Frequent V-/A-N Collocations in RAC

The first research question aims to identify frequent V-N and A-N collocations in published authors’ writing in the field of applied linguistics. To answer the first research question, V-N and A-N collocations appearing more than 5 times PMWs and 1% DP of the texts in RAC were extracted. Table 4.1 shows the numbers of type, overall frequency of occurrences, normalized overall frequency of occurrences, and TTRs of these V-N

40

and A-N collocations identified in RAC.

Table 4.1. Overview of V-/A-N Collocations in RAC.

Type RF (Token) NF ANF TTR

V-N 181 32,721 2,759.4 15.3 0.55%

A-N 248 48,220 4,066.5 16.4 0.51%

Note. RF=Raw Frequency, NF=Normalized Frequency (per-million-word), ANF=Average Normalized Frequency (per collocation type), TTR=Type-Token Ration.

The number of types, ANF, and TTR of the V-N collocations were compared to those of the A-N collocations to explore the differences between the published authors’

use of V-N and A-N collocations. For the number of types, there were 181 types of V-N collocations and 248 types of A-N collocations identified in RAC. The number of A-N collocation types was higher than that of V-N collocation types. This finding corroborates Ackermann and Chen’s results (2013), that is, A-N collocations form the largest lexical collocation group in academic writing. Regarding the ANFs, the ANF of V-N collocations was 15.3 times PMWs, and the ANF of A-N collocations was 16.4 times PMWs. The difference between the two ANFs was not obvious, which meant that, on average, V-N collocations appeared as similarly frequent as A-N collocations did in the academic writing of published authors. A small difference between the TTR of V-N collocations (TTRVRAC3=0.55%) and the TTR of A-N collocations (TTRARAC4=0.51%) was also obtained, showing that the lexical diversity of V-N collocations was similar to that of A-N collocations in the published authors’ writing.

In fact, the subtle differences between the two ANFs and the two TTRs reflect the published authors’ good knowledge of both academic V-N and A-N collocations in the

3 VRAC= V-N collocations in RAC.

4 ARAC= A-N collocations in RAC.

41

field of applied linguistics. The two approximate TTRs demonstrate that the published authors possessed sufficiently balanced lexical repertoire of V-N and A-N collocations, and the two ANFs reveal that the authors consistently varied their use of V-N and A-N collocations to a similar degree.

As pointed out in the precious chapter, one of the aims in the current study is to fill the gap of lacking discipline-specific lists of academic collocations. To examine whether collocations obtained in this study were indeed specific to the field of applied linguistics, V-N and A-N collocations identified in the published authors’ writing were compared with those included in Ackermann and Chen’s Academic Collocation List (ACL, available at http://pearsonpte.com/research/academic-collocation-list/). In ACL, the researcher first extracted 56 types of V-N collocations and 238 types of A-N collocations consisting one of the investigated 29 core nouns, and further identified combinations frequently employed by the published authors in the current study. The comparisons revealed that many of the frequent V-N and A-N collocations employed by the published authors were not included in ACL. As shown in Figure 4.1, among all the V-N collocation types identified in RAC, only 31 (around 17.13%) were included in ACL. As for all the A-N collocation types in RAC, 51 (around 20.56%) were included in ACL. Table 4.2 presents the frequent V-N and A-N collocation types shared in ACL.

Figure 4.1. Venn Diagram of V-N and A-N Collocations Distribution.

42

Table 4.2. Frequent V-N and A-N Collocations Shared in ACL.

Verb-Noun Adjective-Noun

Collocation Ranking in ACL Collocation Ranking in ACL

draw (a) conclusion 568 broader context 161

provide context 1797 cultural context 442

provide (an) example 1817 institutional context 1149

report findings 1966 social context 2131

collect information 268 specific context 2185

contain information 365 major difference 1320

gather information 879 significant difference 2068

give information 917 cultural differences 443

obtain information 1495 individual differences 1117

process information 1768 main findings 1301

provide information 1803 specific form 2190

receive information 1898 additional information 41

address (an) issue 45 detailed information 503

discuss (an) issue 542 relevant information 1944 explore (an) issue 744 specific information 2192

raise (an) issue 1878 key issue 1195

acquire knowledge 32 personal knowledge 1586

require knowledge 1969 previous knowledge 1728

pose (a) question 1661 prior knowledge 1760

raise (a) question 1877 natural language 1441

conduct research 331 appropriate level 89

obtain (a) result 1496 high level 985

assume (the) role 101 low level 1282

examine (the) role 721 complex process 314

play (a) role 1614 specific question 2199

take (a) role 2283 current research 466

conduct (a) study 332 earlier research 579

perform (a) study 1576 empirical research 654

review (a) study 1980 further research 869

complete (a) task 308 future research 873

perform (a) task 1577 previous research 1731

qualitative research 1856

43

longitudinal study 1279 previous study 1733

recent study 1903

The small proportions of V-N and A-N collocation types shared in ACL and in RAC reflected the disciplinary differences in the usage pattern of collocations between the field of applied linguistics and other academic disciplines. As pointed out in previous chapter, collocations in ACL were generated from a cross-discipline academic corpus, whereas collocations identified in the present study were based on texts in the field of applied linguistics only. The huge differences between collocations identified in the current study and those in ACL not only substantiate the existence of discipline-specificity in academic collocation use but also highlight the pedagogical value of the collocations retrieved in this study for student writers in the field of applied linguistics.

This study thus presented all the identified 181 V-N collocations and the 248 A-N collocations in Appendix A and Appendix B respectively. To make the two lists more comprehensible and useful to readers, these collocations were further clustered into different meaning groups. For instance, the verb-noun collocations draw conclusion and reach conclusion were clustered into the same meaning group since they expressed a similar meaning of get into a conclusion. Consultation of the Macmillan Collocation Dictionary and the Oxford Collocation Dictionary for Students of English was carried out during the meaning-clustering process to resolve any confusion of the collocations’

meanings. It is hoped that these lists can serve as useful references for leaner writers in the field of applied linguistics.

In this section, the researcher exhibited the published authors’ general usage pattern of frequent V-N and A-N collocations in the field of applied linguistics, and demonstrated the discipline-specificity of these collocations by comparing with collocations retrieved in a more general academic corpus. The next section then moved

44

on to report findings of the frequent V-N and A-N collocations in the Taiwanese EFL leaners’ writing, and further compared V-/A-N collocations employed by the leaners to those by the published authors in order to uncover the differences between the two writer groups’ use of academic collocations.