Available at

noun phrases are the most challenging linguistic items for interpreter trainees 
working from English into French. As can be seen in Table 23, complex noun 
phrases have the lowest percentage of close renditions and the highest 
percentage of divergent renditions among the four investigated linguistic items. It 
is possible to go even further and underline the fact that there is a particular NP 
sub-pattern, which seems very difficult 
® “N + subclause”, i.e. any noun phrase 
followed by a subclause. Close renditions for this particular sub-pattern account for 
less than 50% of all renditions, which means that less than half of noun phrases 
followed by a subclause were correctly translated. 

---

Results and discussion 
 
page 86 
4.2.3 Longitudinal analysis 
In order to answer the second research question, I will focus on the students who 
carried out at least three interpreting tasks, namely STU01, STU02, STU04, 
STU06 and STU07. I will only take into consideration close and divergent 
renditions as they are the only two types of rendition that can help us determine if 
there is some kind of improvement across interpreting tasks (increase in close 
renditions and decrease in divergent renditions). In other words, the other types of 
rendition, i.e. reduced, zero, expanded, summarized, substituted, and innovative, 
are left aside in this section. As a reminder, IN01 was used on November 3
rd
, IN02 
on November 10
th
, IN03 on November 17
th
and IN04 on December 8
th
, 2016. 
Table 24 – Overview of the percentages of close and divergent renditions 
 
IN01 
IN02 
IN03 
IN04 
Total % 
Close 
rendition 
50.68% 
(50/296) 
58.42% 
(215/368) 
52.13% 
(171/328) 
52.29% 
(217/415) 
53.38% 
Divergent 
rendition 
13.51% 
(40/296) 
16.85% 
(62/368) 
23.78% 
(78/328) 
20.24% 
(84/415) 
18.59% 
As can be seen in Table 24,
there is no clear trend and no marked improvement 
across time here. Although there is an increase in the percentage of close 
renditions from IN01 to IN02, the contrary cannot be said about the percentage of 
divergent renditions. In addition, close renditions drop to 52.12% in IN03 and 
poorly increase in IN04. In general, we can then say that the quality of the 
trainees’ interpretations does not improve across time. However, we can take a 
closer look at the global features we investigated and see how delivery rates, 
lexical density scores and expansion rates can be linked to the trainees’ proportion 
of close and divergent renditions. 

---

Results and discussion 
 
page 87 
Table 25 – Percentage of close and divergent renditions (of the five students 
under investigation)