The Coh-Metrix Common Core Text Ease and Readability Assessor (T.E.R.A.) is a tool designed to analyze the “easability”
and readability, of texts and provide useful information about text features. Most current readability measures are based exclusively on two features of text; the difficulty of the sentences (usually measured by the number of words or clauses per sentence) and the familiarity of the words (usually measured by the frequency of the words in a large database of texts). There are many readability measures and though
each is slightly different, they all focus solely on the surface difficulty of the sentences and the words.
Coh-Metrix T.E.R.A. is very different.
T.E.R.A. analyzes text on five components: narrativity, syntactic simplicity, word concreteness, referential cohesion, and deep cohesion. For a given text, each of these
components is given an “ease” score, which shows how it compares to thousands of other texts. Foremost, T.E.R.A. affords an understanding of what makes a text cohesive (or not). What features make texts more or less cohesive varies from one text to the next. Because of this, we need to read texts closely to find those sections that are less cohesive. This tool can help us do that. Once these sections are found, we can work with students to help them recognize and overcome the obstacles that less cohesive texts might present.
The five components that make up the T.E.R.A. analysis are described below.
Narrativity: Narrativity seems intuitive: the more story-like a text the higher the narrativity score, and the easier the text. Though this is true, some texts will score high on narrativity and not seem very story like. Click here to see an example .
Syntactic Simplicity: Syntactic simplicity is measured through several indices such as average number of clauses per sentence, the number of words per sentence, and the number of words before the main verb of the main clause. Texts with fewer clauses, fewer words per sentence, and fewer words before the main verb will give a text a higher score for syntactic simplicity. T.E.R.A. also measures the similarity of the sentences within each paragraph. Paragraphs that contain sentences with similar structures and verb tenses are easier to read.
Word Concreteness: Concrete words (mask, spoon, forest, ammunition) are words that refer to things you can see, hear, taste, touch, feel, or smell. Abstract words (democracy, appear, success, joy) cannot easily be seen, heard, touched, felt or smelled. A text with relatively high numbers of concrete words is easier to read and will have a high word concreteness score.
Referential Cohesion: Referential cohesion is the overlap between words, word stems, or concepts from one sentence to another. When sentences and paragraphs have similar words or conceptual ideas, it is easier for the reader to make connections between those ideas. Sometimes, however, low cohesion is desirable if you want the reader to create connections to understand the text. Click here to see an example .
Deep Cohesion: Deep cohesion measures how well the events, ideas and information of the whole text are tied together. T.E.R.A. does this by measuring the different types of words that connect different parts of a text. These words are called connectives. There are different types of these connectives: time connectives such as after, earlier, before, during, while, later; causal connectives such as because, consequently, thus. Then there are additive connectives such as both, additionally, furthermore, moreover, what is more. There are also logical connectives; actually, as a result, due to. Finally, adversative connectives are words that connect two phrases or notions that on some level conflict with each other, such as “My favorite sport is baseball however I watch more football” or “Whales are not fish yet they spend their life in the water.” Some examples of adversative connectives are: but, yet, however, although, nevertheless. All of these connectives help to tie the events, ideas and information in the text together for the reader. Click here to see an example .
Conclusion: The dimensions in the T.E.R.A. analyzer are tools to help us look into a text in order to determine what might present difficulties for our students and thus how to help them. A low narrativity score means more uncommon words and possibly more information and ideas. If this is accompanied with a low concreteness score it would indicate less concrete and more abstract words. More abstract words might mean more abstract ideas as well. A low referential cohesion score means students might have trouble seeing how some sentences follow one from the other. If this is accompanied with a low narrativity score this would be even more likely. Just as low cohesion and narrativity scores might make it more difficult for students to fully understand how sentences build on each other, low deep cohesion scores might make it more difficult to comprehend how the ideas, events or information of the text as a whole fit together.
In each of these cases the tool signals what to look for as we prepare the text for our students. Used well it can help guide us to those portions of the text likely to present difficulty and thus where we want to situate our questions and our emphasis. No two complex texts are wholly alike; to prepare students for the complexity demands of the Common Core we need to regularly work with short high quality complex texts. In this way students will gradually come to recognize what makes text rich and complex and how to best absorb what these texts provide. This can only be done with teachers; the T.E.R.A. is designed to help them do this.
Flesch-Kincaid Grade Level
In addition to the 5 component scores, T.E.R.A. provides an estimate of a passage grade level using the Flesch-Kincaid Grade Level readability formula (Kincaid, Fishburne, Rogers, & Chissom, 1975). Traditional readability formulas such as the Flesch-Kincaid have been used for decades to provide educators with an estimate of the difficulty of a text in relation to the grade level or reading ability of the reader. In T.E.R.A., the Flesch-Kincaid Grade Level is computed using a combination of the length of the sentences and the number of letters in the words [(.39 * sentence length) + (11.8 * word length) - 15.59]. This grade level estimate correlates highly with other common readability formulas.
return to top
|