Why are citations so hard to do?

Question

On Wikitree's "Help: Sources" there are two examples that have left me confused: 1) Census, and 2) Web page.

<!--[if !supportLists]-->1)     <!--[endif]-->Census:  Wikitree’s example -- "US Census, 1900", database online. Home Township, Small Town, Washington, USA; pg. 100, family 10, dwelling 15, lines 150-157; June 1, 1900; National Archives Microfilm M-10, Roll 100.

Let’s face it the vast majority of us are cheapskates and we just pull our census reports up on Ancestry or FamilySearch.  So why wasn’t a FamilySearch example used?  Instead it appears this is citing the original source.  Is it ok to ignore FamilySearch/Ancestry and just cite the National Archives?

So I am testing my census acumen with one of my ancestors: Gilbert Russell Day, Anderson County, TX, 1900 Census.

Based on the above example I came up with this:  

“US Census, 1900”, database online.  Justice Precinct 3, Anderson County, TX, page 1A, House 10, line 25, Jun 1, 1900, National Archives Microfilm T623, FHL #1241607.

How did I do? Am I close?

 Or should I stop overthinking and just copy FamilySearch’s citation:

"United States Census, 1900", database with images, FamilySearch https://www.familysearch.org/ark:/61903/1:1:M3L6-BVW : 12 January 2022), Gilbert R Day, 1900.

Interesting that the two examples are so different. I have noticed FamilySearch’s citations are rarely consistent.  Is that because they put too much info in their citations?

<!--[if !supportLists]-->2)     <!--[endif]-->Webpages:  Wikitree’s example -- Wikipedia contributors. "[https://en.wikipedia.org/wiki/George_Washington George Washington]." Wikipedia, The Free Encyclopedia. Accessed 1 Jan 2016.

The example states “Wikipedia contributors”?  Do I assume that is who is responsible for the website?  Based on this URL:  https://sarpatriots.sar.org/patriot/display/106632 this is my attempt:

National Society of the Sons of the American Revolution. “[https://sarpatriots.sar.org/patriot/display/106632].”  Patriot Research System.  Accessed 22 May 2022. 

Your opinions will be greatly appreciated.  Thanks.  John

Comments

At least, the <!--[if !supportLists]-->1)     <!--[endif]--> text is just noise, and can be safely removed. It is apparently a MS Word "feature".

---

I appreciate your desire to construct good source citations.

Wikipedia provides "cut and paste" citations. Look at the list on the left-hand side of the page and click on "Cite This Page." I think the Chicago style source citation is the one most like the example provided on WikiTree's Help page. Whenever I can, I try to go to and cite the "source of the source" for Wikipedia articles by looking at the list of the sources at the bottom of the the article.

I admit to taking the easy route and just copying and pasting the FamilySearch-provided citations, which are sometimes incomplete. For Ancestry.com source citations, I rely on Ron Pavey's Sourcer app. 

For other sites which do not provide source citations, I use Evidence Explained to show me how to construct a good citation. Once I have it like I think it should be, I save it in a word-processing file so I can use it to form the same/similar source citations in the future. I purchased a Kindle copy of Evidence Explained and consider it a good investment. The website has some free examples but only a tiny fraction of what is in the book.

Answer 1

Hey John - I'm far from an expert and must confess to usually just copying and pasting the provided FamilySearch citation however recently (like just today), I found that a couple of FamilySearch citations I referenced in the past now contain broken links - I'm guessing something within the FamilySearch site changed and the original record was moved - I'm still investigating so don't know for sure.

So now I am thinking that as well as providing the FamilySearch citation, it's probably also a good idea to include the original source if that information is available.  I always do this when citing a source from Ancestry as this site usually supplies details of both their own database and the source of the original record.

Anyway - like I said, I'm far from an expert and still get pulled up for sloppy referencing from time to time, but that's my 5 cents worth.

Andrew

Comments

You might want to check out the sourcer app, because it produces citations that are a bit more verbose than the original ones: https://www.wikitree.com/wiki/Space:WikiTree_Sourcer

Answer 2

I thought I would try an experiment: can I find the specific census you are citing using your citation? Here are my results.

Family Search. Working from the catalog I was unsuccessful in finding the 1900 Census for Texas! Using I am pretty good at going from the catalog to a set of records. This needs some more investigating.

Ancestry. I did a search for 1900 Census from the catalog search. Found a link for it right off. Followed the link and started to browse the census, that functionality being on the search page. I could readily get to Texas, then Anderson county, then to Justice Precinct 3, then to sheet 1A. Your ancestor and family is family #6, not #10 (must be a typo), but it does begin at line 25. It took only about a minute to get here.

So, I conclude a well written citation takes you to the record, provided the catalog you are using is properly set up.

Comments

Re-visiting FamilySearch …

I eventually found it, but the final step was stepping through a set of 900 images. And it was in the Precinct and sheet you identified.

The key thing with FamilySearch was having the microfilm number.

The population schedules are cataloged with the Soundex for Texas in a group of 381 rolls of microfilm. The schedules are at the end of the list.

So, at this point I am developing an opinion that FamilySearch has a rather poorly designed catalog. The Ancestry catalog appears far superior.

When citing a record, identifying it should be independent of the repository where it is held.

---

If I remember correctly, that 1900 census specifically was rife with indexing errors on familysearch, so most of the links are now busted

(Not all, just the majority) until the figure out their issues and reindex.

That was just a bad example to pick.

Answer 3

George's example attempt was great, because he demonstrated that websites are unreliable. Some go bankrupt, some change their folder structure, etc, and so having all that extra info about the roll and image and which repository is actually going to help some future researcher locate that record far more reliably than the familysearch link. 

Evidence Explained is the gold standard for that, it explains why this is necessary, teaches a methodology for organizing and building citations and then gives about a million examples that you can flip to directly. 

Worth the money,  but there are a few free samples of "quick checks" if you Google for them. https://www.evidenceexplained.com/content/sample-quickcheck-models

Answer 4

I'd agree with some of the others that your citation for the 1900 census is better than just copying FamilySearch's because it would allow you to fairly easily find the record again if the link gets broken or you need to find it on another website. Personally, I would add two things to your citation though:

1. The name of the person and/or head of household whose record you're citing, transcribed from the image. This helps to find them, especially in cases where the enumerator spelled their name differently or got the name completely wrong.
2. I'd still include a link to FamilySearch or where ever you found the record. The link may not last forever, but in the meantime it makes it much easier for others looking at your research to find the record.

I'll also add that the more you get into the habit of doing citations like that, the easier it will get. It might always feel kind of tedious, but with the records you use frequently, you'll probably get to know what information to include and where to find it without having to think very much about it.

Answer 5

Figuring out how to generate citations yourself is definitely a useful learning experience for using WikiTree.

There are tools to help if you want to generate citations quickly without having to type anything though. The WikiTree Sourcer browser extension generates a citation in one click. It has a lot of options for how to lay out the citation. This is what the default options would generate for that source:

"United States Census, 1900"
Citing Affiliate Publication Number: T623; Line: 25; FHL microfilm: 1241607; Record number: 16306;
{{FamilySearch Record|M3L6-BVW}} (accessed 24 May 2022)
{{FamilySearch Image|S3HY-DT5S-1CC}} Image number 00333
Gilbert R Day (48), married head of household in Justice Precinct 3, Anderson, Texas, United States.

As you can see it:

• Uses the WikiTree Family Search template rather than a link
• Includes more information that the FamilySearch provided citation

Rob

Comments

Rob, don't suppose we could convince you to modify that to be the following, to match the EE quick check for database online First (Full) Reference Note?

FamilySearch (various contributors), "United States Census, 1900," database, Intellectual Reserve, Inc., FamilySearch ({{FamilySearch Record|M3L6-BVW}} : accessed 24 May 2022), entry for Gilbert R Day (48), married head of household in Justice Precinct 3, Anderson, Texas, United States; Citing Affiliate Publication Number: T623; Line: 25; FHL microfilm: 1241607; Record number: 16306;{{FamilySearch Image|S3HY-DT5S-1CC}} Image number 00333.

---

Hi Jonathan,

I'm happy to add new options for formatting. Surprisingly I have had very few requests for different formatting options.

There are some limitations. Because Sourcer has options to include the record "data" (the "item of interest" in EE speak) in various forms (a sentence, a list of fields/values or a table) it makes sense to always have that at the end of the citation.

A few comments on your example:

• FamilySearch ({{FamilySearch Record|M3L6-BVW}} : accessed 24 May 2022)
  This will display on the profile as:
  FamilySearch (FamilySearch Record: M3L6-BVW : accessed 24 May 2022)
  It seems rather unnecessary to have "FamilySearch" twice. Is that what you really want?
• Accessed date: There is already an option of whether to add an accessed date. I could add an option to put the parentheses around the whole link as in your example. It will look a bit odd if the option to put that section on its own line for readability is used - I guess that is why you like the extra "FamilySearch" before the open paren.
• Intellectual Reserve, Inc.
  This is pretty hidden on the FamilySearch web site and seems rather unnecessary to include to me. It isn't part of the FS provided citation. Isn't FamilySearch as the "website creator-owner" sufficient?
• "database"
  I could add an option to include this but these "database" and "database with images" phrases in citations seem rather redundant. When you have a link to FamilySearch or Ancestry isn't it obvious that it is an online database? And if there is an image and I include a link to the image isn't it obvious that there are images? In an example like Find a Grave, is it a database with images even if there is no image for this particular record? I guess to add an option I need to know the rule to follow.

---

I was simply trying to follow the Evidence Explained format.

I'm not entirely sure about the FamilySearch in the front, seems to make the most sense there, supposed to be the Compiler. You could make the argument that it's the LDS church, you could say it's really the Intellectual Reserve, that it's just members and therefore none of those.  

Accessed part was part of the citation format, hence why I put it in there, and where I did that.

"database" - yeah, might be obvious to us now, but it's part of the citation format to describe what it is, and it's for posterity.

Intellectual Reserve as the creator-owner, I think like the "compiler" part, you could probably make that call as they would all be the same for that website. I'm ok with just saying FamilySearch, but put the owner of the copyright for that example.

Answer 6

This is such a great question and helps me feel better about how long it took me to make my template models. Originally I thought, "How hard can this be?" How wrong I was. Eighteen months of drafting models later, I finally have a system I'm mostly comfortable with. 

There were about a dozen reasons why citing is hard but first let's begin with a simpler question. What standard do you aspire to cite? Are you planning to submit to a journal, are you a professional, or are you hoping to be one? If not, then what you need is a system for gathering enough way points that your future self can find the record again (in any format). I use Zotero for citations that don't need pro-level treatment. It's free, supported by open source software, and used by universities. My blogs have citations but are they perfect? No. They don't need to be and I don't need to spend hours making them perfect.

Citations are hard because the ways that we gather genealogical information are almost limitless. Aside from the big aggregators (among them Ancestry, FamilySearch, MyHeritage, Archivenet, etc.), was it an interview, an engraved cup, a photocopy of someone's family reunion speech, a family bible, volume one of a multi-volume parish register, a legal regulation, a set of binders your great-aunt bequeathed to you? Citing is hard because answering the question, "Where did this information come from?" is not straightforward. It often leads to, "...and where did that come from?" and so on down into the rabbit hole of origination. 

I'll share one story. My husband's family has a CD they distributed at a 1990s reunion. On it are the documents gathered to support their genealogical work to date. Not a single document had a citation. How would you cite any one of these? First, who created the record? Is it the person who made the CD or the person who created the digital copy of the record, or the person who created the original record? How do you know? (Answer: you research it.) It took me an hour to write a single citation.

If you're wondering why I spend so much of my time thinking about citations, I'm hoping to achieve my AG designation at some point. 

To close, I have respect and sympathy for anyone wrestling with citations. Thanks for asking this question.

Comments

Great example Linda. Similar situation for me, have a family genealogy that has an excellent narrative built up and mentions some sources used (although not for each of the facts listed) and then just lists of names with b/m/d dates back to 1770s. Clearly, my cousin back in 1940 spent a LOT of time finding the primary and secondary sources to put it together, but I wanted to help confirm it. So, I'm spending years here on Wikitree finding all of those people and the sources supporting those relationships so that my family doesn't have to guess anymore whether it is accurate. Found a few mistakes, but mostly it's spot on. Is my time finding and cataloguing those citations wasted? No, because although I trust the data my cousin compiled, I know that others shouldn't trust her or me to infallibly catalogue the particulars, they should be able to go find that record themselves, and I'm making that 100 times easier for them.

---

Absolutely. What is the first rule in genealogy? Never blindly accept another's work.

Trust but verify.

Your painstaking work does several tasks at once: i) assures YOU that prior work was correct - that you agree with the sources and the interpretation**; ii) may help you learn new source material; iii) brings confidence to the entire body of work; and iv) lays the groundwork for others.

**Document interpretation is its own beast. Did the family member interpret correctly? Is the evidence primary and direct or secondary and indirect? Is the record right or did the scribe make an error? Is there any possibility the handwriting (or blurred and obscured text) was mistakenly transcribed? Even if you have primary information, is the format subject to error, such as an index?

In my own family, I'm having fun exploring the truth behind the lore. It is somewhat surprising to me as a genealogist how many people before me - family, journalists, esteemed historians who only consulted authored material - fell into the same tropes without fact-checking. "Is this true? How can I prove it?"

There is only one way: by doing what you're doing. Sourcing the original evidence.

---

I am sure I am not doing this right.  I am responding to the responses I received.  Thank you for taking the time to help give me some insight into this issue. You were all very helpful, and I will follow up on some of your suggestions. <!--[if gte vml 1]> <![endif]--><!--[if !vml]--><!--[endif]-->

In the meantime, I came up with something.  I created a list of "questions" on an Excel spreadsheet and once answered they are used to create a citation.  Excel isn't the right place for it (it's cumbersome to edit) but it does build a nice consistent citation (unlike FamilySearch).  Now I will have to test it against some of the apps you suggested I try.  

I guess my REAL complaint is -- how inconsistent websites explain the Chicago Method.  Example: One site would tell you to list the author's first name and then their last name followed by a semicolon, while another site tells you to list the last name, then the first name followed by a comma.  So it looks to me that English Professors are lazy.  This isn't rocket science.  The Mormon Church could easily solve this issue by working with a few English Majors and just creating a "Geneology Method".  Then explain and use it on their sites.  The rest of us genealogy nerds would easily fall in line.

Thanks again to all of you for your insight and help.  I truly appreciate it.

---

Yes, it's inconsistent because while there are models and generally accepted practises, EE chief among them, one does not have to follow them provided one creates a consistent style in one's own work. In addition, citations can look weirdly inconsistent in application even if you create a citation model because you are missing some key bits of information. 

For example, you have an original studio photo of your ancestor. It is helpfully stamped with the name and address of the photographer. Great! You know the creator. It might look like this:

firstname lastname, studio portrait, 1915, city, state, [country if desired], [description of item including name and date], [yourname] fonds, [your address]. [If you want to add a second level of citation, you can add where you got it from but for simplicity I'll omit tier two.]

Now you have a second photo but there's no stamp or year. You don't know the creator. You have to omit the first items and attempt to describe the item instead:

[portrait of name], studio portrait, 1915, city, state, [country if desired], [description of item including name and date], [yourname] fonds, [your address].

Newspaper articles are the same. Know the author? Is there a title? No author or title? All of the possibilities make a citation look inconsistent even when you are being consistent.

Census records though are pretty consistent. Layer one follows this format if found on Ancestry:

[country if you work in more than one country], [yyyy] [Canada/England/United States] Census, city/town, province/state, population schedule, enumeration district no. xx, ward xx, city, pg. xx, dwelling no. xx, family no. xx, entry for xx household, digital image, Ancestry.com (https://www.ancestry.com : accessed date); citing [original reference ID info like you're going to the archives].

---

that's actually why I like Evidence Explained, because it's not really a style guide, a "right/wrong" way of citations. It's a methodology that says "what are the things I would need to know to find this again if I had to do so".

The examples are invaluable for saving time, but the first few chapters are really the important part of that work.