Eye-Tracking Processes and Styles in Sight Translation by Wenchao Su

Author:Wenchao Su
Language: eng
Format: epub
ISBN: 9789811556753
Publisher: Springer Singapore

4.3.8 Data Analysis for Local Style 3

Finally, I examined the third dimension of local translation styles , i.e., how translation problems affect the reading-speech coordination in actual sight translation. The coordination behaviors were measured by the Max. and Min. EVS at the word level, which were obtained based on the data analysis of global styles as described in Sect. 4.3.5.

Here, the translation problem was the source text word that met the following two criteria. First, it had to have been among the top 15 words with the longest TFD in actual sight translation for the same participant. Second, it had to have been among the top 15 words with the longest TFD in actual sight translation for at least four of the total 13 novices, or for at least three of the total nine professionals. Thus, the translation problem was the source text word that received relatively high TFDs in actual sight translation. Furthermore, it was termed an actual problem because it occurred in the stage of actual translation (see Angelone 2018). Moreover, since these translation problems attracted relatively high TFDs for most participants within each group, they were deemed to be common actual problems.

Before conducting the inferential analysis of the problem effect on the EVS , the distribution of Max. and Min. EVS was plotted using the sm.density.compare function from the sm package (Bowman and Azzalini 2018).

Following on from this, an LMER analysis was performed to investigate the effect of actual translation problems on the coordination behaviors of each group in actual sight translation. The dependent variables were Max. and Min. EVS at word level, and the test predictor was AOI with two levels (common actual problem [CAP] versus common actual non-problem [CANP]). The task time each participant spent during the preparatory reading was entered as the control predictor in the LMER model. To examine whether the effect of actual problems differed between novices and professionals, I selected the common actual problems for both groups (CAPFBG) and common actual non-problems for both groups (CANPFBG) for further analysis. As a result, the LMER model consisted of group, AOI (CAPFBG vs. CANPFBG) and their interaction as the test predictors, and the task time spent in preparation as the control predictor.

Apart from conducting the inferential analysis of the problem effect on the EVS , it was important to probe into “‘what goes on in between the first and the last fixation on a data point before the onset of output production” as suggested by Timarová et al. (2011, p. 135). Thus, bubble graphs from Excel were used to illustrate what happened during the EVS of a specific translation problem.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.