During my three years of teaching in England, it’s become apparent that students in my American history classes spend a lot of time worrying about access to primary sources. As an undergraduate myself, I knew that upper-level assessment turns on the ability to find and analyze primary sources—that stipulation is no different in this country. What’s clear, however, is that whereas as an undergraduate I could go to the stacks and wander around until I had a pile of travel narratives, diaries, and monograph reproductions related to early American Atlantic history, that’s not necessarily the case here for students taking classes on early American history, Native American history, the Atlantic world, slavery, trade, and colonization. The costs of a Readex subscription, just for example, are too expensive for the library to justify when most students aren’t going to research early American newspapers. As a result, I’ve had to think hard about how to ease students’ worries about locating sources.

Databases, lovely, lovely databases, have been the solution. Especially with the third year students (but I’m looking at you, year 2—you’re next), I’ve incorporated a weekly database exercise that asks students to take a look at a different online repository for primary source research on early American and Atlantic history. Discussion leaders for the week are responsible for taking students on a “tour” of the database; they have to explain how they came up with a question to ask the database (which is useful for thinking ahead to essays and final exams), offer tips about browser compatibility, make recommendations about searching advice, and provide examples of keywords that yielded useful sources. By the end of the semester students have a list of 10-15 databases that they can play around in to look for sources.

For instance, we’ve used Historical Texts for reports on English colonization, The Geography of Slavery in Virginia and Documenting the American South to discuss slave narratives and runaway slave ads, and Indian Affairs: Laws and Treaties, Yale Indian Papers Project, and A Continent Divided when we work on settler colonialism and imperial expansion.

This semester, I asked students to write a reflective essay about using databases. Here’s how I framed the assignment:

“In this assignment, I’m most interested in seeing how your brain works. Your job is to discuss how you have used an online database—from either last semester or this one, or from an independently-located database discussed with me beforehand. Your essay should explain how you formulated a research question using the database, and how your research question changed as a result of the database (or whether it did not change at all). You should then reflect on this experience to say something about the methods historians must use to write the history of early America and/or the Early Republic. “

My students’ essays prompted me to write this post, which, with their permission, uses excerpts from their papers to comment on these databases. And I hope it’s not too presumptuous to share my students’ wish list for the makings of a good database.

A lot of students were enthusiastic about the way these databases made sources easy to find, and readily available. In a paper about Founders Online, one student observed that although the database “can seem overwhelming . . . it is also very easy to narrow down your search.” She particularly liked that “all the letters and writings are transcribed and glossed,” which added “context to the letters” and provided “advice for further reading.” Another student who experimented with Papers of the War Department enthused about the site’s “tips” section, which gave him “an idea of how to find sources in the most efficient way.”

Other students asked good questions and made useful observations about how access to digital databases poses challenges to historians. A student reading the Cherokee Phoenix newspapers available through Georgia Historic Newspapers discovered that although “the quality of these images . . . contained warning notes prior to opening them that they contained ‘illegible text,’” she could “zoom in on the image without overly distorting the text,” and “any occasional words [she] struggled to read” could be discerned by “reading the sentence as a whole.” Another student who worked with A Continent Divided pondered whether databases created a tendency for historians to examine “the same old documents . . . again and again; meaning that the way in which historians use such sources may also become repetitive and unoriginal.” Although he raised this question mostly as a straw man, it paralleled similar concerns that libraries and archives must confront when choosing which of their materials to place online.

Some students did offer critiques of several databases. A student reading Southeastern Native American Documents liked the site in general, but at first “found it difficult to understand because it was not clearly laid out. For example, at the top of the page there is text not related to using the database and just a search bar in the middle of the page.” He wanted the website to “work more like Google: instead of giving me a list of sources with specific words in them, point me in the direction of sources that could be most beneficial.” Another student suggested that, with respect to the Cherokee Phoenix newspapers, more context was required in the search results returned: “The search results pick out passages in which the words searched for feature and present these with just a few words . . . as context . . . mak[ing] it difficult to establish exactly how the article relates to the search and whether it is of any real relevance.” Students who worked with the War Department papers were not always enthused by the challenges of paleography (which I’m less sympathetic about), but also complained about the difficulty of figuring out whether a particular item was available to view based off preliminary search results (and this is something I find tricky about that database, too).

Here are some of students’ main requests when using databases, which I hope might be helpful for me if I ever think about designing one, or for digital humanists just beginning to set one up:

• Always keep the search option accessible, no matter where on the website someone goes; don’t make students return to the main page to refine a search or begin a new one.
• Include a page with searching tips, explaining how your algorithm works, and how to make the most of it.
• If your database compiles sources from different archives, make clear in the search results which documents are transcribed, photographed, and available to view on your database.
• Tag documents with keywords to make clear which documents relate to each other, and to allow students to search by subject as well as keyword.
• When in doubt about how much information to return in search results, err on the side of more, rather than less.
• Make note of which browsers are compatible with your database.

Readers, do you have any databases you’ve found particularly useful for your own classes? If you’ve created a database, what advice would you give to people thinking about doing the same?