Tag Archives: reading tasks

What are reading skills? –They’re not (only) what you think


Enter a caption

When we talk about reading skills, what usually comes to mind? Prediction, reading for gist, reading for specific information and skimming and scanning?

If you’ll forgive the clickbait style title of this post (I couldn’t resist), the fact is that there’s a lot more to the skill of reading than this.

Let’s start by looking at these commonly referred to skills, and then look at what else we could (and should) be teaching our students.


Students are very often encouraged to use the title of the text, or any accompanying pictures to make predictions about the content. There are good reasons for teaching students to do this; the idea is to activate whatever students may already know about the topic and help them to start creating a context. Without context, comprehension is much harder. Think about those moments when you turn on the radio and have no idea for a few moments what is being discussed. It only starts to make sense once you realise what the basic topic is.

Reading for gist

This doesn’t mean telling the students , ‘read the text to get a general idea of what it’s about’. That is never a very helpful task because it doesn’t give the students any reason to read, or any orientation to the topic or context. A gist question should mean that the students have to read the whole text, though not necessarily very thoroughly, in order to answer the question. For example, students might choose which title or picture most closely matches the content of the text. The idea is for students to get an overview of the text, which can help them when they return to the text to find more detailed information.

Reading for specific information

In contrast, reading for specific information means setting questions or a task which require the students to read the text much more carefully. This isn’t really a skill so much as a way of testing comprehension.

Skimming and scanning

These two words sound nice together, and make a memorable pair, but most people are a little vague about what they actually mean. Skimming has some crossover with reading for gist, because they are both about getting an overview. However, I would suggest that skimming and scanning are more properly called strategies rather than skills. They were originally both methods of speed reading. Skimming strategies are often taught as part of EAP (English for Academic Purposes) because these students have to read a great deal of content in English. So students may be taught, for example, to read just the first line of every paragraph, as this is often the topic sentence, which contains the main idea of the paragraph.

Scanning is a way of reading text quickly by only looking for specific bits of information, and not reading everything. For example, looking at the index or contents page of a book to find a specific topic you’re interested in, or picking out the figures in a text to see what the results of a piece of research were.  Again, this strategy can be particularly useful in academic contexts.

Understanding that we read different texts in different ways, using different strategies, is very important for learners. The way we read a novel on the beach is very different from the way we read a legal contract. We probably read the novel much faster, with more of a skimming/gist approach, whereas we are likely to read the contract carefully and slowly, checking that we understand the details. Students reading in a second language often fail to transfer these different ways of reading across from their first language.

Teaching reading skills

However, while looking at different ways of reading, and different reasons for reading,  is important, to what extent are these traditional reading activities actually helping students to read more effectively or fluently? Aren’t we mostly just giving them practice and/or checking their comprehension?

A reading lesson usually goes something like this:

  1. Set the scene/pre-teach vocab/elicit predictions.
  2. Read for gist (hopefully by setting a suitable gist question)
  3. Read again to answer for detailed comprehension questions.
  4. Discuss the content/focus on language in the text.

By encouraging students to predict and to read for gist first, we are perhaps helping them to develop their top-down skills- using context and their previous knowledge to make sense of what they are reading. However, there is a lot more to the reading skill than this.

According to Grabe and Stoller (2011:23) in each and every two seconds of reading fluent readers:

‘1 focus on and access 8-10 word meanings

2 parse a clause for information and form a meaning unit

3 figure out how to connect a new meaning unit into the growing text model

4 check interpretation of the information according to their purposes, feelings, attitudes and background expectations as needed

5 monitor their comprehension, make appropriate inferences, shift strategies and repair misunderstanding, as needed.

6 resolve ambiguities, address difficulties and critique text information, as needed.’

Fluent reading means that readers must be able to carry out all of these reading comprehension processes simultaneously and very quickly.  Just like driving a car, some of the processes taking place simultaneously must be automatized. If the reader is struggling to decode the words, or understand how the words fit into the sentence, or how the sentence fits into the discourse as a whole, their working memory will become overloaded, and they won’t be able to hold onto the overall meaning of the text.

So helping students to develop their reading skills must, I believe, include specific and conscious work on bottom up or decoding skills as well as top down skills, such as predicting.

Decoding skills

Whereas top down skills start from the reader and what the reader already knows or understands, bottom up or decoding skills start from the text. At the lowest level, students need to recognise alphabetic letters and then the words formed from these letters. Clearly, recognising and understanding written words is a key part of reading, even though we would usually deal with it under the heading of vocabulary. According to Paul Nation, in order to understand a written text without any assistance, students need to comprehend 98% of the words. So, if we want to work on developing reading skills (rather than using the text as a way of introducing new language) it is important that students aren’t getting stuck on too many unknown words.

Assuming that students recognise and understand enough of the vocabulary, the next layer of difficulty comes in putting those words together, understanding sentences, how those sentences fit together, and how what we are reading links with what we’ve already read. This is an area where there is huge potential to help students, and yet, outside academic reading courses, it is rarely a focus.

This is why I was so excited to work on Oxford University Press’s new series, Navigate. In this series, there is a clear syllabus and focus on these kinds of decoding skills. At lower levels, this might be understanding conjunctions or pronoun reference. At higher levels, we also focus on understanding complex sentences, ellipsis, paraphrasing and so on.

If we don’t focus on these kinds of skills, we are only teaching students part of what they need to know in order to improve their reading skills, and most of our focus is on testing comprehension rather than teaching.

Other posts on reading:

Learning vocabulary through reading

A lesson from my great-grandfather’s ELT coursebook- extensive reading and vocabulary

Developing meaning-building skills in reading

Repeated reading- some activities to build reading fluency

Listening while reading: can it actually help develop reading skills?

What kind of reading tasks might be worthwhile?

Creating effective reading tasks

Should we pre-teach vocabulary before reading, and if so how?







Filed under classroom ideas, Materials development, Reading, Teaching methodology, Uncategorized

Creating effective reading tasks

In my previous post, I looked at the whole idea of reading tasks and the arguments for and against using them. One conclusion I came to is that, ideally, tasks should bear some relation to the kinds of processes we would use when reading in real life contexts.

There are some quite simple task types which will work for a lot of texts. For example, setting up the topic (using pictures perhaps, or the title) and then asking students think of questions that they would like answered about the topic. Alternatively, ask them to write down what they expect to learn and then read to check their ideas.  This kind of approach has the obvious advantage of no preparation, and is very close to what we would do in the real world.

Similarly, with different kinds of texts, particularly factual ones, we can set tasks which require students to scan and find particular bits of information. Classic examples are small ads, tv listings etc.

More typical comprehension questions,such as we see in EL exams, tend to test what the question setter sees as the main points of the text. Of course, in real life, we are able to decide for ourselves what the main points are, which means that pre-set questions are intrinsically somewhat inauthentic. The format of typical question types can also be rather inauthentic, of course. While at IATEFL this year, I saw John Field, talking about listening, and he made the very valid point that when carrying out multiple choice questions, we are asking students to do something much more complex than simply listen for information. As well as listening for the information we require, they also have to pick up on the distractors and discount them, a sophisticated process.

Some traditional question types are closer to real world reading than others. For example, multiple matching, which is becoming a very popular exam type question. This is a task type which can work very well with short texts. In a comment on my last post, ESL Notes, asked about a possible reading task for these very short tasks (death by twitter). I think multiple matching is ideal for these kinds of texts. For example, for theseparticular texts, you could ask ‘Which text is about a fuel related tragedy?’ ‘Which text tells us about a possible miracle?’ (maybe a rather depressing lesson though? 😉 )

Form completion, note completion and summary completion can also be quite realistic, depending on the students’ context. Form completion is important in an ESOL context, note and summary completion are skills needed by students studying in the English medium.

Other tasks are not necesssarily realistic, but do have a definite value in terms of the skills they teach. For example, matching headings or summaries to paragraphs, which encourages students to read in such a way as to grasp the main point of each paragraph. Incidentally, although it isn’t foolproof, it is worth pointing out to students that the main point is often in the first sentence of the paragraph (the topic sentence).

I am also quite keen on tasks where students have to put in missing sentences or even paragraphs, as these really help develop awareness of how texts are constructed. However, they are quite difficult to construct as you have to make sure that there are very clear clues which guide them to the right answer.

This is the first part (there is more text and more gaps) of an example from Real Life Advanced, an Upper Secondary book, published by Macmillan. The original article is here.

Mark Boyle, 31, gave up using money in November 2008. He lives in a caravan that he got from Freecycle, an organisation which matches people who have things they want to get rid of with people who can use them, and so keeps usable items out of landfills. His caravan is parked at an organic farm near Bristol, where Boyle volunteers three days a week. He grows his own food, has a wood-burning stove and produces electricity from a solar panel (it cost £360 before the experiment started). He has a mobile phone for incoming calls only and a solar-powered laptop. Boyle, who has been vegan for six years, set up the Freeconomy in 2007, an online network that encourages people to share skills or possessions and now has 17,000 members. His book, The Moneyless Man: A Year of Freeconomic Living, is out now.

 1_______________My friend and I were talking about all the problems in the world, such as sweatshops, global warming, factory farms, animal testing, wars over resources. I realised they were all, in their own way, connected to money.

2_______________I sold my houseboat in Bristol and gave up my job at an organic food company. I made a list of everything I bought and tried to figure out which I could get in another way. For toothpaste I use a mixture of cuttlefish bone and wild fennel seeds. Things like iPods you just have to knock off the list, but birds in the trees around my kitchen have become my new iPod.

Everything takes more time and effort in a moneyless world. Handwashing my clothes in a sink of cold water, using laundry liquid made by boiling up nuts on my rocket stove, can take two hours, instead of half an hour using a washing machine.

It was meant to be just for a year but I enjoy the lifestyle so much that I’m just going to keep living like this. 3________________

I had a very normal childhood. I think at first my parents wondered what on earth I was doing. But now they totally support me and they say that they may even try it themselves.

A I’ve never been happier or fitter.

B Just being a vegan cuts down the number of women I’m compatible with, never mind being moneyless.

C It all started in a pub

D My girlfriend isn’t that keen on my lifestyle though.

E Sometimes it is frustrating trying to socialise with no money

F I decided to give up cash.

The answer to the first gap is ‘C’. This is clear if you understand how narratives work- that we start by giving the background- where he was, what he was doing. The second gap is ‘F’- this relates back to his conversation- it was the result- and to the fact that he realised that the problems were all caused by money (relates to cash). It relates forward to the impact of his decision. The third gap is ‘A’ and relates back to what he says about enjoying his lifestyle. (‘D’ might also be possible here, but it is clear from the rest of the text that he doesn’t have a girlfriend).

Multiple choice and true/false are perhaps the most artificial task types. True/false also has the disadvantage of being pretty easy to guess. However, they are also the most common in exams, which many of our students will need to take, so it is perhaps worth giving them practice in these task types for that reason alone. Some tips for writing these types of questions:


  • Don’t use the exact language in the text- paraphrase.
  • Make sure there is only one idea in each statement, so that students aren’t likely to think it is partially true. For this reason, keep statements short.
  • Make sure there is roughly an equal numbe of T/F answers, and be careful not to create a pattern (e.g. T then F then T)

Multiple Choice

  • Paraphrase as above.
  • Make sure all the choices are plausible and can’t be guessed or answered through general knowledge (for example, a well known piece of ESOL material which asked largely Muslim students about what Ramadan is)
  • Don’t make the distractors (the wrong answers) too complex. You are testing what’s in the stem (the first part of the question).
  • Make all the choices a similar length if possible, and don’t make the correct answer the longest or shortest.
  • Make sure the answer isn’t always ‘C’ etc.

Whatever task type you use, in the spirit of demand high teaching, I think it’s always a good idea to devote some time to exploring why students have chosen particular answers. What in the text led them to that conclusion? This will not only prevent guessing, but will, I think, actually go some way towards developing their ability to tackle texts.


Filed under Exploiting authentic reading materials