In the 70s a very popular theory about reading was that skilled readers were using the context of what they were reading and were thus predicting what words were coming. This is what was thought to allow skilled readers to read so rapidly.
From this, the conclusion was draw that if skilled readers read in this way than we should teach children to read in this way. When a child did not know a word or stumbled on a word while reading, you would cue the child  by asking “What do you think it says?” Or you could tell the child to look at a picture, other cues to use context or their background knowledge. If you did this you would be building a good reader.
Reading researchers thought this sounded like a good theory. So they began testing this theory in the 70s and 80s. They were surprised to find that “guessing” and “using context” was not how good readers recognized word. They disproved the context theory. After 40 years of research this old theory is no longer held by reading scientists. They have determined that skilled readers do no rely on guessing a word and in fact guessing a word is more what weak readers do.  AND it is not the best way to help a child to read. There are better ways to teach a child how to read.

The problem is….practitioners still use the strategies based on the old theory.
Beginning readers need to learn how to recognize words, in order to read automatically – rapidly and effortlessly.  This skill is called word recognition, simply knowing  a word when you see it.
How can we best teach children to automatically recognize words so they can read smoothly and effortlessly?

Let’s take a look at what the research tells us and if the old cues of “What do you think it says?” or directing the child to look at the pictures is an approach we should abandon. And then discuss new approaches to help children build reading skill of recognizing words accurately and rapidly.