In libraries, newspapers are treated differently than periodicals (even though newspapers fit the definition of "periodicals" in most respects). The main reason for this difference is that the methods of acquiring, describing, organizing, displaying, preserving, and providing ongoing access to newspapers are very different than the methods used for journals and magazines. Even when newspapers have been digitized, the methods of digitization are very different than the methods of digitizing other periodicals, and digitized newspapers are usually found in specially designed collections with interfaces intended to support the unique demands of searching digitized newspapers.

From the researcher's point of view, a major difference between newspapers and other periodicals is that most newspapers were never indexed, and only one American newspaper (New York Times) was indexed in its entirety. What this meant for researchers was that the only method of discovering articles was by browsing newspapers, issue by issue, page by page, article by article.

Obviously, digitization of newspapers has significantly improved matters for researchers, though even digitized newspapers present challenges that you will probably not encounter with digitized journals and magazines.

Some challenges to expect:

1. Not all newspapers have been digitized. Although it may seem as though everything is online, the reality is that a very large number of newspapers are not yet digitized, and access to many digitized newspapers is restricted by paywalls. For the United States, digitization of (extant) eighteenth and nineteenth century newspapers is much closer to complete than is digitization of twentieth century newspapers. For twentieth century newspapers, you'll primarily find major metropolitan titles through the Library. Tens of thousands more are available through subscription-based services like Newspapers.com, but most of those newspapers are not available through the University Library (at least not the digitized copies) because these companies either do not offer institutional subscriptions, or else the institutional subscriptions are prohibitively expensive (on purpose, because these companies make most of their money by selling subscriptions directly to researchers).

2. Accuracy of keyword searches varies wildly. Newspapers are notoriously difficult to digitize, due to factors like their large format, multi-column layout, article jump-continuations, variety of fonts and graphical material, lack of standard page layout, sloppy presswork, old type, torn pages, crumbling pages, badly mended tears, stained pages, dirty pages, faded ink, clipped articles, bleed-through, gutter shadow, creases, non-standardized orthography (especially in pre-twentieth century newspapers), and other factors, any combination of which will reduce the accuracy of keyword searches.

3. The concept of news. The most common mistake we see students make when working with historical newspapers is applying a twenty-first century idea about "news" to older newspapers. Prior to the 1830s, "news" was commercial and political intelligence, and remember that the latter category would be political intelligence for those who held the franchise.¹ Newspapers began to cover the lives of working Americans, and to court them as subscribers, in the 1830s.

4. Vocabulary. Language changes quickly, and to retrieve newspapers in digitized newspaper collections, you must use the language that was current during the time period you ar researching. For example, in nineteenth century newspapers, strikes were often called "turnouts". Immigrants from China were often called "celestials". If searching for names, bear in mind that first and middle initials were often used in lieu of first names. A married woman was often referenced by her husband's name, e.g. "Mrs. Henry Smith". People could be referenced by their legal first names (e.g. Donald), or by a shortened form (Don, Donny, Donnie), or even by a nickname. You would need to use all forms of the name to make sure you retrieve all relevant articles.

Digitized newspaper collections tend to make rather extravagant claims about the documentary value of newspapers. For example, America's Historical Newspapers claims that it "Chronicles the evolution of American culture and daily life from 1690 to the recent past". Commercial newspaper vendors usually argue that historical newspapers offer some privileged view of "daily life" or "social history", or cite the old cliché that newspapers are the "first draft of history", or a "window onto the past", all of which is ironic since, until the past few decades, historians tended to view newspapers as doubtful sources of information.²

For the major metropolitan newspapers of the post-Civil War era, and especially the twentieth century, the best place to begin is:

For pre-twentieth century American newspapers, the two best collections are:

1. Andrew Pettegree, The Invention of News: How the World Came to Know about Itself (New Haven, Conn.: Yale University Press, 2014); Ross Eaman, Historical Dictionary of Journalism (Lanham, Md.: Scarecrow Press, 2009); Bob Franklin, Key Concepts in Journalism Studies (London: SAGE, 2005).

2. Lucy Maynard Salmon, The Newspaper and the Historian (New York: Oxford University Press, 1923).