What You Read May Shape What You Write

How reading sophisticated sentences impacts your writing

Posted Aug 15, 2016

Educators and employers alike are dissatisfied with writing skills in the classroom and office—and this dissatisfaction is rising. For example, P21, a coalition of civil society groups that work towards building 21st century skills noted that 26.2 percent of college students lacked basic writing skills. Five years ago, the authors of Academically Adrift, a book that posed the fundamental question of whether college was worth it, found only half of graduating students felt their writing was better than it had been in high school.

Even Ivy League compulsory writing programs have come in for criticism in the provocative Chronicle of Higher Education article, “Why Johnny Can’t Write, Even though He Went to Princeton,” which examined the yield from increased writing spread across undergraduate curricula—distinctly underwhelming. In addition, a 2015 survey conducted by the Society for Human Resource Management, an international organization with over 285,000 members world-wide, found that writing was the top skill requirement employers felt most recruits lacked.

Nevertheless, in 2016, more American colleges and universities required writing courses throughout their students’ degree programs than they did 40 years ago. Colleges and schools of business, in particular, have responded to widespread employer dissatisfaction about graduates' writing skills. For example, Stanford and the University of Rochester, among others, employed writing coaches to address student writing skills on assignment across the curriculum, while the Wharton School at the University of Pennsylvania recently bumped its communication requirement to 12 courses—the equivalent of a minor or even a major concentration at some institutions. So, why are the writing skills of students not improving?

Your Reading Influences Your Writing

The problem might have less to do with writing and more to do with reading. Teaching MBAs at the University of Florida, I started performing an experiment. I would read a student’s writing––often as little as a sample page––and I would pinpoint precisely what they read regularly.

Two students aiming for post-graduate careers in biomedicine, for example, acknowledged they read critically acclaimed fiction and classic literature frequently. A commercial banker was incredulous when I told him he must read The Economist and Wall Street Journal cover to cover each week. And a would-be entrepreneur guiltily admitted that, yes, he hadn’t read a book since high school.

Encouraged by this experiment, I surveyed 80 students on their reading habits across an MBA program in fall 2015. In the survey, students reported their regular reading from a range of sources. These sources included business-related general news magazines and newspapers (Forbes, The Economist, and The Wall Street Journal), non-fiction, genre fiction (as in mysteries, science fiction, fantasy novels), critically acclaimed or classic literary fiction, and academic research articles.  Students also identified online sources such as Tumblr, reddit, the Huffington Post, Ozy, Medium, and BuzzFeed.

What You Read May Be More Important than How Much You Write

The study, published in the May issue of the International Journal of Business Administration, found strong correlations between the complexity of students’ regular reading and the sophistication of their writing. One widely-used measure of assessing the sophistication of elementary and secondary school writers has been sentence length. Longer sentences generally include phrases and clauses that require writers to master sentence structure, grammar, and relationships between causes and effects. As a result, my study measured lengths of sentences and clauses. However, the study also analyzed the relative infrequency with which words occurred in publications––another measure of sophistication––to arrive at overall estimates of the complexity of writing.

Simply put, students who reported reading publications and books that scored highly on sentence and clause length, as well as on the complexity of the words used, had higher scores than students who reported reading materials that ranked low on the same measures. 

The whole thing seems so obvious, in retrospect. But, as I discovered when I started this study, researchers have focused on the influence of reading on speaking and between partners in conversation. Other studies have examined how writing skills in a first language influence writing in a second language. But no one seemed to have studied the impact of regular reading habits on writing in adults.

What Makes Writing Sophisticated?

The one challenge of this study was that tools for assessing writing are scarce. Readability formulas survive in some word processors as “Text Analysis.” The Flesch Reading Ease, Flesch-Kincaid, and FOG ostensibly assess the readability of your writing. But these formulas, as I discovered when I wrote The Reader’s Brain: How Neuroscience Can Make You a Better Writer rely relentlessly on counting, especially the syllables in words to assess complexity.

The problem, of course, is that a word like praxis has the same number of syllables as baseball. While most third graders understand baseball with ease, you’d be hard-pressed to find many Ph.D.s who can breezily explain the meaning of praxis.

Instead, I pressed into service software that could analyze 19 measures of the complexity of sentences––from sentence length to clauses, coordinating phrases, and nouns used as phrases, clauses, or adjectives. But I also needed a means to measure the sophistication of writers’ vocabularies to address the praxis/baseball issue and avoid merely counting syllables in words and words in sentences.

Currently, a widely-used platform known as Lexile assesses the reading level and degree of difficulty of over 100 million articles and books world-wide. In fact, Lexile not only measures the difficulty of books assigned throughout primary and secondary education but also scores the articles my students accessed in the likes of the New York Times and Forbes whenever they read them via a library database. Unlike other measures of reading difficulty, Lexile uses its source’s median lengths of sentences, combined with a score reflecting the relative infrequency with which the sentence’s words occur in Lexile’s scanned corpus of some 100 million published documents. Thus, a source like the New York Times, which targets readers with at least an undergraduate degree, averages Lexile scores in the 1400 range. In contrast, coverage of the same story in USA Today, written to be comprehensible to fifth graders, earns an average Lexile score of 1150. However, the same story, covered in online-only outlets, earns an 890 Lexile in The Huffington Post and a 910 Lexile on reddit. (In another study, I validated Lexile’s overall reliability against 19 features of syntactic complexity—more on that outcome later.)

You Write What You Read

In concrete terms, readers who devoured reddit, BuzzFeed, Tumbler, and the Huffington Post regularly but read nothing else, aside from textbooks in class, had the lowest scores on the length of their sentences and clauses, as well as the lowest Lexile scores. In contrast, students who read academic articles, non-fiction, or The Wall Street Journal, The Economist, and the New York Times had the highest scores on their writing.

Studies of people conversing suggest a kind of mirroring behavior may be at work, where our conversational rhythms and word choices reflect those of the people we converse with.

Other researchers have found an alternative explanation in structural priming––a phenomenon where we retain a strong memory of the sentence structures we read that may carry over into our writing.

Ultimately, my study's findings are consistent with over six decades of research on elementary and secondary school students into the influence of reading on writing, with researchers finding that reading proved as influential or even more influential on the quality of students' writing than the amount of writing their schooling required.

Assign Books and Articles, Not Textbooks

Regardless of what causes the influence of reading on writing, the implications are clear. Educators should require students to read more articles, in addition to challenging fiction and non-fiction books.  Unlike the often simplistic or stilted writing in so many textbooks, well-written articles and books lead to better writing. And these sources can also help students  experience reading for pleasure, even if they’ve sworn off reddit for the month.

Our students might well be turning to online content not just to stay connected. They might be sticking with Tumblr because textbooks have schooled them to see reading as something inherently unchallenging after you enter high school.

I still pinpoint who reads what. Now I just know I’m seeing a bona fide causal relationship: sophisticated reading equals better writing.