      In this part, I explain the findings described above, regarding the differences and/or the similarities of the types and the functions of fillers used by the first and the seventh semester students.
First of all, the fact that the total number of fillers produced by the first and the seventh semester students is quite significant, which is 316 fillers and 424 fillers, may mean that the longer the studying time or the higher the semester, the more fillers they produced. This might happen due to the length of the conversation because the seventh semester students, as the higher semester, averagely produced longer conversation than the first semester students, as the lower semester, since the longer the conversations, the more fillers occurred.
Regarding the number of silent pauses used, there is also a significant difference between the first and the seventh semester students in producing unit (.), double unit (--), and treble unit (---), which is 10.76% : 12.50%, 5.70% : 2.12% and 1.90% : 0.71%. This may suggest that the length of studying time gives a significant result on the use and the frequency of silent pauses. Then, it may also imply that seventh semester students are better than the first semester students in using silent pauses appropriately since the less occurrence of double unit (--) and treble unit (---) may mean that they are able to fill the gap quickly with other kinds of fillers, such as ee and ehm, instead of letting the silence fill the pause or letting the listener wait for the next utterance. When the seventh semester students need to fill the pause, they probably prefer to use unit (.) which does not take longer time as well as double unit (--) and treble unit (---). No wonder that the numbers of unit (.) by the seventh semester students is bigger than the numbers of unit (.) by the first semester students.
Moreover, there is a small difference in unlexicalized filled pauses since the similarities are very big. The first semester students produced 27.85% ee while the seventh semester students produced 38.68% ee. Then, it may imply that the first semester students are better that the seventh semester students since out of a lot of types of fillers, the seventh semester students preferred to use ee. It is supposed to be good if they are able to use, not only ee, but also the other types of fillers. Besides, the more the ee occurred, it may sound as disfluency.
Moreover, in lexicalized filled pauses, there is also a great difference between the first and the seventh semester students in producing I think, which is 4.11% : 1.42%. This may mean that the first semester students are better than the seventh semester. It may suggest that the seventh semester students are not accustomed to use I think when stating their opinions. Maybe they are accustomed to directly state their opinion by saying "I disagree with..." or "I don't agree with..." while the first semester students are accustomed to fill the pause and to hesitate by producing I think.
In addition, the fact that the first semester students did not produce oh, I guess, like, sort of, what is it, and what else, while the seventh semester students did not produce ah, huh, nah, and well does not result on a significant difference because the difference range is small, compared to the similarities. This insignificant difference may suggest that the length of studying time does not result on the appearance of some examples of fillers produced. This might be because the students are not accustomed to produce those fillers. Probably each speaker has her/his own favorite fillers that s/he produces when s/he is speaking, so probably s/he does not produce the other fillers besides her/his favorite ones. For example, there are two students from each semester produced I guess instead of I think although both have the same functions. Maybe those fillers above, such as oh, I guess, like, sort of, what is it, what else, ah, huh, nah, and well did not appear for some students in different semester because maybe there are other fillers that can be used similarly, for example kind of for like and sort of; what's that for what is it and what else; yeah or ya for ah, huh, nah, and well; and eh for oh. Possibly, the disappearance of some fillers above is a matter of being accustomed to produce those fillers or not.
Still related to lexicalized filled pauses, interestingly, the use of well and ya gives a significant difference for the first and the seventh semester students with the comparison 6.96% : 0% and 2.53% : 9.43%. This probably happen because the seventh semester students may prefer to produce ya instead of well which actually has the same function. No wonder that none of the seventh semester students produced well, but the occurrence of ya is very high.
Interestingly, ya is not a native-language filler, but it belongs to Indonesian one since we have the word ya in Bahasa. The use of well (native-language filler) by the first semester students and the use of ya (Indonesian filler) by the seventh semester students may imply that the first semester students have a better quality in using fillers than the seventh semester students, because the more similar to the native speaker, the better the use of fillers. Â
Besides ya, there is also an Indonesian filler, which is nah. This nah occur one time only in the first semester student's occurrence. Nah does not belong to the native-language fillers, such as huh, ee, err, and so on, but it belongs to Indonesian fillers since nah is commonly used by Kalimantan people. This nah appeared when the speaker was talking her experience enthusiastically. "... when I go to Petra ... when I go with, you know, like a general transportation like bemo, nah, ee, when you ah, when you use it, ee they they just stop it and they need some passenger....". The occurrence of nah by the first semester students may imply that when the speaker is talking her/his experience or narrating her/his story enthusiastically, sometimes s/he does not realize that the local fillers may appear due to the speaker may focus on the plot and the clearance of the story instead of the use of fillers. Maybe these local fillers do not appear when the speaker is getting involved in a light spoken interaction since s/he could control the use of non-native language fillers.
Although there are some differences explained above, actually there are also some similarities between the types of fillers by those students. For example, both the first and the seventh semester students used (.) as the most frequent type of silent pauses, followed by (--) in the second position, and (---) in the third position. Probably it is influenced by the topic discussed in the conversation, which is about Petra Square Apartment, which may not be too difficult, so to search the right word, they did not need more than 3.2 to 16 seconds. No wonder, (.) appeared the most in the conversation.
For unlexicalized filled pauses, there is no difference for the two most frequently used types, which is ee as the most frequently used type, followed by ehm in the second position. This may imply that both the first and the seventh semester students are not good enough since out of several types of unlexicalized filled pauses, they just used two types only as the most. Perhaps people sometimes do not realize when they say ee and ehm because maybe ee and ehm are the simplest and the easiest fillers to produce, compared to others.
