We Don’t Need No Stinkin’ Exercises*: Listening to Stories is More Efficient Than Direct Instruction for Vocabulary Acquisition

Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInEmail this to someone

Study reviewed:

Loftus-Rattan, S. M., Mitchell, A. M., & Coyne, M. D. (2016). Direct vocabulary instruction in preschool: A comparison of extended instruction, embedded instruction, and incidental exposure. The Elementary School Journal, 116(3), 391-410. (pay wall)

Everyone agrees that reading storybooks to young children helps them build vocabulary. But lots of researchers think they can improve on storybook reading, making it more “effective” as a way of teaching children new words.

Loftus-Rattan and colleagues compared three storybook reading conditions to see which was more effective in teaching a group of preschool children (N = 25) a set of three unknown words. Those conditions were:

  1. Incidental Exposure: Teachers read the storybooks as they usually did without any vocabulary explanations or “follow-up” activities related to the target words.
  2. Embedded Instruction: Teachers asked students to listen for the three words and raise their hands when they came up in the story. The teacher then gave a definition of the word (e.g. “A weald is forest or some woods” (p. 398)), re-read the line in which it appeared, and had the kids repeat the word.
  3. Extended Instruction: Teachers did everything in the Embedded Instruction condition, plus follow-up activities with additional direct instruction on the three target words.

The children were randomly assigned to one of the three storybook conditions, and heard the story (Goldilocks) three times over a period of one week. They were then tested on their word knowledge, including a meaning recall test that required them to give a definition of the word. The words were all rare and unknown to the children before the experiment. Post-tests were given one week and six weeks after the treatment.

The researchers gave time estimates for the instructional conditions. For the Embedded Instruction, teachers spent about two minutes on each word. The story was read to the kids three times, so we get a total of 18 minutes for the three target words (two minutes X three story readings X three words).

For the Extended Instruction condition, teachers spent seven minutes on each word, which gives us a total of 63 minutes for the three target words (seven minutes X three story readings X three words).

No time estimate was provided by the researchers for the Incidental Exposure condition. Experimental psychologists typically count only the sentence an unknown word appears in as the “context” for that word, so by that standard the estimate should be around 5 seconds per word. But as Krashen (1999) pointed out, information found before (or even after) the sentence containing the unknown word may help the reader understand its meaning.

In an previous study with a nearly identical design (Coyne et al., 2009), the researchers used 10 seconds per word for the incidental condition (for a total of 30 seconds).** I have used that same figure here. That gives us a total of 1.5 minutes per set of three words (10 seconds X three story readings X three words).

Table 1 shows how many of the three words the average child knew on the post-test and the delayed post-test for each condition (from Table 2, p. 402). I’ve adjusted the raw scores from the recall test in order to show the estimated number of words learned. Also shown in Table 1 is the total amount of time per condition for those three words, and the words per minute calculation (words learned/time spent in minutes), which tells us how efficient each condition was.

Table 1: Word Knowledge Gains in Three Storybook Reading Conditions

 Extended InstructionEmbedded InstructionIncidental Exposure
Recall Scores (Post-Test).94.30.10
Recall Scores (Delayed Post-Test).76.28.08
Time on Treatment63 minutes18 minutes1.5 minutes
Words per minute (Post-Test) .014.016.067
Words per minute (Delayed Post-Test).012.015.053

Recall maximum score = 3.

The researchers looked at the number of words gained in each condition and concluded that Extended Instruction was the most “effective” method for teaching kids new words. .

But it’s clear from the words per minute calculations shown in Table 1 that direct instruction is the most costly approach in terms of time. Simply listening to the teacher read the story is more than four times as efficient as direct instruction in acquiring new vocabulary (.067/.014 = 4.7).

If you have all the time in the world, then direct instruction is for you.

Direct instruction is more work for the teacher, more painful for the students, and less efficient for vocabulary growth. So thanks, researchers, but teachers and students don’t need any more of these kinds of “improvements.”

*With apologies to The Treasure of the Sierra Madre.

**I have corrected an earlier version of this post in which I had Coyne et al. giving the equivalent of 90 seconds per word over the course of the three readings for the incidental condition (.5 x 3 story readings).

P.S. I am not saying teachers should withhold explanations of words from children during storybook reading, especially when the kids ask for them. That would be really annoying for the children, even disrespectful. I know of no study that has looked at the effects of explaining words that the kids themselves ask about, however, which says something about the mindset of researchers (to be fair, there may be one out there that I missed).

A case can be made that short, teacher-selected “on-the-fly” explanations of words may sometimes be useful, but that, too, would need to be examined with an eye toward its efficiency. Biemiller and Brote (2006) reported that during their storybook reading experiment with kindergartners and 1st graders, “1 or more children expressed complaints about interruptions for explaining word meanings” when the story was read to them the first time (p. 48). Perhaps we should listen to the customer.

Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInEmail this to someone