The Douma readability formula is a Dutch-language adaptation of the well-known Flesch Reading Ease test. Because Dutch syllable and word lengths differ systematically from English, applying the original Flesch constants would give misleading results, so W. H. Douma recalibrated the formula in 1960 specifically for Dutch prose.
How it works
The score uses three counts: total words, total sentences, and total syllables. From these it derives average sentence length (words per sentence) and syllables per 100 words. The Douma formula is:
RE = 206.84 − 0.77 × (syllables per 100 words) − 0.93 × (words per sentence)Higher scores mean easier text. The scale runs from 0 (extremely difficult) to 100 (very easy), using the same bands as the English Flesch formula so scores are directly comparable across languages.
Syllables are estimated by counting vowel groups while treating Dutch digraphs such as ij, ei, ui, ou, oe, aa, and oo as single vowel sounds (one syllable nucleus each). This vowel-group method is a reliable approximation for native Dutch vocabulary.
Score bands — what each range means
| Score | Label (Dutch) | Typical audience |
|---|---|---|
| 90–100 | Zeer makkelijk | Young children, beginners |
| 70–90 | Makkelijk | General public |
| 60–70 | Redelijk makkelijk | Informed general readers |
| 50–60 | Redelijk moeilijk | Professional/higher-educated adults |
| 30–50 | Moeilijk | University level, specialist readers |
| 0–30 | Zeer moeilijk | Academic, legal, scientific text |
For most public-facing Dutch web copy, news articles, and government communications, target 60 or higher.
What drives the score down
Sentence length carries the heaviest weight in the Douma formula (the 0.93 coefficient). A single 40-word sentence lowers the average sentence length term sharply. Dutch writing, like German writing, tends toward long complex sentences with subordinate clauses stacked at the end — this is the primary culprit for low readability scores in formal Dutch copy.
Long compound nouns are the second factor. Dutch compounds like leesbaarheidsonderzoek (readability research), verzekeringsmaatschappij (insurance company), or belastingdienst (tax authority) are each one word but carry many syllables, pushing up the syllables-per-word ratio. Breaking them into phrases or using shorter synonyms lifts the score.
Practical example
- Easy: “De kat zit op de mat. De kat is moe.” — short sentences, single-syllable words → score near 95
- Hard: “Het leesbaarheidsonderzoek naar de begrijpelijkheid van overheidscommunicatie heeft aangetoond dat burgers moeite hebben met lang samengestelde zinnen die meerdere bijzinnen bevatten.” — one 25-word sentence, many long compounds → score near 20
How to improve a low score
- Break long sentences at natural clause boundaries — aim for an average of 15–18 words per sentence.
- Unpack compound nouns: verzekeringsmaatschappij → het verzekeringsbedrijf or de verzekeraar.
- Prefer short synonyms where available: mogelijkheid → kans, werkgever → baas.
- Put the main clause early, not at the end of a chain of subclauses.
- Use active voice; passive constructions tend to add syllables and words.