Íslenskt mál og almenn málfræði - 01.01.1983, Page 109

Learning about -ari 107
1. Method
1.0
The experimental method used in this study was modelled on that
used in Clark and Hecht (1982). I used the same procedure, materials
and subject age groups with the exception of minor, language-specific
adjustments as necessary. For example, one of the verbs used in the
English version of the study, bounce, does not have a familiar single-
word counterpart in Icelandic, so it was replaced by another common
Icelandic verb (að vökva ‘to water’) and pictures appropriate for this
verb were substituted in the experimental materials.
1.1 Subjects
The 48 children participating in the study were all from middle-
or upper-class homes in the greater Reykjavík area. They were di-
vided into four age groups with twelve childen (six boys and six girls)
in each group: I 3;0-3;8 (mean age 3;4), II 3;9-4;5 (mean age 4;1),
III 4;6-5;2 (mean age 4;10), IV 5;3-6;0 (mean age 5;7). All children
were native speakers of Icelandic.
1.2 Materials
Using the English verb list given in Clark and Hecht (1982) as a
basis, I drew up a list of 40 Icelandic verbs meeting the criteria that:
(2)1 they are likely to be familiar to two- and three-year old chil-
dren;
2 they are (with the exception of þvo) two syllables (i.e. stem
+ a infinitive marker);
3 they are not derived from nouns; and
4 nouns formed by adding -ari to these verb stems are not
likely to be familiar to young children.
These verbs are listed in Table 1. For each verb, two pictures were
drawn, one of a human agent performing the action named and one
of an instrument performing the same action. For the most part, these
pictures were identical to those used in the Clark and Hecht study.
(See the sample pictures in Figure I.) Each picture appeared on a
separate 3” x 5” card.