Student-Generated Comprehension Questions

February 8, 2012

Paul Emmerson, the TEFL writer / trainer has an excellent section on his website with tips on using authentic texts in class which involve zero preparation.

I’d been looking for ways to make work incorporating authentic reading and listening texts purposeful and fun without depending too heavily on worksheets (last thing my production workers want either before or after their 9-hour shift is a worksheet; and part of the ongoing “unplugging” of my teaching!) and was delighted to find Paul’s ideas. I’d like to add to Paul’s ideas with a couple of my own.

Image-based Text Summaries.
Show students a collection of photos related to the article / report that you’re going to use (a lot of online newspapers have news galleries that you can use – see the Guardian ones, for example). Instruct students to write down a few words /phrases that come to mind with each picture. Get students to pool their ideas and write a 2-or-3-sentence summary of what they expect to see / hear in the story. When students are done, give them the text to check their ideas. They can then amend or add to their summaries.

Key Info Prediction Task
Activate students’ prior knowledge and topic knowledge by looking at the headline and photo of a news article and pre-teach any necessary vocabulary. Then, write some of the key details from the text on the board and have students try to guess what the details refer to – students can discuss their predictions pyramid style and come up with a final list. Then, give out the text and get students to check their answers.

Jumbled Articles
Find two different news articles and cut them up into sections which each contain 2 or 3 sentences. Give each pair of students the two jumbled stories and have them separate and order the texts. Once they have finished (and you have checked that the texts are in the right order) give one of the complete texts to one of the pair, and the other complete text to the other partner (i.e. text 1 to student A, and text 2 to student B) and have them write 3 or 4 comprehension questions based on their text. Finally, have students exchange texts and questions and get the students to answer each other’s questions about the text.

Typical Texts
With texts that tend to have a fairly predictable structure, such as film reviews or job ads, have students write questions that they expect the text to answer, then give out the text and have students answer their own questions (i.e. What happens in the film? Who stars in the film? What’s the salary? What are the main job responsibilities?) Here’s a sample text that you could use for this:

Would love to hear more ideas from you…

The Reading Process Riddle

July 31, 2009

Yesterday, my wonderful and beautiful wife (who, incidentally, is also an amazing photographer) forwarded me an email containing the old “Count the number of F’s” brainteaser. Of course, I didn’t fall for the puzzle this time, owing to the fact that I had done it before. However, it occurred to me that the puzzle does a great job illustrating a couple of important points for language teachers about the reading process.

The Puzzle
But, before we get on to the boring teacher stuff, here’s the puzzle, in case you haven’t done it before:

Quickly count the F’s n the next text – how many are there?


Most people only count 3, failing to count the F’s in the 3 Of’s. There are actually 6 F’s (I think!).

For Language Teachers

Now, 3 reasons why I deem this to be worth blogging about:

1. It shows that when fluent readers read, they don’t process texts letter by letter, word by word – rather they take in all the letters simultaneously, recognising all the letters in the word at once. Furthermore, research suggests that we recognise related pairs of words more quickly than unrelated pairs or words.

2. It demonstrates that fluent readers don’t process all of the words in a text – according to studies, we process fewer than half of the function words (words such as of, the, and to which don’t contain lexical meaning) in a text and around 80% of the content words.

3. More generally, it highlights the inadequacy of bottom-up approaches (on their own) for explaining the reading process and provides support for the view that reading more is likely characterised by the continuous interaction between top-down and bottom-up processing skills.

As Celce-Murcia and Olshtain (2000) point out:

“Good and effective reading must…be viewed as combining both rapid and accurate recognition and decoding of letters, words, collocations, and other structural cues with sensible, global predictions related to the text as a whole”

There are obvious implications here for work that we do with written texts in the classroom.

Now, what about the one about the colour and the tool?