Archive for December, 2017

Direct FJP Grading

December 9, 2017

By Nirmaldasan


James N. Farr, James J. Jenkins, and Donald G. Paterson suggested a New Reading Ease Index in their article of 1951 titled ‘Simplification of Flesch Reading Ease Formula’. They replaced the syllabic count in the Flesch formula with a monosyllabic count. This irked Rudolph Flesch himself and the readability expert George Klare. The creators of the new formula responded to the criticism and produced fresh data to show that both the formulae yielded ‘substantially equivalent results’.

Since the number of monosyllables is fewer than the number of syllables in any passage, the Farr-Jenkins-Paterson (FJP) formula is a fine simplification enjoying a high correlation of 0.93 with the Flesch formula.

This is how it works. Take a sample of 100 words from the passage to be tested for readability. Count the number of monosyllabic words (M). Also calculate the average words per sentence (AWS). Substitute the values in the formula:

FJP Reading Ease Index = 1.599*M – 1.015*AWS – 31.517

The formula yields a score which may be converted to Grade Levels by looking up a conversion table – the same that is used for the Flesch formula. Since the scoring system is the same, it becomes easy to compare the old and the new formulae. The authors tested the formula and found ‘perfect agreement for 237 of the 360 paragraphs’ with the Flesch formula. They say: “There is a disagreement of only one step for 119 paragraphs,” and add: “In only four instances is there a disagreement of two steps (in one instance the old index was ‘Fairly Easy’ and the new was ‘Fairly Difficult’, and in the other three instances the old index was ‘Standard’ and the new index was ‘Difficult’).”

This formula, in spite of all its decimal points, is not as intimidating as the Flesch formula. However, the use of the conversion table along with the formula is certainly a trouble that needs to be eliminated with little expense to accuracy.

Readability critics may say counting monosyllables is ‘baby talk’ or ‘primer style’. What will they say about counting non-monosyllables? Surely, they have to agree that this is neither ‘baby talk’ nor ‘primer style’. Then show them this exact equation: M (monosyllables) + N (non-monosyllables) = W (words). That may silence them.

But for those who wish to use the FJP formula without the conversion table, here is my simplification called Direct FJP Grading = 0.2*AWS + 0.3*N – 4.

AWS is the average words per sentence and N is the number of non-monosyllabic words in a passage of 100 words.