Cloze Tests to Determine Reading Comprehension

Cloze Tests to Determine Reading Comprehension When teachers wish to measure how well a student comprehends a reading passage, they often turn to Cloze tests. In a Cloze test, the teacher removes a certain number of words that the student then needs to fill in as they read through the passage. For example, a language arts teacher might have their students fill in the blanks for the following reading passage: _____ mother is upset with _____ because I got caught  _____ a rainstorm. Sadly, I ______ my umbrella at home. _____ clothes got soaked. I ______ I wont get sick. Students are then instructed to fill in the blanks for the passage. Teachers are able to use the student’s answers to determine the reading level of the passage. Why Readability Formulas Are Not Enough While readability formulas can tell teachers how complex a reading passage is based on vocabulary and grammar, it does not reveal how difficult a passage might be in terms of reading comprehension. For example: He waved his hands.He waived his rights. If you were to run these sentences through readability formulas, they would have similar scores. However, it is obvious that while students might easily understand the first sentence, they might not comprehend the legal implications of the second. Therefore, we need a method to help teachers measure how difficult a particular passage is for students to comprehend. History of the Cloze Test In 1953, Wilson L. Taylor researched closure tasks as a method to determine reading comprehension. What he found was that having students use context clues from the surrounding words to fill in the blanks as in the example above has a high correlation with how readable the passage is for the student. He called this procedure a Cloze Test. Over time, researchers  have tested the Cloze method and found that it does indeed indicate reading comprehension levels.   How to Create a Typical Cloze Test There are a number of methods that teachers use to create Cloze tests. Following is one of the most common methods used: Replace every fifth word with a blank. This is where the students are to fill in the missing word.Have students write only one word in each blank. They are to work through the test making sure to write a word for each missing word in the passage.Encourage students to guess as they go through the test.Tell students that they do not need to worry about spelling errors as these will not be counted against them. Once you have administered a Cloze test, you will need to ‘grade’ it. As you explained to your students, misspellings are to be ignored. You are only looking for how well students understood what words to use based on contextual clues. However, in most instances, you will only count an answer as correct if the student answers with the exact missing word. In the example above, the correct answers should be:   My mother is upset with me because I got caught  in a rainstorm. Sadly, I left my umbrella at home. My clothes got soaked. I hope I wont get sick. Teachers can count up the number of errors and assign a percentage score based on the number of words that the student guessed correctly. According to Nielsen, a score of 60% or more indicates reasonable comprehension on the part of the student. Using Cloze Tests There are a number of ways that teachers can use Cloze Tests. One of the most effective uses of these tests is to help them make decisions about reading passages that they will be assigning to their students. The Cloze procedure can help them determine what passages to assign students, how long to give them to read specific passages, and how much they can expect students to comprehend on their own without additional input from the teacher. Note, however, that Cloze tests are diagnostic. Since they are not standard assignments testing a student’s understanding of  the  material that has been taught, the student’s percentage score should not be used when figuring out their final grade for the course. Source Jakob Nielsen, Cloze Test for Reading Comprehension. Nielsen  Norman  Group, February 2011