Monday, November 24, 2014

The Ultimate Challenge - Building a Family Tree From Sources - Post 1: Crawford Family

Noted geneablogger James Tanner wrote Building a Pedigree From Sources -- The Ultimate Challenge on his Genealogy's Star blog on 22 November 2014.

1)  In his blog post, James makes the points that:

*  Traditional genealogy began with a pedigree chart and a search for names. From my experience, nearly all beginning researchers start out the same way. They begin filling in names and they follow the suggestions from countless books, classes and websites that teach the Research Cycle.

*  It is past due time to revise this traditional approach and realize that it is no longer necessary... [an alternate way is] building a pedigree strictly from sources.

*  Today, [John] Doe can go onto any one of several online database websites and begin by filling out his name and his parents' names on a pre-constructed family history form. He could use, or other programs. 

*  If he is using either or or, all he has to do is enter a minimal amount of information about himself and his parents and perhaps a name or two of grandparents. The programs then begin to suggest sources for further information.

2)  Is this correct?  Is it foolproof?  What about "proving" your research?  My answers are:

*  Yes, it is correct.  A person can often build an online family tree and the tree provider can provide records for persons in the tree - for example, leaf Hints, or MyHeritage Record Matches, or FamilySearch Record Hints.  

*  No, it's not foolproof.  You may get stuck and have no leafs, hints, or matches.  Or you may get only one more generation back.  It all depends on the names and data entered, and whether there are records available or other researchers with trees with those persons in them.

*  "Proving your research"  is really a concept that beginners usually have no understanding of, but if they continue they eventually should come to understand that they need to cite their sources, look in other resources, consider the evidence collection for each conclusion made, etc.

3)  However, in my opinion, it is a viable way to start a family tree, and to find records and relationships for ancestors in a relatively quick process.  The records found are "low hanging fruit."  Finding other records for persons and their life events in other resources may be a longer process.  

One problem is that the leaf Hints an and Record Matches on MyHeritage requires a subscription.  Building an Ancestry Member Tree is free, and building a small tree on MyHeritage is also free.  Adding information to the FamilySearch Family Tree is free, as are the Record Hints for persons in the tree.

This online family tree process may "hook" beginners on finding out more about their family history. That is good for the beginner, for their family members, for local repositories, for local genealogical societies, and for online data providers.  It's a win-win for almost everybody for a relatively small amount of money.  

4)  I decided to try it out.  I couldn't use my own ancestral families, since I already have online family trees that would be easily found.

I chose persons born between 1900 and 1940 who were in the 1940 U.S. census.  Those would be representative of parents and grandparents of persons interested in starting genealogy research today.  I used the San Diego County census pages near those of my grandparents.  

Initially, I chose four census entries.  I may do more in the future.  I will write a separate blog article about each census entry.

My plan was to chose a person, create a new Ancestry Member Tree (kept private), enter as much information as possible from the 1940 U.S. census record, and then follow the green leaf Hints that provides nearly instantaneously.  How far back could I go? 

5)  Here is the 1940 U.S. Census page for the Earl Jas. Crawford family in San Diego, California:

I started with Betty Lee Crawford (granddaughter of head, age 5 months born in California), whose parents were Lowell A. Crawford (son of head, age 29 born in Minnesota) and Betty R. Crawford (daughter-in-law of head, age 18 born in Illinois), and Earl Jas. Crawford (head of family, age 57 born in Minnesota) and Eva A. Crawford (wife of head, age 59 born in Minnesota).  This gave me three generations to start with.

Here is the Family view in my new Ancestry Member Tree:

After just a minute or two, I received green leaf Hints for Lowell A. Crawford and Earl James Crawford.  There were no Hints for Betty R. (--?--) Crawford, mother of Betty Lee, or for Eva A (--?--) Crawford, mother of Lowell A. Crawford.

I clicked on the Hints for Lowell A. Crawford - there were 6 Unreviewed Hints:

There was a Hint for Ancestry Member Trees, one for the California Death Index, a U.S. City Directory for San Diego, a 1940 Census record, a 1920 Census record, and a Find A Grave record.

There were two existing Ancestry Member Trees with Lowell A. Crawford in it:

I could select one or both of the trees and click on "Review selected tree hints" to add names and events to my new tree.

At this point, I don't know if any of the information in these existing trees are correct, but they do have all of the information that was in the Hints for Lowell A. Crawford, and have the parents correct.  I added the family members to my tree (mainly siblings of Lowell), and then did the same for his father Earl James Crawford and his mother Eva A. Crawford.  The information provided birth and death dates for Lowell and Earl, and information on his mother, Eva Adelle Jennings.  I was able to go back two more generations on both the Crawford and Jennings lines in a short period of time.

I accepted all of the Hints except for some of the Ancestry Member Trees.  By accepting the Hints, events were added to each person, and a Source was added for each of the Events accepted.

I stopped there, although there were leaf Hints for all four of the earliest generation, and for siblings of the direct line:

If there had been no existing Ancestry Member Trees, I would have been able to add most of the information above from the leaf Hints provided.

However, there were no Hints for my starting person, Betty Lee Crawford or her mother Betty R. (--?--) Crawford.  So further research in other resources (family, repository, online) is necessary to find information about them.  There is a chance that the daughter Betty is still alive.

All in all, this effort took me about one hour - but I knew what I was doing.  What I did was not a new process for me.  I have been making and using Ancestry Member Trees for years.  

6)  In this first example, the process worked really well.  I was able to add two generations in one line in a short period of time.  If I was working with a client, I think that they would be somewhat overwhelmed and pretty happy with what was found.  I would emphasize that more information was required to draw conclusions about each of the names, dates and places - birth, marriage and death certificates, other census records, cemetery records, obituaries, military records, family records, etc.  But its a pretty good start on  a family tree.

In a real case with a client or colleague, they would probably have at least some family information, such as a Bible, vital record certificates, newspaper clippings, or living relatives that they could glean information from.

7)  I will do another one of these each day this week trying to uncover problems and find benefits to building a pedigree with sourced information.  

I won't do a similar process with MyHeritage because it takes some time for their search engines to find Record Matches for persons.  I may check out the FamilySearch Family tree to see if this family (starting with Lowell and Betty R. (--?--) Crawford) is included.  

The URL for this post is:

Copyright (c) 2014, Randall J. Seaver


James Tanner said...

Thanks for mentioning my post and many thanks for pointing out some of the limitations still inherent in the process of using the automated source hints. You always have such valuable comments.

Geolover said...

Interesting trials, Randy.

As far as your point 2), "Is it foolproof?" One hazard of the internet-provider site "hints" is alluded to in your "Proving your research" element.

One can easily follow a same-named couple via hints, but be quite unaware that there was more than one same-named couple of similar ages and birth places until really far along in gathering material. Thus life-paths of different persons can be conjoined (in the tree), and determining who was actually who can require finding a more broad scope of data. And if the couples are related, their children can all be John, Richard and Elizabeth . . . .

Having personally assumed wrongly that "same name = same person" more than once, I can attest that the same is possible in internet trees as well as in the algorithms that generate the hints.

Jan Murphy said...

One obvious drawback is that the Ancestry Hint system is over-weighted in favor of people who have the same names. When people bite on these sound-alike hints, then copy all the unexamined compiled information onto a "virtual cemetery" memorial on Find A Grave, Ancestry then serves it back up as a fresh hint. The newbie researcher thinks (as with online trees) that "all these people say this is true, so it must be so!" -- completely unaware that there may be many people all copying the same erroneous Ancestry shaky leaf hint.