Context: Good News and Bad

By a slight margin, most of the 87 voters thought context explains word meaning about 25% of the time, for 1 in 4 words. That's about right, if we count less explicit clues, leading to a more general understanding of the word (Swanborn & deGlopper, 2002).Context varies, but in general, it does not provide an 'in-house' explicit definition. Case in point: When Tiger Woods apologized for his 'transgressions' all of Googledom scrambled to comprehend the word. The top search term for 12-2-09 included 'transgression meaning' and 'transgression definition.' Gobsmacking, isn't it? Even knowing the context folks had to look up transgression (and they did look it up, because they were sufficiently interested--proving the power of motivation). 

As an aside, one might approach transgressions through morphology: The Latin root -gress- means 'to step.' Thus, transgress is to step across (to cross the line). Regress is to step backwards, while the opposite is progress, literally to step forward. Then there is Congress, a body that might yet learn how to step together (but I digress). Another approach is through classroom conversation, using transgress as a substitute for 'break the rules' when explaining the class norms or discussing an offense. This would situate an archaic but important word in classroom context.

"Context works well if you have a lot of it" (Kevin Feldman, email). The more we know about any subject, the more apt we are to utilize context to leverage our understanding. But in general, how reliably does context explicate word meaning?  Sometimes, context is direct and explicit but more often it is too general or even misleading (Beck, McKeown, & Kucan, 2002). According to a landmark study by Nagy, Herman, and Anderson (1985), average and above-average readers in eighth grade could infer unknown word meaning from context clues only 11% of the time, on average (but as mentioned earlier, Swanborn and deGlopper's estimate, allowing for partial knowledge, arrived at about 25% of the time).

Here is more promising news: Fukkink and de Glopper (1998) examined 21 research studies and determined that instruction in context clues yields a medium-sized effect on ability to derive word meanings. Thus, we can be taught to infer word meanings from context, but only when the context itself is sufficiently explicit and our knowledge base sufficiently robust. Want to increase the probability? Teach students to infer meaning from context clues and morpheme clues, integrating the two types of information (more on this later). Build background knowledge and schema, too (Nagy, Anderson, & Herman, 1987).

Provide multiple exposures to the novel word in varied context. As David Pearson said, "Read it! Write it! Talk it! Do it!" Persist. There are several effective and enlightening ways to teach vocabulary, and only one truly feckless approach--assigning a list of decontextualized words to be learned via the dictionary. That's been proven ineffective in several studies (e.g., Bos & Anders, 1990; Bos et al., 1989; McKeown et al., 1985). 

With that, I make my egress.