spend a few hours dealing with online family trees, then you should document what information you viewed and analyzed. I would not say that you have to cite every tree that copies information from other trees, but rather look at the claims in the tree and try to identify the “original tree” in which the information appears. You can then assess the claims of fact, instead of the tree itself. You should of course note that the claims were copied to x number of trees, but that these other trees did not offer any additional detail (if that is indeed the case).
I hope this helps to answer your questions.
OK, that sounds like a reasonable amount of effort so that the useless entries aren’t being ignored but too much time isn’t spent on them either. Thanks!
Pingback: The “literature search” in the Internet age « Planting the Seeds