Beware the Derivative

My mind naturally compiles things–sorting like things into some kind of order. I see the attraction and admire the dedication of people who take records from a location and compile them into a collection that is easily researched. Most of what is searchable on is some form of compilation of records. We need them to do our research. I’m all for benefiting from the work of others. I’m learning to avoid becoming dependent on either the work of others or compilations.

Recently I have encountered examples of why it is not a good idea to take compiled data at face value, even when it is sourced. The first kind of error that creeps into compiled data is the transcription error: missing tick marks in the census, gender and race marked incorrectly, etc. To their credit allows users to add corrections on most fields (although not those). Nothing replaces looking at the image to verify the information for ourselves. I’ve done transcription for; it is challenging and easy to make mistakes.

Just today I discovered an error on a record on

Schoolfield, Henry incorrect year

Below is the image used to produce the above transcription.

Schoolfield, Henry missing year

There is no year. Other dates on this same page range from 1745 to 1768, but most are in the 1760’s. The document itself is a transcription done about 1899 from the original church records that were not filmed by the LDS church.

Another type of compilation is a book. Spotting errors in books is more of a challenge, and close to impossible to correct. All too often the publisher is defunct and/or the author is deceased, leaving no one to update any later editions.

One book I rely on a lot in my current research project is Land Records of Worcester County, Maryland 1666-1810 by Ruth T Dryden. This book represents a staggering amount of work drawing information from various sources into a readable format that is unique in its focus on individual tracts of land in the county. I was lucky enough to find a library willing to lend it long enough for me to photograph all the pages relating to my Schoolfields and Houstons. It is impossible for a book this size to be error-free. It’s a human thing.

Land record error

The above is from page 575 of the book that begins to track the history of this tract of land through 1810. The wording of the 1720 entry alerted me to double-check the Will since this entry indicates both recipients received the ‘upper’ part of the property.

Land record correct
Somerset County, Maryland Liber 16 page 96

Checking the Will that was recorded in the Will book, we see that the lower part was given to his brother Joseph. The next page does indeed show that the upper part of Smith’s First Choice was given to Armwell Robert Vigerous. Bear in mind the Will book is itself a transcription of the original Will. Sometimes Will books are transcriptions of transcriptions of earlier versions that have deteriorated, or were kept separately by the county. Some archives possess the original Wills, and will, for a fee, provide a photocopy or photo of it. If there is any doubt about the transcription, it can be worthwhile to attempt to get a digital image of the original to see if it is legible. The above transcription has handwriting from the time period, so it is likely to be the first derivative of the original. In this instance, the error was easily resolved by viewing this image.

Example 3:

Somerset County Maryland Marriage References and Family Relationships, 1666-1800 by Lyndeth Esgar (2013) is another compilation I take with a dose of caution.

marriage record mess

I was looking for clarification about Katherine Givan as the mother of John Schoolfield’s children. I’ve written about this family before because this John Schoolfield was a possible parent to my Dolly Schoolfield. The names John Schoolfield and Mary Richeson/Richardson are in the water supply as Dolly’s parents. I have a post devoted to debunking this.

I wanted to know why Katherine Givan was named as the mother of John’s children. If the above image from that book were to be believed, then George and Day Givens were her parents. I looked for and found a Will for George Givan on in Volume 25 Wills, Liber DD #4, pg 472 written 15 Oct 1748. The image is poor quality, but it reveals that George and Day Givan are brothers to Cathren Scoffel. George mentions his ‘cousin’ Thomas Givan Scoffel. Thomas Givens Schoolfield is mentioned in the Will of the above-mentioned John Schoolfield as his son. The Will was written 13 Jan 1772 and appears in Worcester Co Maryland Will Book 4 pg 132. I looked at the other Givan Wills from the correct time period and found that Katherine is the daughter of Robert Givan. His Will appears in Volume 21, Liber T & D, pg. 451 and is dated 24 May 1735. Katherine is not married at that point.

Also in that entry is the last name Gatchel for daughter Mary. It’s Satchel. The one that bugs me, though, is the name Miriam as one of the daughters. The Will Book shows the following:

Schoolfield, John list of children

One other time in this Will, her name is shown as Merin. Below is part of a marriage bond for Robert Schoolfield’s daughter Merin in Bracken County, Kentucky.

Schoolfield, Merin marriage bond Kentucky, County Marriage Records, 1783-1965

She appears to have been named for Robert’s sister. (It is likely that the witness John Schoolfield is Robert’s son.) Yet the name Miriam persists in records and trees.

That makes at least three errors in one entry from a compilation. One problem comes from at least one of the sources the author cites also being a compilation, making it a few generations from the original.

The lesson here is to view compilations as finding aids, not established facts. Go to the sources that are cited, then review them for accuracy. Get as close to the original as possible. Then cite the earliest version rather than the compilation.



Genealogy Citations For The Rest Of Us

SuperscriptI’m new to citing sources in genealogy. It’s not a natural or organic process for me at all. So consider me someone who is still surmounting the learning curve, as I suspect most family historians are. Part of me wishes I’d started citing my sources in the beginning. Yet in some ways I’m glad I didn’t because I would have spent far too much time doing something I didn’t enjoy, and might have lost interest in genealogy all together if I thought it was just too tedious to pursue.

I have no interest in becoming a professional genealogist. It’s too much work and I’m not built for having paying clients. Still, I see the value in citing sources. I have spent some time reading articles about this necessary function, and have arrived at some conclusions about a reasonable approach to keep from going crazy trying to get it right.


Here are some examples of a citation for a U.S. Census record.

This one is from Family Search:

United States, Bureau of the Census, 15th Census, 1930, “Kansas, 1930 Federal Census: Population Schedules”, Washington, D.C.: Bureau of the Census Micro-Film Laboratory, [199-?], [FHL film 2340440], E.D. 40-6, p. 30, Family History Library, 35 N. West Temple Stree, Salt Lake City, UT 84150-3400, USA !Name: Frank C. Dilts.

This one is from

1850 United State Census (Free Schedule), Pitt Township, Wyandot County, Ohio; p. 233, family 86, dwelling 79, lines 967-977; June 1, 1850; National Archives Microfilm M-19, Roll 719.

This one is from Evidence Explained (this is a short version):

1870 U.S. census, Wilkes Co., Ga., pop. sch., Washington P.O., p. 223 (stamped), dwell. 19, fams. 19-20, Cozart and Hogue.

Here’s the one I use. It’s what the Daughters of the American Revolution uses for its applications.

1850 U. S. Census, Illinois, Edgar Co, District Nineteen, pg. 116B, ln 27.

The DAR standard starts with the date, then the largest district: U.S. and lines up by the next largest district: State, County or Parish, Enumeration District  (abbreviated ED) or Township (abbreviated Twp), page number, then either family number or line number. It’s clean and simple and easy to remember once you get the hang of it.

What dictated the decision I made about what style of census citation to use is something I learned from another area of my life: Know Your Audience. I spend time helping people join the DAR, so more often than not, my audience is comprised of the staff genealogists at the DAR. Ask yourself under what circumstances would someone else be reading your citations? Citations for a peer review journal would be written entirely differently than citations that are hyper-linked in a family tree website. Different lineage applications have different citation conventions as well. It saves me headaches knowing I don’t have to make a complex citation for most real world applications.


Love, Love, Love the Citation Machine. It’s handy for citing books, newspapers, and websites. I just plug in the information I have from a title page for a book, for example, and it creates the citation for me. It does require proofreading the finished product, but it saves me having to remember or look up the style details every time I want to cite something. I did learn early on to request title pages for books, although I just did forget to do that the other day. Fortunately, I have a fair number of title page images on my hard drive, so I don’t have to re-search for them whenever I need one. Using the Citation Machine makes short work of producing readable citations.

Clearly not everything we need to cite in genealogy fits nicely into the census, book, website, newspaper system. We encounter all kinds of interesting records that have no easy way to cite.

Some websites do give you a ready-made citation for the record.

Family Search does that for the online images they provide.

Family Search Citation box

See the nifty citation at the bottom of the record description? Just copy and paste that into foot or end notes, family group sheet or citation notes in genealogy software, and it’s good to go. It has the added advantage of having a URL imbedded in it, so if the citation gets published on a website or in a .pdf, all the reader has to do is click on it to be taken directly to the record. How cool is that? is a great place to get already composed citations.

Ancestry citation

I copy and pasted the information in the box below the ‘Save’ button on the results page, and added the Accession number, and voila, I had my citation. In the absence of a subscription to Ancestry, or some other site that gives this kind of composed citation, I’ve learned to look for a composed citation to use at the website that provided the document. If one isn’t there, then I try searching “How do I cite (fill in the blank)? I just did Google: How do I cite BLM-GLO records? I found this handy guide from the St. Louis Genealogical Society. Help is out there. This doesn’t have to be difficult.


Seriously, that is the best advice I’ve seen yet on all the websites and documents I’ve read and videos I’ve watched on citations. Whatever my style is, I need to commit to it and make it my own. My non-conformist streak kicks in whenever I try to force myself to use someone else’s system for doing pretty much anything from exercise to cooking to citations. As long as I’m consistent, then I believe I’ll be okay. Developing my own style for citations is a work in progress. Like any work in progress, I have learned to give myself permission to get it wrong. Making mistakes is how we learn. I’m not going to be submitting my work for peer review publication, so as long as my citation informs me and any potential reader how to find the record I referenced, then the citation is valid.