Word length, familiarity No. 1 predictor of readability
Mabel Vogel and Carleton Washburne of Winnetka, Illinois, were the first researchers to statistically correlate writing traits with readability.
In 1928, they published 19 writing attributes that reduced readability. The top three centered on word familiarity:
- Number of different words ( 77% correlation)
- Median number of words that appear on a list of the most frequently used words in the English language. (-70% correlation, which means familiar words increase readability)
- Number of wordsnot occurring on that list of familiar words (67% correlation)
Since Vogel and Washburne’s research, researchers have proven in the lab — again and again — that 1) word familiarity and word length and 2) sentence complexity and length are the two strongest indicators of reading ease.
Familiar words easier to understand.
Let’s take a look at some of those studies …
1. Comprehension increased with word familiarity and ease, found Ralph Ojemann in 1934. Four vocabulary factors affected comprehension:
- Percentage of words on the familiar words list
- Percentage of words understood by 70% to 90% of school children
- Average difficulty per word
- Average difficulty of different words
2. Jargon and hard words reduce comprehension more than all other factors combined, found Ralph Tyler and Edgar Dale in 1934.
The more difficult words a passage included, the harder it was to understand, they learned. They found that the top two predictors of comprehension were:
- Jargon, or number of different technical words
- Hard non-technical words
These two factors alone affected readability as much as all of the other 10 factors they reviewed combined.
3. “Vocabulary load is the most important [predictor of] difficulty,” wrote Irving Lorge in 1994. He based a readability index on just three elements, including number of hard words.
4. Hard words and sentence length alone predict comprehension with a 92% percent chance of accuracy, found Edgar Dale and Jeanne S. Chall.
In 1948, they published a readability formula measuring only words not found on a list of common words and average number of words per sentence. That formula scores among the highest of all readability formulas.
5. You can achieve a virtually perfect reading grade level correlation by measuring just the percentage of one-syllable words and three other elements, found Edmund B. Coleman in 1965.
Coleman used cloze scores, where people fill in the blanks in a sentence, instead of multiple-choice tests for his study. Cloze scores are considered a more accurate way to measure a person’s understanding of a passage.
6. Five word characteristics affect comprehension, found G.R. Klare in his1976 analysis of 36 readability studies:
- Proportion of functional words, as opposed to “content” words. Functional words include articles like the and a; pronouns like he, him, she and her; prepositions like on, under, against and during; conjunctions like for, and, nor, but, or, yet and so. These functional words make messages harder to understand.
- Word frequency, familiarity and length. Short, simple, recognizable words make copy easier to understand.
- Concreteness. Concrete words, like schnauzer and Pepto-Bismol, are easier to understand than abstract words, like bored and trouble.
- Association value. The easier the words are to remember, the easier the copy will be to understand.
- Nominalizations. Converting a word to a noun — like put out to output — decreases comprehension. So don’t commit verbicide.
7. Shortening and simplifying words and sentences made messages six grade levels easier to read, found Thomas Duffy and Paula Kabance in 1981. The revisions reduced the reading grade level from 11th to 5th, making the revised copy six grade levels easier to read.
Choose words to be read.
Bottom line? To make your copy easier to read and understand, choose words that are:
Sources: William H. DuBay, Unlocking Language, Impact Information (Costa Mesa, Calif.), 2006
DuBay, Working with Plain Language, Impact Information (Costa Mesa, Calif.), 2008
DuBay, Smart Language, Impact Information (Costa Mesa, Calif.), 2007
DuBay, The Principles of Readability, Impact Information (Costa Mesa, Calif.), 2004