Chapter 4 Findings
4.1 CI Results
4.1.1 Note-taking preferences
The first research question is: Does FSR affect student interpreters’ note-taking preferences in CI? Five dependent variables were chosen to represent note-taking preferences in this thesis: the distribution of language-based notes to symbols, the distribution of full words to incomplete words, the distribution of notes in Chinese to notes in English, the number of horizontal lines, and the number of total note units. The aim was to see whether there was a significant correlation between the SR change and each of the five variables.
First, the results of between-subject design showed that there were no statistically significant differences between Y1 and Y2 participants in each analytical result (see Appendix vii.). This means that Y1 and Y2 groups shared similarities among all the note-taking preferences. Therefore, it was unnecessary to perform analysis on the two groups separately. So, within-subject design was then applied to all the participants, the results of which are as follows.
4.1.1.1 The distribution of language-based notes versus symbols
According to the result shown in Table 4.1, there was a significant effect of SR difference on the ratio of language to symbol (P=.01).
Table 4.1 ANOVA on the effect of SR on language/symbol ratio
The results from descriptive statistics as shown in Table 4.2 showed the average in SSR (2.22) and FSR setting (2.92). The ratio of language-based notes to symbols in the FSR setting was higher; that is, on average the participants wrote more language-based notes over symbols in the FSR setting.
Table 4.2 Descriptive statistics of language/symbol Ratio
Language/Symbol grade M SD N 4.1.1.2 The distribution of full words versus incomplete words
According to the result shown in Table 4.3, there was an insignificant effect of SR difference on the ratio of full words to incomplete words (P=.662). This indicates that the participants wrote neither more full words nor more incomplete words in the FSR setting.
Table 4.3 ANOVA on the effect of SR on full word/incomplete word ratio
SS df MS F p
Within-Subjects Effects
SR and F/N ratio 0.67 1 0.67 0.20 .662
Error(SR) 64.14 19 3.38
* P value < 0.05
4.1.1.3 The distribution of English and Chinese
According to the result shown in Table 4.4, there was an insignificant effect of SR difference on the ratio of English words to Chinese words (P=.494). This indicates that the participants did not write more notes in either English or Chinese in the FSR setting.
Table 4.4 ANOVA on the effect of SR on English/Chinese ratio
SS df MS F p
Within-Subjects Effects
SR 2.28 1 2.28 0.49 .494
Error(SR) 84.32 18 4.68
* P value < 0.05
4.1.1.4 The amount of horizontal lines
In this study, the two source texts were designed to differ in information density to achieve the SR difference. By default, there were more information that could be written down in the FSR setting. In order to know in which setting the participants were inclined to draw more lines for the same amount of information, the amount of
horizontal lines per proposition in the source texts was chosen to be investigated.
According to the result shown in Table 4.5, there was a significant effect of SR difference on the amount of horizontal lines per proposition (P=.002).
Table 4.5 ANOVA on the effect of SR on horizontal lines per proposition
The results from descriptive statistics as shown in Table 4.6 showed the average in SSR (0.10) and FSR setting (0.08). This shows that on average the participants drew fewer horizontal lines per proposition in the FSR setting.
Table 4.6 Descriptive statistics of horizontal lines per proposition
Grade M SD N 4.1.1.5 The amount of total note units per proposition
In order to know in which setting the participants were inclined to write more notes for the same amount of information, the amount of total note units per proposition in the source texts was chosen to be investigated.
According to the result shown in Table 4.7, there was a significant effect of SR difference on the total note units per proposition (P=.000).
Table 4.7 ANOVA on the effect of SR on total note units per proposition
The results from descriptive statistics as shown in Table 4.8 showed the average in SSR (0.95) and FSR setting (0.76). The amount of total note units per proposition in the FSR setting was larger; that is, on average the participants took fewer notes per
proposition in the FSR setting.
Table 4.8 Descriptive statistics of total note units per proposition
Grade M SD N
The third research question is: Does FSR affect student interpreters’ accuracy in CI?
The scores of every participant in both settings was fed into SPSS for analysis.
First, as shown in Table 4.9, the result of between-subject design showed that there was no statistically significant difference (P=.833) between Y1 and Y2 participants in the analytical result. This means that Y1 and Y2 participants’ accuracy was affected by FSR to a similar extent.
Table 4.9 ANOVA on the effect of grade and SR on accuracy
SS df MS F p
Within-Subjects Effects
SR 5118.91 1 5118.91 108.72 .000
SR * Grade 2.16 1 2.16 0.05 .833
Error(SR) 847.51 18 47.08
Between-Subjects Effects
Grade 658.53 1 658.53 7.02 .016
Error 1689.66 18 93.87
* P value < 0.05
Since the result in Table 4.9 revealed no difference between the two groups of participants who have spent different years at the program, it was unnecessary to perform analysis on the two groups separately. A within-subjects design was applied to all the participants. According to the result shown in Table 4.10, there was a significant effect of SR difference on the score of accuracy (P=.000).
Table 4.10 ANOVA on the effect of SR on accuracy
SS df MS F p
Within-Subjects Effects
SR 5118.91 1 5118.91 114.47 .000*
Error(SR) 849.67 19 44.72
* P value < 0.05
The results from descriptive statistics as shown in Table 4.11 showed the average in SSR (80.70) and FSR setting (58.08). The score of accuracy in the FSR setting was lower.
Table 4.11 Descriptive statistics of accuracy 4.1.3 Note-taking preferences and accuracy in the FSR setting
The forth research question is: Do certain note-taking preferences correlate with better CI accuracy in the setting of FSR? To understand whether better performers have certain note-taking preferences in the FSR setting, all the participants were divided into two groups based on the scores given by the two raters. Participants who scored above the mean of scores were in the high scoring group, and those who scored below the mean were in the low scoring group. Next, the note preferences of each group were analyzed.
The five dependent variables to represent note-taking preferences were the same as in 4.1.1: the distribution of language-based notes to symbols, the distribution of full words to incomplete words, the distribution of notes in Chinese to notes in English, the number of horizontal lines, and the number of total note units. Besides, language-based notes, symbols, full words, incomplete words, Chinese words, and English words were tested on independently, too, to achieve a more detailed understanding of the relation between note-taking patterns that may have led to better accuracy.
4.1.3.1 The distribution of language versus symbol
The numbers of note units in the form of language and symbols in the FSR setting were fed into SPSS for analysis. As shown in Table 4.12, the result of the
within-subjects design for high scoring versus low scoring participants found no
significant interaction effect between the language/symbol gap and score (P=.332). This indicates that there was no significant difference between participants from the high scoring and low scoring group in the gap between the numbers of note units in the form of language and symbols.
Table 4.12 ANOVA on the effect of FSR score-group on language-symbol gap
SS df MS F p
Within-Subjects Effects
Language-symbol gap 6553.60 1 6553.60 63.73 .000 Language-symbol gap * Group 102.40 1 102.40 1.00 .332
Error(SR) 1851.00 18 102.83
Between-Subjects Effects
Group 52.90 1 52.90 0.66 .427
Error 1440.00 18 80.00
* P value < 0.05
4.1.3.2 The distribution of full word versus incomplete word
The numbers of note units in the form of full word and incomplete word in the FSR setting were fed into SPSS for analysis. As shown in Table 4.13, the result of the within-subjects design for high scoring versus low scoring participants found no significant interaction effect between the full word-incomplete word gap and score (P=.519). This indicates that there was no significant difference between participants
from the high scoring and low scoring group in the gap between the numbers of note units in the form of full words and incomplete words. Namely, the full word-incomplete word gap between the two groups was similar.
Table 4.13 ANOVA on the effect of FSR scoring group on full-incomplete word gap
SS df MS F p
Within-Subjects Effects
Full word-incomplete word gap 476.10 1 476.10 2.45 .135 Full word-incomplete word gap *
Group 84.10 1 84.10 0.43 .519
Error(SR) 3500.80 18 194.49
Between-Subjects Effects
Group 102.40 1 102.40 1.95 .180
Error 946.20 18 52.57
* P value < 0.05
4.1.3.3 The distribution of Chinese versus English
The numbers of note units in the form of Chinese and English in the FSR setting were fed into SPSS for analysis. As shown in Table 4.14, the result of the
within-subjects design for high-score versus low-score participants found no significant interaction effect between the Chinese-English gap and score (P=.405). This indicates that there was no significant difference between participants from the high scoring and low scoring group in the gap between the numbers of note units in the form of Chinese and English.
Table 4.14 ANOVA on the effect of FSR score-group on English-Chinese gap
SS df MS F p
Within-Subjects Effects
Chinese-English gap 4473.23 1 4473.23 25.86 .000 Chinese-English gap * Group 126.03 1 126.03 0.73 .405
Error(SR) 3113.25 18 172.96
Between-Subjects Effects
Group 75.63 1 75.63 1.88 .188
Error 725.25 18 40.29
* P value < 0.05
4.1.3.4 The amount of horizontal lines, total note units, and all the other categories
The numbers of note units in each category in the FSR setting were fed into SPSS for analysis. In this case, independent T-tests were performed. As shown in Table 4.15, there was only a significant effect in the horizontal line category (P=.031). According to the descriptive statistics also shown in Table 4.15, the mean of the high scoring group (11.40) was higher than that of the low scoring group (6.40), meaning that those who scored higher in the FSR setting used more horizontal lines than those who scored lower.
Table 4.15 T-tests on the effect of FSR scoring group on all categories
Low High
M SD M SD t p
Language 47.50 9.710 42.00 8.179 1.370 .188
Symbol 18.70 12.910 19.60 6.150 -.199 .844
Full word 22.20 10.758 16.10 8.144 1.430 .170 Incomplete word 26.20 13.983 25.90 10.796 .054 .958
Chinese 11.40 9.548 12.20 7.510 -.208 .837
English 36.10 14.410 29.80 8.443 1.193 .248
Horizontal line 6.40 6.077 11.40 2.989 -2.335 .031*
Total 83.60 20.495 87.50 9.536 -.546 .592
* P value < 0.05
4.2 Interview Results
Retrospective interview was the second phase of data collection in this study. It was conducted after the experiment with the purpose of soliciting feedback from the participants in regards to the experiment. Each participant was asked five questions regarding the research questions:
(1) Is faster speech rate a difficulty for you? On a scale of 1 to 5, 1 being not difficult at all and 5 being very difficult, how would you rate the difficulty brought about by faster speech rate in the experiment?
(2) What challenges did the faster speech bring in the phase of note-taking?
(3) Did you notice any differences in your note-taking preferences between the two settings?
(4) Did you consciously resort to different strategies when you were taking notes in the two settings? What were the differences?
(5) Did the faster speech affect your final output performance? If yes, please give the reasons.
After the data were analyzed, the qualitative analysis showed that there were similarities and differences concerning FSR note-taking in Y1 and Y2 participants. In addition, there were also similarities and differences in the way Y1 and Y2 participants identified FSR challenges. The following are the findings arranged in the order of the five interview questions.
4.2.1 FSR difficulty perception
The difficulty level of FSR was rated by the participants on a scale of 1 to 5.
During the interview, the researcher reminded the participants that they should not consider factors other than SR, such as speech text difficulty, when selecting the difficulty level. The average of Y1 participants was 3.6; the average of Y2 participants was 3.25. With FSR considered only slightly less difficult by Y2 participants, this shows that there was not a wide gap in difficulty perception of FSR between Y1 and Y2
students in this experiment.
4.2.2 FSR challenges to note-taking
The following are some common challenges mentioned by participants. A lot of the challenges commented on by participants were interrelated, but could not be grouped in
only one category. This is reasonable, considering the fact that the Efforts in CI often happen in close temporal proximity to each other (see Chapter 2). In view of the
multitasking nature of interpreting, it is only natural that one problem may be directly or indirectly tied to another.
4.2.2.1 Inefficient notes due to inadequate comprehension
Six participants (30% of all participants) revealed that a major challenge to note-taking efficiency came from comprehension issues, in addition to or instead of note-taking techniques themselves. This was reflected in comments from both Y1 (A1, A2, A7, A10) and Y2 (B1, B4) participants. Participants stated that FSR made it
extremely hard to mentally grasp every detail of the speech (A7), or to take notice of the tone of the speaker (B4, B13).
For some participants, even if they managed to write something down, sometimes the note units could not remind them of the whole picture in the source text, because they did not even understand it in the first place. The lack of adequate comprehension reported by the participants, as reviewed in Chapter 2, was possibly a result of FSR disrupting the Listening and Analysis Effort, resulting in compromised understanding of the source text (Gile, 2009). The difficulty of understanding the logic of the passage was brought up multiple times in the interviews. Participants reported only noting down proper nouns or numbers while they were not able to follow the logic of the speech. As A10 has said, “I didn’t quite understand some of the messages the speaker had said, so I could only write down keywords.” This was viewed as problematic by the participant, saying that “When I didn’t really understand something when I took notes, it’s likely
that I will get them wrong when I produce the target.” Another instance was B1’s comment on how she had at times failed to grasp the logic between messages in the comprehension phase in the FSR setting. “In an FSR setting I’m accustomed to taking note of nouns or bigger concepts instead of logic and narrative because I don’t have enough time” (B1). As B1 has said, that was the reason why she “was forced to drop a lot of links between segments and beginnings of sentences” during note-taking. Yet another Y1 participant remarked on noting down the numbers without understanding the context (A2). She commented that “I panicked a little after I heard the numbers, because I didn’t understand what the speaker was talking about when he spoke the numbers.”
In short, notes that could not help retrieve the messages in the source text due to comprehension difficulties was reported by both Y1 and Y2 participants as a problem, while more Y1 participants mentioned the difficulty of understanding the context around numbers in particular. Moreover, some Y2 participants commented on not being able to grasp and write down the speaker’s tone. This was not reported in any comments from Y1 participants.
4.2.2.2 Difficulties in attention allocation
Perhaps the crux of the comprehension problems in 4.2.2.1 lied in the fact that interpreting is a multi-tasking activity. Eight participants (40% of all participants) commented on the imbalanced attention allocation among different Efforts affecting note-taking efficiency. While FSR to a certain degree may not impair comprehension of listeners, for interpreters, a SR over a certain degree may cause difficulty in moving
balance between listening and note-taking. They generally felt that in the FSR setting, more attention should be allocated to listening and understanding the message. However, they commented on how this took a toll on note-taking. It was reported that “listening alone occasionally took so much brain power (B20)” that note-taking was affected. The pressure of “listening and thinking about what to write down at the same time (A6)”
was felt by several participants. In this case, because of FSR, attention was compelled to move from the Listening and analysis Effort to the Note-taking Effort so fast that at least one of the Efforts showed signs of declining processing capacity available.
On one hand, inadequate processing capacity for the Note-taking Effort can happen when interpreters do know what to write down because the brain is busy listening and understanding the message.
“I understood the message, but I didn’t have time to think of a keyword to write down to help me remember the message, so the only option was to write down the proper nouns.” (A1)
“I was compelled to focus on listening, so I didn’t know what to write down.”
(A4)
“FSR made note-taking somewhat more difficult. The quantity of notes was limited due to lack of time, so I had to really think about what to write down.”
(A5)
On the other hand, inadequate processing capacity for the Listening and Analysis Effort can happen when interpreters do not have enough time to understand a segment
because that makes note-taking extremely difficult. However, notes produced this way was at high risk for becoming inefficient retrieval cues, because the messages were not necessarily understood (see 4.2.2.1).
“I used almost all my mental energy on note-taking so that little was left for listening and understanding. That’s why I was compelled to write something down as soon as I heard the word.” (A3)
“I didn’t have the time to fully understand a segment before writing a few words as retrieval cues, so I had no choice but to write exactly what I’ve heard.” (B19)
“When the speaker started to list a lot of examples, there was no time to mentally process the message. I could only write down the things I’ve heard.
But I didn’t really grasp the speaker’s tone, so I think there were some omissions [in my target]” (B4)
In addition, a participant mentioned FSR affecting the Short-term memory
operations Effort. She speculated that “I think there was a possibility that FSR affected my memory. I had to keep thinking of effective keywords to write down, so my brain was too tired from all the multitasking (A1).” When a message is neither stored in short-term memory nor in the form of notes, it is likely that it will not be rendered in the target text.
Lastly, some participants also reported that there was not enough time to write the things they wanted to write. It was sometimes because “I was still busy writing the last
segment. (A8)” or simply because “the speed was so fast that I could not finish a word (A6).” When this happened, B12 recalled trying to quickly add details to a previous segment while she still remembered, but “sometimes I couldn’t put it down correctly because I forgot the message or didn’t remember it correctly.”
In sum, a number of participants acknowledged that one of the difficulties in note-taking came from the need to allocate their mental energy perfectly among the different Efforts in the comprehension phase. There was no apparent difference in perception of this difficulty based on the description of Y1 and Y2 participants, despite the fact that more Y1 participants reported this difficulty.
4.2.2.3 Messy note structure and handwriting
Four Y2 participants (20% of all participants) observed difficulty in presenting note units in a neat form. First, in regards to note structure, B5 claimed that in his notes he was “not able to present the logic in the speech by, for instance, placing horizontal
Four Y2 participants (20% of all participants) observed difficulty in presenting note units in a neat form. First, in regards to note structure, B5 claimed that in his notes he was “not able to present the logic in the speech by, for instance, placing horizontal