University of Twente Student Theses


The Flesch Readability Formula: Still alive or still life?

Wolf, R.A. (2013) The Flesch Readability Formula: Still alive or still life?

[img] PDF
Abstract:Ever since schools divided students into grades, readability has been a subject of importance. By being able to quickly and easily match a text to a learner, a lot of time would be saved for the teacher, and the amount of potential study material would vastly increase. L.A. Sherman was the first to study readability by analyzing literature from a scientific perspective (1897). After careful research, he came to the conclusion that average word length and average sentence length correlate highly with how easily texts can be read by a certain target audience. Based on this work (and work by several others, most notably E.L. Thorndike), several readability formulas came to existence, the most important of which was presented in 1943 by Rudolph Flesch in his PhD dissertation. In 1948, the formula was revised to look like this: RE Score = 206.835 – (1.015 x ASL) – (84.6 x ASW) In this formula, ASL (Average Sentence Length) is the average number of words per sentence, and ASW (Average of Syllables per Word) is the average number of syllables per word. Since this publication, the formula has been applied to a number of subjects and studies. For example, Florida law requires that any legal contract has a RE score of 45 or above, a score that corresponds to a difficulty suitable for high school graduates. However, despite the frequent use of the formula, not much has been undertaken in forms of validation of the formula. G.H. McLaughlin states in his 1974 article that until then, only six validation studies had been undertaken. To solve this, my bachelor thesis will once again examine the validity of the Flesch Readability study. The study consisted of two parts, namely two different reading tests which were administered. The first test was a pre-validated test created by the American state of Texas, while the second test consisted of five texts taken from British municipal websites, on five different subjects. Of this last test, there were five different versions, bringing the total up to 25 different texts. Each participant was administered the pre-validated test and one of the five versions of the ‘website test’. When the testing was finished, the results from the first test was used to validate the Flesch formula. The results were surprising: It only managed to attain a correlation of r = 0.075. Apart from this, an ability score was assigned to each participant based on the performance on the first test. The results from the second test were analysed using both the RE score of the texts and the ability scores of the participants. This analysis confirmed that the RE score holds very little predictive value. The ability score on the other hand was a good predictor for the number of questions participants could answer correctly on a reading test. While the study was not flawless, it does seem there is no life left in the Flesch Readability formula. Other, later formulas may have stood the test of time, and it is time to move on to them, or perhaps create something entirely new.
Item Type:Essay (Bachelor)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:77 psychology
Programme:Psychology BSc (56604)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page