Blogs About Reading
Shanahan on Literacy
Literacy expert Timothy Shanahan shares best practices for teaching reading and writing. Dr. Shanahan is an internationally recognized professor of urban education and reading researcher who has extensive experience with children in inner-city schools and children with special needs. All posts are reprinted with permission from Shanahan on Literacy.
Teaching Students to Use Context
I’m writing this blog because of the disarray I see over the topic of context instruction and the poor instructional practice that it seems to manifest.
One confusion is already well recognized, but merits some mention here. The other befuddlement usually goes without remark, and yet it, too, has unfortunate consequences for young readers.
Let’s dispatch the first problem forthwith. This one I’ll refer to as the three-cueing problem. Research found that when students err in reading a word, they often try to use various kinds of information to resolve the difficulty. Essentially, when something goes wrong, readers try to make things work one way or another. They don’t try to read the word as much as to get it right anyway possible. They turn to context — trying to guess the word by the meaning of the other words, the pictures, the syntax. Whatever it takes.
That fooled some authorities into encouraging the teaching of those solve-it-at-all-cost strategies, instead of teaching them to read. A bad idea, since poor readers are more likely to turn to semantics, pictures, and syntax to guess words, while better readers rely on the letters and sounds. Why teach kids to read like poor readers?
Context does have a role in the decoding process, but as an evaluation check rather than a word reading tool. Meaning (or the lack of it) reveals the success of decoding. If decoding worked, the reader keeps rolling. In cases of failure, the reader must look at the word again to decide among the decoding alternatives (“maybe this is a schwa sound and not a long vowel?”). Even when the meaning has said, “try again,” the next try depends upon letters and sounds not, context.
Using context to determine meaning
Okay, no context in decoding.
What about in meaning?
Everyone seems to agree that context can be quite helpful for determining the meaning of words and phrases. And yet ...
I spent a lot of time this week reading research on context and meaning. For the most part, I was disappointed.
My take? The research community has been spinning its wheels. Most of their questions have been decidedly academic (in this context, academic means useless for any practical purpose).
Older research usefully revealed that poor readers were not efficient or proficient in deriving meaning from context (McKeown, 1985). Studies also showed that students gain a lot of word meanings just by reading and using context (Nagy, Anderson, & Herman, 1987).
The practical outcomes from such studies have largely been limited to the development of context-taxonomies with little practical value in real reading. Research continues to emphasize various types of context clues (e.g., definitions, antonyms, synonyms, comparison and contrast, examples, lists, cause and effect, inferences).
I use context a lot in my reading, and never try to identify or classify the types of contexts that is available. I can’t imagine that categorizing context clues would improve my reading.
Nevertheless, these schemes have attracted and continue to attract research. Their results — despite all logic — may even persuade some of their pedagogical value.
Fortunately, cooler heads have prevailed on this one, even if those cooler heads are too often ignored. A research synthesis (Kuhn & Stahl (1998) determined persuasively that students benefited from context instruction. However, it found that it wasn’t the category training that helped, only the actual practice in figuring out word meanings from context.
Maybe curriculum designers could employ those frameworks to generate a diverse set of practice items. The students could then engage in practicing making sense of context but would need have no explicit truck with the categories themselves.
I have more basic issue with this research. I think it focuses on the wrong learning outcomes. The researchers emphasize how well the students learn the word meanings, testing their later knowledge of those definitions. That seems wrongheaded to me.
The real purpose of using context is to comprehend the text, not to learn word meanings. Context use also improves efficiency and reduces the burden of having to look up so many words. If we teach context use effectively, then reading comprehension, and perhaps, reading rate, should improve. Students may also end up knowing more words than they would without such teaching, but that would be a secondary outcome, not the primary one.
Context instruction should be aimed at facilitating reading, rather than as a delivery system for explicit vocabulary teaching.
When I confront an unknown word in text, I first determine whether it matters. imagine I come to the following sentence:
“There are many serious diseases of the stomach and duodenum, including gastritis, gastroenteritis, gastroparesis, non-ulcer dyspepsia, peptic ulcers, and gastric cancer.”
I don’t recognize all those diseases. But I’m not in medical school and given my reading purposes, ignorance of gastroparesis won’t diminish my understanding. In this case, it’s enough that I recognize that these are diseases, so I keep reading.
When do we teach that kind of vocabulary conscience — both the importance of recognizing when we don’t know the meaning of a word and the need to make reasonable judgements about how to address our ignorance most appropriately?
I set out to understand a text — not to memorize the author’s words.
The only context studies I could find with a focus on comprehension were older studies that focused on fill in the blanks in cloze exercises. Cloze may be an index of comprehension, but it is one overly aligned with context training. Context use should improve the ability to answer questions about a text or to write a summary of it.
Most context instruction emphasizes whether kids can arrive at the right definition of a word from sentence or paragraph context. For example,
“John was so hungry that he didn’t leave a particle of the muffin on the plate.” Define particle:
Such exercises are common, and they do offer some context practice. I’d challenge such work in terms of its “over consistency.” Such exercises never include any words that can’t successfully be figured out from context. Not exactly how it really works in a reading situation.
“As I looked around the room desperately, the teacher started handing out the papers.” Define desperately:
I think it would be a good idea to mix in items like that one in context exercises. Kids need to recognize when context could help and when it is not likely to.
I suspect that such consistency just teaches students that such training has little to do with reading and that it can safely be ignored. Rarely are the sentences in authentic texts written to, so obviously, reveal the meaning of a word.
Teaching word meanings has a deserved place in the curriculum. Teaching context is something different from that. In the earlier example that I gave, the one with the word particle, I would prefer it if the student were trying to interpret the sentence rather than the word. Crumb might be a good synonym for particle in this case, but so would the word anything. Admittedly, anything is a lousy definition for particle, but such a response would show that the student had been able to interpret the author’s meaning — even when they could not articulate a definition for particle.
While it is apparent that we could improve context instruction by dropping the categories, adding some items in which context fails to help, and by focusing on comprehension rather than vocabulary instruction, there is something even more basic that could and should be done.
Reading lessons usually begin with pre-reading vocabulary introduction. Teachers spend a lot of time familiarizing students with words that will come up in the text they are about to read. This is supportive of comprehension — kids are more likely to understand text when they already know all the words they’ll need. That might be a good idea if teachers plan to continue on with their students into later life, anticipating any words they may need just in time to read each text. I’d rather try to make students more independent than that — you know, teach a man to fish…
There should be less pre-teaching of vocabulary. Let’s end the pre-reading introduction of any words that can reasonably be determined from context. The words should become targets of the post-reading questioning. If students can’t answer such questions, then go back and help them figure those words out. Over time, they should improve in those abilities.
That means context instruction should be a daily experience for kids — not a semi-annual worksheet.
Context is not the best avenue to decoding, but it can play an important role in comprehension — if we teach it as avenue to that, rather than to vocabulary learning. Focus such instruction on sense-making, not word learning.
Comment from Jo
Would you recommend backing away from front-loading vocabulary with even multilingual learners? This is, of course, an element of the SIOP (Sheltered Instruction Observation Protocol) Model for teaching students who are learning English.
Reply from Tim Shanahan
Yes, I would. Students who are learning a language need to learn to deal with context too. Very much what I said above: if there are important words that you think can't be determined from context and/or morphology, and the words are important — feel free to front load. Even better with most kids, would be to provide them with a glossary of those words.
Comment from Richard
What about having students try to determine which words they are struggled with in the reading? I have had some success with this. I am getting many students who have been taught to skip unimportant words but not how to determine if it's important.
Reply from Tim Shanahan
I’m a big fan of that. There isn't much research on that, however. When a student identifies such a word, the teacher then has to decide how best to respond — tell the definition, or guide use of context and morphology.
Comment from Patrick
Don’t leave out connotation.
Reply from Tim Shanahan
What a great point. I sure missed the opportunity on that one. Connotation is one of the great reasons to teach context use. Thanks.
Comment from Joan
Teaching students how to use the context to try and determine the meaning of an unfamiliar word is often referred to as a "word learning strategy" — in my work training teachers I sometimes note that it more aptly might be called a "word figuring-out strategy" although that doesn't sound very smooth. Use of context is also referred to as trying to find clues outside the word. Efficient readers not only look at these "outside" context clues, but they combine this with "inside the word" clues — i.e., use of morphology to recognize meaningful units within a word. I agree completely with your suggestion that the goal of helping students become better readers is best served by giving them guided practice to become adept at using context. Let's also keep in mind what Isabel Beck pointed out years ago that we should teach students that not all contexts are helpful. In her book "Bringing Words to Life" she noted that context can vary from enough information to be able to determine the meaning, to context that provides some information about a new word but not enough to determine its full meeting, to what she called "misdirective" — i.e., misleading clues about a word. She offered this example of a misdirective for the word "grudgingly": "Sandra had won the dance contest, and the audience's cheers brought her to the stage for the encore. 'Every step she takes is so perfect and graceful,' Ginny said grudgingly as watched Sandra dance." (p.4) Part of providing guided practice for students in use of context (and word parts) should be pointing out that it doesn't always work!
Baldwin, R.S., & Schatz, E.K. (1985). Context clues are ineffective with low frequency words in naturally occurring prose. National Reading Conference Yearbook, 34, 132-135.
Baumann, J.F., Edwards, E.C., Boland, E.M., Olejnik, S., & Kame'enui, E.J. (2003). Vocabulary tricks: Effects of instruction in morphology and context on fifth-grade students' ability to derive and infer word meanings. American Educational Research Journal, 40(2), 447-494. doi-org:10.3102/00028312040002447
Baumann, J.F., Font, G., Edwards, E.C., & Boland, E. (2005). Strategies for teaching middle-grade students to use word-part and context clues to expand reading vocabulary. In E. H. Hiebert, & M. L. Kamil (Eds.), Teaching and learning vocabulary: Bringing research to practice (pp. 179-205). Mahwah, NJ: Lawrence Erlbaum.
Blachowicz, C.L.Z., Fisher, P.J.L., Ogle, D., & Watts-Taffe, S. (2006). Vocabulary: Questions from the classroom. Reading Research Quarterly, 41(4), 524-539. doi-org:10.1598/ RRQ.41.4.5
Dessenberger, S., Wang, K., Jordan, E., & Sommers, M. (2022). Lexical inferencing as a generation effect for foreign language vocabulary learning. Memory & Cognition. doi.org:10.3758/s13421-022-01348-5
Duffelmeyer, F. A. (1984). The effect of context on ascertaining word meaning. Reading Research and Instruction, 24(1), 103-107.
Fukkink, R.G., & de Glopper, K. (1998). Effects of instruction in deriving word meaning from context: A meta-analysis. Review of Educational Research, 68(4), 450-469. doi.org/10.1111/0023-8333.00162
Fukkink, R.G. (2005). Deriving word meaning from written context: A process analysis. Learning and Instruction, 15, 23-43.
Hafner, L.E. (1965). A one-month experiment in teaching context aids in fifth grade. Journal of Educational Research, 58(10), 472-474.
Kuhn, M.R., & Stahl, S. A. (1998. Teaching children to learn word meanings from context: A synthesis and some questions. Journal of Literacy Research, 30(1), 119-138.
Nagy, W.E., Anderson, R.C., & Herman, P.A. (1987). Learning word meanings from context during normal reading. American Educational Research Journal, 24(2), 237-270.
Schatz, E. K., & Baldwin, R. S. (1986). Context clues are unreliable predictors of word meanings. Reading Research Quarterly, 21(4), 439-453.