stories from memory. In their pretend reading of the stories, the children were observed
to use the language of the stories including words considered unusual in kindergartner’s
discourse. This was taken as evidence by Eller et al. (1988) that the children had acquired
new vocabulary from listening to the stories.
However, use of the novel words in only one situation does not prove that had acquired
the full meaning of the words. Otherwise, they should have employed different tasks
which would show whether they had grasped all the different aspects of the words’
meanings. Furthermore, it is unclear whether this model of development they propose
would be valid for the acquisition of new lexical items since it is not clear, if we consider
the criteria they used, that the words were unknown to the children. The data for this
investigation were extracted from a larger study, and consequently no pretesting of the
target vocabulary words was undertaken, so the children’s previous knowledge of these
words was not known. Furthermore, no post-testing was undertaken to see whether this
type of performance was still present a week later. Besides, because “bona fide”
children’s literature was used, the number of encounters with the target words could not
be controlled.
Leung and Pikulski (1990) studied the extent to which kindergartners and first graders
could identify the meanings of novel words that they heard in stories. They replicated the
Eller et al’s (1988) study by using the same two picture storybooks, but they also used a
pretest / post test design that included controls who did not hear the stories. After hearing
each story, participants from the experimental group were asked to give the meanings of
20 target words from the stories before and after the experimental treatment.
Results support Eller et al’s (1988) findings that repeated exposure to stories increased
children’s use of target words in their pretend readings. However, there was no significant
difference between the experimental and the control groups in vocabulary gain as
evidenced by subjects’ ability to verbally define the target words. Leung and Pikulski
suggested that vocabulary gains might have been demonstrated if the design had used a
multiple choice test of word meanings. In other words, a different measurement of word
knowledge (a more moderate one such as a multiple choice test) could alter the results.
73