be of high and low frequency based on Thorndike’s index (1944). That index was chosen
since it gives the frequency of the words present at different age groups by taking into
account their oral vocabulary.
Prior lexical knowledge (naming and comprehension) tests
Those tests were given as measures of baseline vocabulary knowledge (expressive and
receptive). They were also used as pre-test for children’s knowledge of the target words by
measuring the number of accurate and incorrect responses. The screening test was very
similar to the one used by Dockrell, Messer, & George, (1997). The order of the words’
presentation was randomised. Practice trials were not scored.
Each of the two vocabulary tests had 22 trials and four additional practice trials (see
Appendix 7.1). The naming vocabulary test included the same items as the comprehension
vocabulary test. The test stimuli belonged to two different semantic categories, animals and
artifacts. Eleven of the items were words for animals and eleven of the items were words for
artifacts. Ten of the words were of high frequency and twelve of the words were of low
frequency. Four out of the twenty two words were the target words used in the study. A
laptop computer was used, where the pictures of the items were presented on the screen. A
scoring sheet, a paper and a rubber were also used.
Post-test measurements
Naming task: Four different cards presenting the pictures of the target items were used.
Multiple choice task: Four plates were used. On each plate four pictures were presented (see
Appendix 7.2). They included a picture of the target word, a semantic foil, a phonological
foil and an irrelevant foil.
Association task: Twenty four cards presenting pictures of different items as well as pictures
of the target items were used.
Contrast task: Pictures of the four target words as well other pictures (from the association
task) were used.
Story generation task: The four pictures of the target words were used.
7.3.4 Procedure