Check Your Record Hints Carefully

We have to remember that Record Hints are hints. There is nothing about a Record Hint that supposes that FamilySearch has done anything to verify the content of the hint. A "hint" is a suggestion, nothing more, nothing less. The accuracy of the hints, by the way, is based on the accuracy of the information already in the FamilySearch.org Family Tree. For example, if the name in the family tree is incorrect, even if the program provides a "Record Hint" the hint will probably be for the wrong person.

There are a number of other reasons why the record hints for any of the programs that provide them might be wrong. One of the most common is the lack of records. For example, if I am searching in England or Scandinavia I may run into a situation where there are hundreds of people with the same or similar name is my ancestor. The existing records will also contain very little information differentiating people with the same name. In this situation, the computer program creating the record hints will often suggest people who have the same name and similar information but not be the same person.

Advanced searching techniques can make the searches more accurate such as those done by MyHeritage.com, but inevitably there will always be "false positives." A false positive is when the program indicates a match when no match actually exists. The goal of the programmers is to reduce false positives but in reality because of the discrepancies in genealogical records, this task is nearly impossible as we go further back in time. It is relatively easy for a record matching program to find a match for a person living within the last 200 years in an English-speaking country. But as we go back in time, the number of records decreases, the number of people with the same names increases and the difficulty in finding accurate matches eventually becomes nearly impossible.

Record match technology is a tremendous aid to research. FamilySearch.org has done an incredible job in improving the technology on the website for the Family Tree. But, there will always be limitations and as the program evolves there may be difficulties in maintaining accuracy. As the records come online and the indexes are done. We should also remember that the accuracy of the record hints depends on the accuracy of the indexing. Obviously, Record Hints cannot be supplied for unindexed records. Therefore, indexing is crucial for the expansion of record hints technology.

All this means that you should closely examine record hints for accuracy and consistency. In every case, be sure and incorporate the information you learn from the accurate records. Determining whether or not a record hint is accurate may involve a degree of additional research.

In summary, do not blindly rely on record hints in any of the online programs.

Reclaim the Records liberates New Jersey Marriage Index, 1901-2016

With help from a volunteer, the Reclaim the Records organization has done it again! They have "liberated" the entire New Jersey Marriage Index for 1901-2016. Here is a quote from their 17th Newsletter:

Introducing the NEW JERSEY MARRIAGE INDEX, 1901-2016! These records are now totally digital, and totally free -- forever! Now you can research anyone who got married in the Garden State right from your home, still in your pajamas. 

We've posted these images at our favorite online library, the Internet Archive (archive.org). You can skip right to any year you want and flip through all the images, or you can download the records to your hard drive as JPG's, PDF's, and/or other formats. Each file is listed year-by-year (or occasionally by a year range), and then the marriages are listed alphabetically by surname. 

Just to be clear: these are images of the index, so this isn't a real text-searchable marriage database just yet. But rest assured that the usual genealogy websites we all know are going to start indexing projects and will make that happen eventually. (Yes, the Internet Archive does run automatic OCR on the text contained in the images, but the recognition quality isn't that great, so you're probably better off just reading through the images instead of trying to text-search.)

I am vitally interested in this effort as evidenced by the fact that my wife and I are preparing to move to the Washington, D.C. area to be full-time FamilySearch missionaries/record preservation specialists for FamilySearch and its sponsor The Church of Jesus Christ of Latter-day Saints. Our responsibilities will include helping to digitize records which will then be available to genealogists.

Here is a screenshot of the Introduction to the New Jersey Marriage Index, 1901-2016 on the Internet Archive or Archive.org:

https://archive.org/details/njmarriageindex?sort=titleSorter

I strongly suggest reading the entire newsletter article. It is really a primer about how to go about helping make even more records available across the country. 

When is a source not a source? A Family History Challenge

A genealogical source citation should answer, at least, two questions:

• Where did the information come from?
• How can someone else find the same information?

How those questions are answered has developed into a major issue among genealogists. At one end of the spectrum are an extremely small number of academic/professional genealogists who have developed an elaborate "citation" system based primarily on various published systems such as the Chicago Manual of Style. 2017 and others. The professional/academic system of recording sources has become so elaborate that my 15th edition of the Chicago Manual of Style is 956 pages long.

Clear at the other end of the spectrum is the system of citations included on the standard family group record for many years Here is an example of the source field:

Unless you use very small letters, there isn't enough space here to encourage the entry of a source. 

Hmm. It is time to start defining my terms. 

The word ambiguous is defined as a word that has more than one meaning or interpretation. The word "source" as used by genealogists falls into that category. One meaning of the word focuses on where the information used by the genealogist to enter names, events, and other information was obtained. Another use of the word "source" is used as a synonym for the word "citation." The first use of the word "source" answers the first question above. Citing the source involves recording the identity of the place where the information was obtained. For example, if I find a reference to my ancestor in a book, the source is the book. When I use that information to enter names or dates or whatever and to my own records, i.e. a genealogy database program, the citation is the information I enter telling others about the book.

A genealogist's failure to record where the information was obtained renders the information useless to subsequent researchers. The concept here is that any information recorded about our ancestors, i.e. historical figures, needs to be documented with detailed information about where the recorded information was obtained so that subsequent researchers can verify whether or not the information is correct and also determine the degree of reliability of the information. 

Yes, I do have to keep repeating myself in order to make sure that what I'm saying is absolutely clear. No, I do not have detectable dementia yet. :-)

So when is the source not a source?

Even if the information supplied is referred to as a "source" if it fails to tell where the information was obtained it is essentially not a source. For example, here is an entry from FindAGrave.com:

Accompanying this entry are three photographs showing the grave marker for Dr. David Shepherd. The only information on the great marker is the name "Dr. David Shipherd" and the two dates. The information contained on grave markers is commonly considered to be a "source." However, the origin of the remaining information in the entry except for the identity of the cemetery is missing. It is helpful to have a citation to this FindAGrave.com entry and to see the photo of the grave marker, but there is no real way to determine where the information came from or when it was recorded. There is a link to David Shepherd's wife's grave in the same cemetery. However, that entry although also detailed, does not has the same three photos shown for her husband. Although a reference or citation to the FindAGrave.com memorial tells us where the researcher found the information, it does not tell us how we can find the same information since the FindAGrave.com information must have come from some other source. 

So, simply copying information that does not answer the questions above, does not help us determine the origin or reliability of the information and therefore has little or no value. By the way, grave markers are not necessarily accurate as to birth information. You cannot also assume that they are accurate as to the death information either. The marker could have been placed many years after the actual events. 

The real issue here is our ability to determine the reliability of the information we find in genealogical compilations such as online family trees and other types of publications of collected information. 

How to Analyze a Record in the FamilySearch Family Tree: Part Three -- What is missing?

In this post, I am continuing to show the process of analyzing a record in the FamilySearch.org Family Tree. Continuing from the earlier posts, the easy part is now done. The entry for Dr. David Shepherd Jr. has received its first clean up and the Record Hints have been added. What is next?

I am still concerned about the name including the title and the designation as a "Jr." The grave marker photo on FindAGrave.com has the name spelled "Shipherd" with the birthdate and death date as shown in the details above. There is also a second grave marker with the name "D Shipherd" and yet a third photo of a grave marker with the name, "Dr David." Does the existence of three grave markers imply three different people?

None of the sources suggested in the Record Hints gives any specific and timely information about his birth or the original spelling of this name. The information on the FindARecord.com website is not supported by any documents. You would have to make the assumption that the grave marker with the "Dr David" and the one showing "Shipherd" with the dates of birth and death, to be the same person. The date of burial reflects the place in FindAGrave.com but the date is estimated from the death date. So we are missing any documents or records that show the birth and death information other than the guesses.

In addition, the name here includes both a title (Dr.) and a suffix (Jr.) neither of which show up on any of the records so far.

From my experience, most beginning genealogists would think they were finished at this point. The entry is complete. There is a "source" for the information and it would seem to be time to move on to other people. Is that realistic? Not at all. It is still time to do some more research. According to what is in the record already, he lived until 1877. So he should be in the 1850, 1860 and 1870 U.S. Census records. This is an obvious record to find. Here is the 1870 U.S. Census record.

This supports the designation as "Dr." because his occupation is defined as "Physician." It also provides support for his wife's name. The record has "Sally Smith" but Sally is a common nickname for Sarah. The ages of David and his wife are also approximated by the ages in the census record. That provides a little more information and confirms what we already have. Next we find the 1860 U.S. Census:

This additional record raises some questions about the names of the children, but otherwise supports what is already in the Family Tree. However, David Shepherd is not shown as a physician, but as a farmer. Checking on Ancestry.com, I found eleven more Record Hints.

Here is the 1850 U.S. Census record from Ancestry.com.

There are a number of additional records and the children still need to be corrected and sources added. This is an ongoing project. I can keep adding records and correcting the entries until the sources become much harder to find.

As I continue to add sources, the confidence level of the entries become higher and higher.

Here are the previous parts of this series:

http://rejoiceandbeexceedingglad.blogspot.com/2017/10/how-to-analyze-record-in-familysearch_8.html
http://rejoiceandbeexceedingglad.blogspot.com/2017/10/how-to-analyze-record-in-familysearch.html

How to Analyze a Record in the FamilySearch Family Tree: Part Two -- Cleaning up the Entries

I selected Dr. David Shepherd Jr. as my "test subject" in this series focusing on analyzing the records in the FamilySearch.org Family Tree because of a lack of sources and some obviously serious issues. By the way, finding such a situation in the Family Tree is extremely easy for me with my ancestors because even after many years of cleaning up the entries, there is still a huge amount of work to be done. One reason, as I mentioned previously, for selecting this person as an example is that the entry is apparently complete and detailed even with the lack of sources. The information must have come from somewhere.

A little bit of research in the form of reading about the family in my daughter's blog, TheAncestorFiles, disclosed that there is apparently a book about the family. Here is the citation.

Barnett, Eula Mae Garrett. 1983. Shepherd family history, 1605-1966. Salt Lake City, Utah: Filmed by the Genealogical Society of Utah.

The book is a typed manuscript that has been microfilmed and the only copy is in the Family History Library in Salt Lake City, Utah. I have not yet made a trip to Salt Lake to view the microfilm, but I am guessing that the information came from that book.

The first step in cleaning up and adding sources to this record is to look at the list of "Birth Names." As I explained, these names are merely an artifact of the numerous submissions over the years and unless there is a real dispute about the person's birth name, these duplicate entries can be deleted especially the one about "Dr." being a title of nobility.

Before:

After:

I could also delete the reference to the census as a source, but the date and place might be helpful, so I left that entry alone.

Next, there is a whole list of sources. Here is are the before and after screenshots.

Before:

I had already added one source that had a reference to both a birth and death date. The FindAGrave.com entry is technically not a "source" because the information is usually not supported by documentation other than a photo of the grave marker but in this case, adding the entry supports the information already in the Family Tree. Here is a screenshot of the original entry from FindAGrave.com.

There is a long list of Record Hints.

The question always comes up about whether or not to add the Record Hints if they appear to be duplicates. It is a really good idea to add all the Record Hints, whether or not they appear to be duplicates in order to notify the computer program that it has made a right selection.  If it bothers you, you can detach the "duplicates" later, but be sure they really are duplicates and not just references to the same record. Duplicates come from the same source website etc.

Most of the sources were actually for children in the family. Here is the after screenshot of the sources added.

There were several copies of the same marriage record but the record was of a marriage license and would not necessarily change the date of the marriage. In addition, some of the records had been mistranscribed and had the date of 15 December

The entry was now "cleaned up" but the only record of his birth and death dates came from the FindAGrave.com record that had no further source for the information. Further research is certainly necessary, especially for the children.

Here is the first part of this series:

http://rejoiceandbeexceedingglad.blogspot.com/2017/10/how-to-analyze-record-in-familysearch.html

How to Analyze a Record in the FamilySearch Family Tree: Part One -- The Initial Review

This record of Dr. David Shepherd Jr. has no supporting sources in the FamilySearch.org Family Tree website. Although the information appears to be detailed and complete, without sources, everything here is open to question and revision. Further, there is nothing showing that this person is even my relative. The titles of "Dr." and "Jr." make the entry look more certain, but where are the records that would support either designation?

The idea here is to carefully examine and analyze the information in the Family Tree for consistency, supporting sources, reality, and believability. It is also important to make sure the dates and places are standardized and that the places reflect the names of the locations at the time of the events. This process may seem difficult and time-consuming, but with practice, you can spot the problem areas almost immediately.

Some things in the Family Tree readily tell me that no one has been working on this entry with any degree of knowledge of the Family Tree or with any sophistication in genealogical research. They are also indicators that the information may be incomplete or inaccurate. This is both good and bad. It is good because that means that I have an opportunity to do some research and possibly add names to the Family Tree and bad because the amount of clean up necessary is probably overwhelming.

Of course, the first indication of a lack of involvement is the lack of sources does not automatically mean that the information is not correct. Research may support the entries or change them.

The lack of sources even more significantly shows that this entry has, so far, been neglected because there is a whole page of Record Hints waiting to be processed.

Here is the present status of Dr. Shepherd's family.

The blue icons show the availability of Record Hints. The purple icons indicate that there are no records and the red icons show serious data errors. It also appears, from the temple icons,  that someone has come through this part of the Family Tree and harvested the "green icons" but has yet to complete the temple work.

As I examine the entries further, I see a long list of alternate "Birth Names."

This is one of clearest indicators of the lack of attention to these entries. As I have written many times, these "Birth Names" are a remnant from the historical contributions to FamilySearch and its predecessors over the past hundred years or so. They are usually either inaccurate or duplicative. In some cases, not apparent here, these entries should be modified to reflect nicknames or other real alternative names. But in their present form, they should be deleted.

If you are unfamiliar with any of the procedures to work with the Family Tree at this level, I suggest you refer to The Family History Guide for step-by-step instructions.

Why do I care? My experience has shown that as I become involved in doing research on these parts of the Family Tree, I will inevitably find additional missing ordinances and individuals and can add them in.

So, the first steps are almost too obvious: standardize and clean up the entries and then evaluate and add in all of the available sources if the Record Hints turn out to be valid. Focusing on Dr. Shepherd, this is exactly what I started to do. Remember to correct or add in any information found in those sources.

Where did all this information come from if there are no sources? I am aware that there is a surname book about the family but at this point, I have yet to discover a copy anywhere. I speculate at this point that any such book has likely included this Vermont family in with Shepherds from Virginia or elsewhere, but perhaps not.

I intend to "clean up" this family and report on my progress. Stay tuned.

Serious FamilySearch Family Tree Challenges

It would seem from this view of the FamilySearch Family Tree that the entries are rather ordinary. Most of the entries appear to be somewhat complete and although there are some evident data errors with red icons, the rest of it appears to be manageable. Appearances can be deceiving. I spent two full days working on sorting it out just a few of the data problems with these families. I have not even begun to unravel the problems.

I have previously done some research on David Sheppard (b. 1750, d. 1832). The research was sufficient to establish in fact that his parents were unknown and to confirm his death date and place of burial. The family had been previously confused with one in Connecticut where the father had the same name. The main issues that remained concerned his children. Here is a screenshot of the family as it now appears:

Some of the children are shown to have been born in Castleton, Vermont. But two of the children are shown to have been born in Bennington, Vermont, approximately 65 miles away. Further research also shows that some of the family members moved to Geauga County and Cuyahoga County in Ohio in the 1830s. However, there are several large gaps in the family and we can presume that there may be additional children.

Since much of this information comes from traditional sources, i.e. submissions to FamilySearch and its predecessors over the past 100+ years, much of the information that was already in the Family Tree had to be verified with historical records. The real difficulties begin with the next generation. Here is another example:

The existing records show that Henry Sheppard had two wives both of whom he married in Cuyahoga County, Ohio. However, the only child shown to have been born to Henry and his wife Cynthia is shown as being born in Vermont. It is possible but unlikely that the couple traveled over 500 miles from Cuyahoga County to Castleton, Vermont to have a baby only to have the wife die and the husband travel back to Ohio and get married again the next year. This is especially true since Cuyahoga County was only established in 1810. Additionally, construction only began on the Erie Canal in 1817 so all of the travel would've had to have been done overland.

Further, records for Andrew McClain Shepherd, the child, show that he was born in Ohio. However, I've yet to find any record of his birth other than mentions in the US census records.

 These kinds of problems only become evident as I compare the sources attached with the information in the Family Tree. In this case, the connection between Andrew McClain Shepherd and his parents is completely missing. The only connection is that it may have been born in Ohio.

Significantly, every single child in this family has similar problems. The challenge here is that the information appears to be somewhat complete and supported by sources. It is only when the information is carefully examined that the missing information connecting the family members and the dates associated with each of the family members reveals that there is no supporting data for most of the information that has been inferred. For example, the existing information in the record shows that Andrew McClain Shepherd was born in Vermont however the only sources showed his birth in Ohio. The Ohio birth explained the problem of the traveling parents.

These examples show the importance of looking carefully at each source added to the record and correcting and adding the information to the details shown for the individuals. Many of the apparent contradictions can be resolved merely by following this procedure. The challenge, however, in dealing with the Family Tree, is that this procedure has to be applied consistently to every single member of the family. Given the condition of the Family Tree, there is no way to avoid doing this work.