Chapter 1 the study of collocations

Table 29. Translation accuracy distinguishing proficiency levels 
Group 1 
Group 2 
Group 3 
Collocation Types 
Collocation Types 
Collocation Types 
 11. 
SV(O) Prep O** 1. 
Noun Prep** 
 5. 
Adjective Prep* 14. 
SVV-ing** 
   27. 
Verb Noun (creat)** 
   5. 
Adjective Prep* 
**: Significantly more accurate than the other two groups 
*: Significantly more accurate than one other group 
 
4.2.1.2.3  Implicational Scaling for the Translation Data (All Groups) 
 
For the implicational scaling analysis, following the Guttman procedure, 
 each subject was coded as having translated accurately (1), or not (0), each of 
the 6 types of collocations in the translation test.  A criterion of 80% accuracy 
was used for the coding of the data (see also Andersen 1978; Anderson 1978).  
That is, if a subject was 80 to 100% accurate in translating the particular 
collocation type, she/he was coded as 1.  Accuracy less than 80% was coded as 
0.  As with the essay data, the two axes of the matrix for the implicational 
scaling consisted of the six items ranked from the most accurately translated by 
 
268

---

all subjects to the least accurately translated, and the 275 subjects ranked in 
order of their accuracy of response to all types of collocations, from  subjects 
translating accurately the most types to subjects translating accurately the 
fewest types.  This matrix is given in Appendix M and summarised in Figure 6 
below.  The coefficient of reproducibility for this analysis was .92.  The 
coefficient of scalability was .578 and so approached significance (see Andersen 
1978; Hatch & Lazaraton 1991:212). 
 
0
.0 5
.1
.1 5
.2
.2 5
.3
.3 5
Ty pe1 4
Ty pe1 6
Ty pe1 1
Ty pe1
Ty pe5
Ty pe2 7
Col l oca t i on  Ty p es
 
      Accura cy   - Tra nsl a t i on Test  - Al l  Group s