Incorrect Gunning Fog Calculation

[bradfordlynch]

In the FOG calculation, complex words are defined as those words having 3 or more syllables. Not that Wikipedia is the best source of information, but see list item 3 here. But, as you can see in the snippet below from textstat.py the threshold for a difficult word is: not in the easy_word_set and more than one syllable. (Lines 254-255)

    @repoze.lru.lru_cache(maxsize=128)
    def difficult_words(self, text):
        text_list = re.findall(r"[\w\='‘’]+", text.lower())
        diff_words_set = set()
        for value in text_list:
            if value not in easy_word_set:
                if self.syllable_count(value) > 1:
                    diff_words_set.add(value)
        return len(diff_words_set)

    @repoze.lru.lru_cache(maxsize=128)
    def dale_chall_readability_score(self, text):
        word_count = self.lexicon_count(text)
        count = word_count - self.difficult_words(text)

        try:
            per = float(count) / float(word_count) * 100
        except ZeroDivisionError:
            return 0.0

        difficult_words = 100 - per

        score = (
            (0.1579 * difficult_words)
            + (0.0496 * self.avg_sentence_length(text)))

        if difficult_words > 5:
            score += 3.6365
        return legacy_round(score, 2)

    @repoze.lru.lru_cache(maxsize=128)
    def gunning_fog(self, text):
        try:
            per_diff_words = (
                (self.difficult_words(text) / self.lexicon_count(text) * 100))

            grade = 0.4 * (self.avg_sentence_length(text) + per_diff_words)
            return legacy_round(grade, 2)
        except ZeroDivisionError:
            return 0.0`

[bradfordlynch]

I created PR #71 for a fix to this.