For the fourth straight year, Washington, D.C. is the most literate city in the United States, according to a recent study on literacy. The study, by Central Connecticut State University (CCSU), examined how well Americans used their literacy skills in the nation’s largest cities. Rounding out the top five were Seattle, Minneapolis, Atlanta and Pittsburgh.

CCSU ranked the cities based on six categories: bookstores, residents’ educational attainment, newspaper circulation, use of online resources, the library system, and periodical publishing resources. The most literate cities were largely in the Northeast, and they generally had a well-educated and well-paid population.

Click here to see America’s Most Literate Cities

Click here to see America’s Least Literate Cities

The focus of the study was not on reading test scores, but on reading culture, explained Dr. John W. Miller, head of the study and CCSU president. “This isn’t about whether or not people can read, it’s about whether they do read,” Miller said.

But while Dr. Miller’s study does not consider students’ reading test scores, education still plays a key role, said Miller. All but one of the 10 most literate cities were in the top quartile nationwide for the percentage of people with at least a bachelor’s degree. In Seattle, the second most literate city, 57.7% of the population had a college degree, the highest among all cities considered. But a well-read city should not just be made up of degree-holders, Miller stressed. “It’s not all one way,” Miller said, noting that cities where a large portion of residents are college educated can still slip in the rankings if, simultaneously, many other residents fail to complete high school.

At the other end of the spectrum, cities with poor literacy were also less educated. The 10 cities with the poorest reading habits were also in the bottom quartile nationwide for the percentage of people with a college degree and for the percentage of residents with a high school diploma. Many of these cities were in the bottom quartile nationwide for the percentage of people with a college degree, as well as for the percentage of residents with a high school diploma. In Anaheim and Fresno, less than 75% of residents had a high school diploma, while in Stockton just 17.9% of residents had a college degree — all among the worst rates for major cities.

Another factor the study found that often plays a role in the development of a highly-literate city is income, Miller noted. “Seattle, Minneapolis, they have higher income levels than a lot of cities.” Some cities, however, still scored well despite lower income levels. Miller highlighted New Orleans as an example. The city was in the top quartile for literacy, although its median household income was just $34,361, well below the U.S. median of $51,731.

In order to have a high-paid, well-educated workforce, cities must also have the jobs necessary to bring-in or retain that talent. According to Miller, “Cities that are more successful in terms of business development also have more a more educated population, [and] also have higher income.” In a number of the most literate cities, high-paying professional, scientific, and management occupations are especially prominent. Atlanta, San Francisco, Seattle, and Washington, D.C. all are among most literate cities and in the top five cities for such jobs — measured by percentage of total employment.

For the nation’s least literate cities, the opposite is often true. “Cities that are at the bottom have lower levels of business formation, [and] consequently lower levels of good jobs [and therefore] lower salaries,” Miller told 24/7 Wall St. California cities Fresno, Stockton, and Bakersfield, and El Paso, Texas, were all among the nation’s least literate cities. They also had among the 10 lowest proportions of professional occupations.

Based on the report published by Central Connecticut State University, 24/7 Wall St. reviewed the literacy ranking of 77 U.S. cities with populations of at least 250,000. The study reviewed city literacy based on six categories — library systems, bookstores, educational attainment, digital readership, and newspapers and other publications. Education metrics considered by Miller are from 2012, library system data is from fiscal year 2010, and circulation, publication, and bookstore figures are from 2013. To determine the availability of each reading material, the study measured circulation relative to the size of the population. 24/7 Wall St. also reviewed 2012 data from the Census Bureau, including income, and poverty.