Statistically, 1919 Was a Bad Year

For this post, I’m pretending my blog is the op-ed page of the New York Times. This past year, beginning in January, the Times ran occasional opinion essays on the year 1919. I got the idea for this post when, last June, my history students completed a digital history project in which 1919 emerged, as a surprise to the class, as the second most turbulent year in 20th-century U.S. history.

In a January 2019 web article at, George Washington University professor Matthew Dallek asks, “Was 1968 America’s Bloodiest Year in Politics?” My high school history students tried to answer that question. More generally, my students tried to determine, quantitatively, what years within 20th-century U.S. history were the most turbulent. It turns out, based on their research, that 1968 was 4.5 times more turbulent than the year 1919, which edged out 1970 for second place in my students’ list of bad years. They were surprised at finding 1919 near the top. Toward the end of the school year, as the curriculum extended into the late 20th century, the class learned how the Vietnam War and the counterculture of the Sixties and early Seventies had torn the nation apart. With events like the Chicago “police riot,” Nixon’s election, and MLK’s and RFK’s assassinations, 1968 made sense to them. And, with Kent State and Nixon’s widening of the Vietnam War into Cambodia, 1970 made sense. But 1919, in my students’ minds anyway, hadn’t involved such turmoil. Did it really deserve to be runner-up to the bloodiest year in politics? My students’ research project, A Quantitative Ranking of Turbulent Years, which was based on levels of nationwide newspaper coverage of 20th-century events, shows that the year 1919 ranks near the top due to numerous stories about race riots, white supremacist terrorist attacks, violent strikes, and anarchists bombings.

Before researching the events of each year, the class first established a definition of the term “turbulence.” This was done in order to help students, who worked on the project both independently and in small groups, make consistent decisions about what events to include in the study. The definition, which students set parameters for by looking at web articles dealing with violence and civil unrest, included characteristic actions like riots, group violence, terrorism, insurrections, and large labor strikes. With this definition in hand, my students used Wikipedia articles and my school library’s digital reference collection to create a list of turbulent events for each year from 1900 to 1999.

In order to assign value to each event, and to ultimately arrive at a quantitative ranking of turbulent years, students searched for each event within and recorded into a spreadsheet the amount of coverage (i.e. search result occurrences) each event received. Because the number of newspapers within varies by year, students created a multiplier in order to equalize the opportunity for coverage. In other words, for those years where the number of papers was below the maximum number of papers, each event’s coverage was multiplied by the percent difference (e.g. x 1.35), while events occurring in the year with the most newspapers were simply multiplied by 1. Regardless of the year, therefore, each event had equal opportunity for coverage because total occurrences were increased by the multiplier. Once the multiplier was applied to each year and total coverage for the year was determined, that total was then multiplied by the number of events for the year. This was done in order to reduce the chances of any one individual event from overly influencing the shape of the final ranking.

In 1919, the Great Steel Strike, the Boston Police Strike, and the May Day Riot received the most newspaper coverage. For the year, the total number of qualifying events was nineteen, which far surpassed the overall per-year average of 2.4 for the rest of the century. While the Steel Strike of 1919 ranked seventh overall, behind events such as the Oklahoma City Bombing (1995), it’s the total number of different stories in 1919 that mostly explains how the year rates so high in terms of turbulence. Indeed, only 1968, with twenty-one turbulent events, has a greater total. In the article, Professor Dallek remarks that in 1968 Americans asked themselves, “Was the United States a society far more prone to violence than all other industrialized nations?” In 1919, Americans who read their daily newspaper certainly could have asked themselves the same question.