Márta Lesznyák: A longitudinal investigation of translation trainees’ translation and post-editing competence development I
›
3 Results and Discussion
›
3.6 Discussion of changes between the beginning and the end of the training period
›
3.6.2 Changes in performance indicators (time on task and error numbers – research questions 3.4 and 3.5)
›
3.6.2.2 Differences in error numbers between the first and the second data collection waves

3.6.2.2 Differences in error numbers between the first and the second data collection waves

Jegyzet elhelyezéséhez, kérjük, lépj be.!

A key question of our longitudinal investigation is how translation and post-editing performance change between the beginning and the end of the training and whether there are differences between the two groups.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

In line with our expectations, most error numbers decreased by the end of the training in both groups. The only exception was terminology errors, which increased in both groups and reached statistical significance in the PE group. Despite the growth in terminology errors, all other error categories showed a decrease in both groups. However, the extent of this reduction varied across error types. In the HT group, the decline in accuracy errors was the most apparent and approached statistical significance. By contrast, in the PE group, fluency errors and the total number of errors dropped significantly.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

These findings have several important implications for training. First, it is obvious that post-editors have profited from traditional translation training even without specific training in post-editing. In fact, their improvement in performance is more pronounced than that of human translators.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Secondly, the difference between PE students and HT students lies in the specific error category in which they showed improvement. Notably, the overall performance of PE students improved significantly – a development that is clearly relevant for professional practice. It is also interesting to observe that in the PE group, the majority of the overall improvement results from a drop in fluency errors. This finding indicates several changes between the beginning and the end of the training. On the one hand, students have more mature views on what is linguistically acceptable in a target text, and, on the other hand, they have the cognitive capacity to focus on this aspect of the text, spot errors and correct them.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

In contrast, HT students’ most pronounced development can be observed in the accuracy category. Given the key importance of accuracy, this is a welcome development. Nevertheless, mean values indicate that at the end of their training, HT students make, on average, one mistake more (8.30) than PE students at the beginning of the training (7.19). This suggests that MT provides such an overwhelming advantage that human translators can hardly compete with it, particularly if the MT output is of good quality.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

Finally, the increase in terminology errors should be explained, which is particularly alarming in the PE group, as the difference between the first and the second wave reached the level of statistical significance; as a result, it cannot be dismissed as a measurement error. All the more so, because Sermann’s analysis of the same student sample, translating and post-editing the legal text, showed that students tended to use online dictionaries and sources of questionable quality, or occasionally no sources at all. Sermann’s findings may explain our results, as low-quality sources will inevitably lead to terminology errors in translated and post-edited texts.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

In addition to students’ inefficient search behaviour and resource use, other tentative explanations can be given for poor performance on terminology. One such explanation could be that terminology is an area where the lack of explicit training in post-editing may lead to negative consequences. It is reasonable to assume that students had never been warned about the weaknesses of MT concerning terminology, and this lack of awareness might have led to inappropriate research and decision-making strategies when checking terms in the MT output.

Jegyzet elhelyezéséhez, kérjük, lépj be.!

To sum up, both post-editing and translation competence show development between the two data collection rounds, but as Figure 8 shows, the students in the two conditions show improvement on different factors, and PE students have a constant advantage over HT students on most factors.
 
Figure 8 Changes in Mean Error Numbers from the Beginning to the End of Training in the HT and PE Groups
chevron_left
chevron_right
Tartalomjegyzék navigate_next
Keresés a kiadványban navigate_next
A kereséshez, kérjük, lépj be!
Könyvjelzőim navigate_next
A könyvjelzők használatához
be kell jelentkezned.
Jegyzeteim navigate_next
Jegyzetek létrehozásához
be kell jelentkezned.
    Kiemeléseim navigate_next
    Mutasd a szövegben:
    Szűrés:
    Kiemelések létrehozásához
    MeRSZ+ előfizetés szükséges.
      Útmutató elindítása
      delete
      Kivonat
      fullscreenclose
      printsave