The purpose of this study was to determine the comparative validity of seven readability formulas and revisions of three of these formulas. The formulas employed were the following: the Lorge, the Yoakam, the Flesch reading ease, the Dale-Chall, the Farr-Jenkins-Paterson, the Tribe, the Fry readability graph, and the Powers-Sumner-Kearl revisions of the Flesch reading ease, the Dale-Chall, and the Farr-Jenk1nsPaterson formulas. The hypotheses tested in the study were that there were no significant differences (1) among readability scores predicted by the original formulas; (2) between each revised formula and its original version; (3) among the readability formulas and revisions of formulas developed since 1956. The formulas being tested were applied to six books from two series of high-interest, low-vocabulary readers to provide evidence of the consistency of measurements made by the formulas. The books were rated approximately third, fourth, and fifth grade levels by the publishers through the use of word lists. Analyses of variance, t-tests, and the Kendall coefficient of concordance (W) were run on the scores yielded by the formulas. All null hypotheses tested were rejected on the basis of statistical evidence derived from the data collected as a result of this study. The following conclusions were justified by the data provided in the study:
1. There were significant differences among readability formula scores in all combinations of formulas tested. This indicates that exact grade-level scores are not predicted by the readability formulas.
2. There was a significant agreement among formulas in respect to ranking the books in order of difficulty.
3. There was a significant difference among mean grade-level scores of the two reading series tested by all formulas and for all books. This indicates disagreement between publishers as to the actual difficulty of a grade-level designation.
4. There was a significant difference among the readability scores of only one of the two series of books tested. It is felt that the other series cannot be relied upon to provide reading material of increasing levels of difficulty.
It is suggested that the grade-level designations provided by the readability formulas or by the publishers should be used in conjunction with the teacher's experience, wisdom, and knowledge of her pupils rather than as absolute values.
Level of Degree
Teacher Education, Educational Leadership & Policy
First Committee Member (Chair)
L. Helen Walters
Second Committee Member
Third Committee Member
Harold D. Drummond
Fourth Committee Member
Miles V. Zintz
Edelman, Joanne Erwin. "A Comparative Validity Study of Readability Formulas." (1969). https://digitalrepository.unm.edu/educ_teelp_etds/371