One has but to Google the phrase "citing sources for genealogy" to find many wonderful takes on the subject. Perhaps my favorite is from familylocket.com; it's entitled "Source Citations: The Good, the Bad, and the Ugly." So what can I possibly add to the wealth of great advice already out there?
In keeping with a trend here, I'm going to begin addressing citations in part by discussing examples of what is less good. It's all very well to look at sterling exemplars and aspire to them in our own work, but "sterling" is not the standard you commonly find on Ancestry and collaborative trees online. Sometimes it helps to compare the abysmal, the excellent, and much in between. In that spirit, I am going to use a case study which concerns my great great aunt, Annie Ross (1867-1898), who tragically died of an as-yet unconfirmed illness at the age of 31. For her short obituary, I'll show versions of the source citation on a scale from abysmal to excellent.
At this point, I had better get a certain issue out of the way. While there are a number of components that belong in a good source citation (see below), the way in which these components are arranged in it can vary widely. This refers to a citation's style. Many historical disciplines use The Chicago Manual of Style. While there are multiple variations on this style, Elizabeth Shown Mills' Evidence Explained is widely considered to be the genealogical gold standard.
Every genealogical statement of fact must come with a source citation. The fact that we will assume here is the death date of Annie Ross: 7 September 1898. According to her Drouin baptism record, she was born "Marie Adelaide Ross" on 8 January 1867 in Maskinonge, Quebec. The evidence presented by multiple other documents suggests that she emigrated with her parents and siblings to Marquette, Michigan in about 1879. She lived in Baraga, Michigan from about 1883 until the time of her death. On 3 November 1887, she married William Levithure, with whom she had several children. She was commonly called "Annie," and occasionally "Devina."
For a long time I did not know Annie's exact death date, and had only family tradition and approximations to go on. So far I have been unable to locate a death record for her at local repositories (this search is ongoing!), but I have been able to find an obituary and a gravestone, which together present reasonably solid evidence as to her date of passing.
Here is my scale of abysmal to excellent for citing her obit (along with some annotations for each):
No citation whatsoever.
Comments: This is self-explanatory.
2. Not Much Better
"Yeah, I think I found it in a newspaper somewhere."
Comments: Even a bad citation is better than no citation. But our aim should be higher than providing bad citations.
3. Still Bad
"One of the local newspapers. From 1898 or 1899, I forget which."
Comments: This is only marginally better than the last example. We have an approximate date range, and we know to look at "one of the local newspapers" (in case we really couldn't figure this much out on our own), but the citation remains unacceptably imprecise. There is no publication title, no article title or description, no firm date, and no repository.
4. Not Terrible But Still Not Very Good
"Annie Ross's Obituary, L'Anse Sentinel, 1898."
Comments: Here we have a publication name and a rough designation for what the article is. We even have a hard date year. But we still don't have a specific date, a repository, and some other important information. How easily could the reader find this? It would take some digging.
"Obituary of 'Mrs. William Levitue,' L'Anse Sentinel, 10 September 1898."
Comments: Now we are approaching a respectable citation, but we're still not quite there. What we do have are several essential components. The untitled article's description is exact (down to its misspelling of Annie's married name), as are the publication title and date. But we still have no page number, no repository, no date of access, etc.
"Obituary of 'Mrs. William Levitue," L'Anse Sentinel, 10 September 1898, pg. 5; accessed at www.chroniclingamerica.loc.gov."
Comments: The added page number and repository put this in a class above the previous citation. Here we have very specific information (minus an optional column number) about where in the newspaper to find this article. We even know where specifically the researcher accessed the publication. But we're still missing some information about when this source was accessed, which is important since we now know it was seen online.
"Obituary of 'Mrs. William Levitue," L'Anse Sentinel, 10 September 1898, pg. 5; accessed at www.chroniclingamerica.loc.gov on 9 November 2019.
Comments: This is probably better than the majority of user-crafted citations you'll see in online trees. In most cases, someone looking at it would easily be able to find the referenced source again. Nearly all of the necessary components are there. What's missing? Let's go to our last example.
"Obituary of 'Mrs. William Levitue,' L'Anse Sentinel, 10 September 1898, pg. 5, column 2; digital image, Chronicling America: Historic American Newspapers Online Collection (https://www.chroniclingamerica.loc.gov; accessed 9 November 2019).
Comments: This citation is high on specifics and has every necessary component. (Unless your obit comes with a rare author name, this is typically left out.) Sure, one could argue about where to put Mrs. William Levitue's name, where to put "obituary," and debate several other points regarding style. But those things will be quibbles. As it is written, this citation not only explains where its source was found, it also makes clear the source's specific nature and how it was accessed. The experienced researcher knows that the date something was accessed online is important information, since websites can easily change or cease to be accessible. He/she also knows that it matters where or how easily something was accessed, and that explicitly describing the collection in the citation presents valuable information and further possibilities for research. He/she even knows that there are subtle differences between accessing an online image of something versus accessing an original copy, and that these differences can be important in certain situations.
Beginners often get nervous about citing their sources. After all, they may think, citations are intimidating and this takes away the fun of genealogy. I'm not going to lie: creating quality citations can be intimidating. I fuss over mine all the time. I still don't like some of those I have in various trees, and I'm always tweaking them. But I believe that this care is necessary for good research. Part of the fun of genealogy should be tackling the challenge of writing citations that will help to make your research stand on its own long after you're gone. Going back to my second post, why do you do genealogy after all? Is part of it that you want your work to have maximum value and impact for a very long time? If yes, then a big part of that is taking your citations seriously...and embracing opportunities to improve them.