Chapter 1 the study of collocations

CHAPTER 4 
 
ANALYSES AND RESULTS 
 
4.0  Introduction 
 
This chapter describes the results from the language proficiency 
measures and the main analyses performed to address the two hypotheses 
listed in 2.5.  The presentation of the results is organised around the three sets 
of data used to address each of the hypotheses: the free production essay data, 
the elicited production translation data and the elicited production blank filling 
data.  In section 4.1 the results from the language proficiency measures 
performed on the essay data are reported.  The aim of these measures was to 
screen the data and establish clear proficiency differences among the three 
groups, using different measures of language proficiency on the essay data.  
This initial screening of the data was considered necessary, since the 
proficiency differences between groups are the major independent variable for 
the present study. 
 
In 4.2 the results of the main analyses of each set of data are described.  
The analyses and results for Hypothesis 1 are first reported in 4.2.1.  The aim of 
these analyses was to examine evidence for developmental differences in the 
knowledge of collocations, assessed both in terms of ability to use collocations 
in the essay data, and in terms of accuracy of response to questions eliciting 
collocations in the translation and blank filling data.  These analyses involved 
 
233

---

comparisons of collocation use and accuracy between groups using Kruskal-
Wallis tests.  The Kruskal-Wallis one-way analysis of variance by ranks is a 
non-parametric test for deciding whether a number of independent groups are 
from different populations (Siegel & Castellan 1988:206).  Evidence for 
acquisition orders was then sought using implicational scaling of:  i) the use of 
collocations in essays across all groups, and ii) mean accuracy of response to 
collocation types on the translation and blank filling tests across all groups.  In 
this way it was hoped to show what differences  existed in the subjects' 
knowledge of collocations across different proficiency levels, and how 
knowledge of collocations developed.  This evidence was used to address the 
first hypothesis, which states that there are patterns in the development of 
collocational knowledge across all groups.  
 
Analyses and results for Hypotheses 2 are described in section 4.2.2  
These analyses involved comparisons of collocation use and accuracy within 
each group using Friedman Repeated Measures tests, which is a parallel non-
parametric test for repeated-measures ANOVA.  Evidence for acquisition 
orders within groups was then examined using implicational scaling of 
collocation use and accuracy for each group.  The aim of these analyses was to 
show what developmental differences and sequences existed in the use of 
collocations within groups.  These results would reveal any group-specific 
patterns in the development of collocational knowledge. 
 
The results of all the analyses are summarised in 4.3 in relation to the 
two hypotheses of the study.