|Year : 2019 | Volume
| Issue : 1 | Page : 11-17
A study on the most frequent academic words in high impact factor english nursing journals: A corpus-based study
Department of English Language, Faculty of Medicine, Lorestan University of Medical Sciences, Khorramabad, Iran
|Date of Web Publication||7-Dec-2018|
Mr. Yadollah Pournia
Department of English Language, Faculty of Medicine, Kamalvand Campus of Lorestan University of Medical Sciences, Khorramabad-Boroujerd Road, Khorramabad, Lorestan
Source of Support: None, Conflict of Interest: None
Background: The ability to comprehend a text depends primarily on the knowledge about its words. This study investigated the most frequent words in high impact factor (IF) English nursing journals. Materials and Methods: This corpus-based study was conducted on the articles of 13 English nursing journals with an IF of over 0.7 from November 2014 to September 2016. After the typographical errors were corrected and the tokens (running words) in each journal were equalized, the tokens were analyzed using the Range software. Finally, a word list was extracted from the final 2851 articles and 8196,953 tokens to reach the optimal 98% vocabulary coverage. Results: A word list consisting of 1081 word families and 3175 word types with 5.24% coverage was extracted, which fulfilled the 98% vocabulary coverage. In other words, the coverage of the 1081 word-family list (5.24%), the coverage of the 1st 3000 English word families (87.55%), proper names, marginal words, compound words, and abbreviations related to the software (3.29%), and the coverage of the new proper names (1.13%), new compounds (0.02%), new abbreviations (0.72%), and letter–number combinations (0.05%) totaled 98%. Conclusions: By learning the 1st 3000 English word families and the 1081 word families introduced in this study, a nursing student can comprehend the texts of articles in high IF nursing journals without any considerable help from other resources.
Keywords: Journal impact factor, nursing, vocabulary
|How to cite this article:|
Pournia Y. A study on the most frequent academic words in high impact factor english nursing journals: A corpus-based study. Iranian J Nursing Midwifery Res 2019;24:11-7
|How to cite this URL:|
Pournia Y. A study on the most frequent academic words in high impact factor english nursing journals: A corpus-based study. Iranian J Nursing Midwifery Res [serial online] 2019 [cited 2019 Jan 21];24:11-7. Available from: http://www.ijnmrjournal.net/text.asp?2019/24/1/11/247045
| Introduction|| |
Learning the vocabulary of a language is an essential part of any educational program to learn that language, because the ability to comprehend a text is mainly dependent on the knowledge about the vocabulary of the text., However, teachers of English for specific purposes (ESP) educational programs are often uncertain about what vocabulary their students need to learn., To provide learners of ESP with the required language to study texts, they have to be provided with the vocabulary they really need.
In 2000, the Academic Word List (AWL) consisting of 570 word families with a nearly 10% coverage in academic texts was introduced by Coxhead. The word list was obtained from the General Service List provided by West in 1953. The effectiveness of the AWL has sometimes been questioned, since the word list had the lowest coverage in the texts of sciences including medical sciences with a 9% coverage, and many of the most frequent words in this list rarely appear in the articles of medical sciences. In addition, the effectiveness of the frequent 2000 word-family General Service List from which the AWL was obtained has also been criticized due to the age of the list and the small size of the relevant corpus. Considering these problems of the two lists, some researchers put emphasis on finding frequent discipline-specific vocabulary consistent with the requirements of the discipline.,,
Real, comprehensive, public, private, verifiable, and applicable knowledge about humans is obtained through scientific researches, the results of which are often published in the form of articles in scientific journals. In terms of being dynamic, scientific journals are not comparable with textbooks as the contents of many textbooks become outdated even before their publication and they have little educational value for their readers. Journals with high impact factors (IFs) play an important role in providing this dynamic knowledge, and IF is the main and most common indicator to judge the quality of a journal., The yearly IF of a scientific journal is the average of citations to all of the articles published in the journal in the previous 2 years and is obtained by dividing the number of citations to the articles published in the journal by all of the articles published in the journal in the previous 2 years.
The importance of nursing scientific journals cannot be ignored. The spread of novel nursing knowledge, which stems from scientific research and is published in nursing journals, is vital for the development of the nursing profession, and this knowledge can have a much higher quality and can be provided much faster compared to what is stated in nursing textbooks. In recent years, due to the emphasis placed by modern medical education on “evidence-based medicine,” reading scientific and research papers has been highlighted. In fact, the aim of teaching English to the students of medical sciences in higher levels is mainly to help them in reading and then writing research papers.
Several studies have been conducted on the most frequent vocabulary in nursing texts. Takakubo (2003) analyzed a list of 2650 words at the end of 10 nursing textbooks. Budgell et al. introduced a list of 1000 frequent words by studying nursing articles consisting of 250,000 tokens (running words). Mukundan and Jin provided a list of 1004 specialized words by analyzing 3490,417 tokens. Nor Mohamad and Jin extracted a 2000-word list of frequent words by working on 3640,760 tokens in 7 nursing textbooks. In addition, Yang, by doing research on 252 nursing articles consisting of 1006,934 tokens, introduced a list of 676 frequent word families. Most of these studies have investigated nursing vocabulary in nursing textbooks, and none of them has investigated frequent vocabulary in a large number of nursing articles in high IF nursing journals.
Given the importance of vocabulary in learning any language,,,, the importance of scientific research, the importance of high IF scientific journals,, the importance of nursing scientific journals,, and the need of medical sciences students to English for reading and writing scientific articles in medical journals, this study investigated the most frequent words in high IF (>0.7) English nursing journals and introduced a word list as the High-impact Nursing Academic Word List (HI-NAWL). In addition, the coverage of the 1st common 3000 word families of English and the Coxhead's AWL in nursing scientific journals with an IF over 0.7 was also investigated.
| Materials and Methods|| |
This quantitative corpus-based study was conducted from November 2014 to March 2015 (downloading the articles) and from July 2015 to September 2016 on articles downloaded from the ProQuest Nursing and Allied Health Source section of the ProQuest database. The ProQuest database contains thousands of articles published in hundreds of scientific journals and is composed of several smaller databases. The ProQuest Nursing and Allied Health Source includes hundreds of health and nursing scientific journals. First, the general word of “nursing” was searched by ticking the “Full-text” and “Peer-reviewed” tabs so that the found articles would possibly be the most relevant full-text and peer-reviewed articles in nursing. Then, in the journal section which showed the names of the journals in which the found articles had been appeared, journals with an IF higher than 0.7 were selected based on the list of IF in 2012. This cutoff point was chosen based on the number of the journals with an IF in the search results. Most of the journals did not have an IF, and some of the journals with an IF did not have enough full-text articles to be downloaded. IF in nursing journals is generally much lower than that in journals of medicine so that the highest IF in nursing journals (IF = 2.50) belonged to the Journal of Oncology Nursing Forum. Therefore, an IF over 0.7 is considered high in nursing. In total, there were 13 nursing journals with an IF over 0.7 in the list of the journals [Table 1]. Subsequently, the articles of each journal were arranged and downloaded “most recent first.” In fact, the nonprobability consecutive sampling method was used. Downloading was continued to guarantee the collection of the intended sample size of at least 630,000 tokens for each journal and the corpus (collection of texts) size of at least 8 million and 190,000 tokens for the 13 journals. The size of at least 8 million tokens was specified arbitrarily because the present study aimed to be the largest ever done study on nursing vocabulary. The number of the articles needed to cover the words in each journal was not exactly clear. Therefore, the number of the articles downloaded (6000 articles) was more than necessary. The time period of the downloaded articles was between the years 2002 and 2014.
First, the articles were downloaded with Word or PDF formats. The file of each article was converted to Text format and the sections of authors and affiliations, abstracts, acknowledgments, references, tables, and figures were removed from each file. In accordance with the requirements of the software Range used in the study, the necessary corrections were made in each file. Then, the articles of each journal were collected and saved in one Text-format file, and new articles were added to the file to collect the intended tokens and corpus. To conduct corpus-based studies, millions of tokens are required to ensure the availability of large volumes of texts and language samples. The tokens in each journal were counted using the software repeatedly after adding each article to the file of the journal. In this step, the number of the tokens for each journal was more than the minimum required because the number of the tokens would change in the next step due to the required corrections.
In the next step, typographical errors in each journal file were corrected using the Word software. Then the tokens in each journal were recounted to guarantee the presence of 630,000 tokens with a difference of at most 1000 additional tokens in each journal. The additional tokens more than the range of 630,000 + 1000 were removed from the end of the file of each journal. All the files prepared for the 13 nursing journals with roughly equal number of tokens (630,000 + 1000) were analyzed using the software Range, a free software developed by Heatley, Nation, and Coxhead through the 29 word lists of the software including the 25,000 word-family lists of The British National Corpus and the Corpus of Contemporary American English (BNC/COCA) to determine the frequency and range of all the tokens or word families. The 29 lists of the software include the 1st 25,000 word families of English compiled in 25 lists, each with 1000 word families, along with four lists of proper names, marginal words, compound words, and abbreviations. These 29 lists have been prepared based on two big corpora or collections of texts in American and British English. When a text is run on the software, the software measures the frequency and range of each word and the word family to which that word belongs separately based on the words in the 29 lists. When a word does not exist in the 29 lists, the software labels that word as “Not in the Lists.” Frequency refers to the number of occurrences of a token or word family, and range refers to the number of the files (journals) in which the token or word family is repeated. In this step, some of the tokens were not in any of the lists of the software and were classified in five separate files of new word families, proper names, compound words, abbreviations, and letter–number combinations in accordance with the requirements of the software.
All the corrected files for the 13 journals consisting of 8196,953 tokens were analyzed through the software through the 34 word lists consisting of the 29 word lists of the software and the 5 word lists prepared in this study to determine the frequency and range of all the tokens or word families.
In the last step, the most frequent word families in the 25,000 word-family lists of BNC/COCA of the software excluding the 1st 3000 word families – which are expected to be learned before entering the university or at most after passing the general English credits before the specialized credits in the university – and the word list of new words prepared in this study were selected using the software.
In this study, the concept of “word family” was used to select the words. In other words, a basic word and its inflected forms and derivations (if any) based on Level 6 of the scale of Bauer and Nation were considered as a word family. For example, the basic word of activate and its following inflected forms and derivations including activated, activates, activating, activation, activator, activators, inactivation, reactivate, reactivated, reactivates, reactivating, reactivation, reactivations, and unactivated are composed of one word family with 15 word types.
If all the above 15 word types appear, for example, 100 times in all the texts, there will be 100 tokens or running words. Therefore, in this example, there will be one word family for the basic word of activate, with 15 word types, and 100 tokens or running words.
To easily select the most frequent words, the 1st three 1000 word-family lists were converted into one 3000 word-family list and the next twenty-two 1000 word-family lists into one 22,000 word-family list. The 25 lists were converted into two lists because of two reasons. First, the conversion would give a better analysis of the 1st 3000 word families of English combined. Second, having one list instead of 22 lists would save a lot of time in selecting the frequent words since the words could be selected once in one list, not 22 times in 22 lists separately. Moreover, the other four lists of the software including proper names, marginal words, compound words, and abbreviations were converted into one list. In total, there were eight lists including three combined lists of the software and five lists prepared in this study. The most frequent words were selected from the 2nd 22,000 word-family list of the software and the fourth list of new word families prepared in this study.
Two criteria of frequency and range were considered to select the frequent words, and the criterion of the range was before frequency. In other words, words with high frequency would be selected only if they appeared in at least more than half (seven) of the journals. The selected words had to have a range of seven or higher, meaning that they had to be used in at least 7 or more of the 13 nursing journals. The selection of the frequent words was continued to reach 98% coverage, which is the required coverage for the optimal comprehension of language texts without any help from any other sources. In other words, the total of the coverage of the 1st 3000 word families of English and their related compounds, the coverage of the words without any meaning loads such as proper names, marginal words, abbreviations, and letter–number combinations, and the coverage of the selected frequent words had to be 98%.
Due to the large number of the whole words, the selection of the minimum or cutoff frequency required to select the frequent words needed a lot of calculations and trial and errors. After many calculations and trial and errors, it was found that the minimum frequency was 92. In other words, the members of the selected word families had to be repeated at least 92 times in the 13 nursing journals. After removing the word families with a range of <7 using the software of Excel, a list of 1081 word-families was selected as the HI-NAWL, which fulfilled the required 98% coverage. In addition to selecting this list of frequent words, the frequency of the 1st 3000 word families of English and the frequency of the Coxhead's AWL were investigated in the corpus of the 13 nursing journals.
This article was derived from a research project (no. 1948-2015) approved by the Research Committee of Lorestan University of Medical Sciences. The study was first approved by the Research Committee of the Faculty of Nursing and Midwifery of the university.
| Results|| |
The number of all the articles of the 13 journals in the final stage, after the corrections and classifications, decreased to 2851 articles [Table 1] with 8196,953 tokens. The information of all the tokens of the articles in the 34 word lists is shown in [Table 2]. According to this table, the 8196,953 tokens of the nursing journals consisted of 82,145 word types and 62,148 word families.
|Table 2: Place of all the tokens of the 13 nursing journals in the 34 word lists|
Click here to view
The analysis of all the tokens after converting the 34 word lists into eight word lists is presented in [Table 3]. According to this table, the first 3000 word families of English (List 1) covered 87.55%, the next 22,000 word families (List 2) 6.52%, other lists of the software including proper names, marginal words, compound words, and abbreviations (List 3) 3.29%, new word families prepared in this study (List 4) 0.71%, new proper names (List 5) 1.13%, new compound words (List 6) 0.02%, new abbreviations (List 7) 0.72%, and letter–number combinations (List 8) 0.05% of all the tokens in the journals.
|Table 3: Place of all the tokens of the 13 nursing journals in the eight word lists|
Click here to view
The selection of the frequent words was performed based on the data presented in [Table 3] to reach 98% coverage. The frequent words were selected from the 2nd 22,000 word families of English (List 2) and the new word families prepared in this study (List 4). The results showed that a list of 1081 word families [Supplementary File] consisting of 3175 word types with a range of seven or higher with 5.24% coverage, named the HI-NAWL, fulfilled the required 98% coverage. In other words, the total of the coverage of the 1081 word-family HI-NAWL (5.24%), the coverage of the 1st 3000 word families of English (87.55%), the coverage of other words of the software such as proper names, marginal words, compound words, and abbreviations (3.29%), the coverage of new proper names (1.13%), the coverage of new compound words (0.02%), the coverage of new abbreviations (0.72%), and the coverage of letter–number combinations (0.05) were equal to 98%, which is the required coverage for the optimal comprehension of language texts without any help from any other sources.
The result of the coverage of the Coxhead's AWL showed that 569 word families of this 570 word-family list covered 11.75% of all the tokens in the 13 nursing journals.
| Discussion|| |
This study was conducted on 2851 full-text and peer-reviewed articles consisting of 8196,953 tokens in 13 English nursing journals with an IF over 0.7. The number of the words in this study (8196,953 tokens) is much more than the numbers in the studies conducted on nursing words by Takakubo, Budgell et al., Mukundan and Jin, Nor Mohamad and Jin, and Yang. The number is even more than twice the number of the words in the studies by Mukundan and Jin with 3490,417 words and Nor Mohamad and Jin with 3640,760 words, who have used the highest number of words so far. The reason for using a very high number of words in the present study is that, in corpus-based studies, millions of words are required to ensure the availability of large volumes of texts and language samples. Moreover, the number of the articles in this study (2851 articles) is much more than the number of the articles in Budgell et al.'s study conducted on the articles in one volume of six nursing journals, and in Yang's study with 252 nursing articles. In addition, instead of using a limited number of nursing textbooks similar to the studies by Takakubo, Mukundan and Jin, and Nor Mohamad and Jin, the present study was conducted on many articles (2851 ones) from 13 HI journals. The reason for using numerous articles instead of a limited number of textbooks is that the high number of articles written by numerous authors will solve the problems of individual styles in writing and the high frequency of specific words by a specific author on a specific subject or field of study. Moreover, in terms of being dynamic, scientific journals are not comparable with textbooks as the contents of many textbooks become outdated even before their publication, and they have little educational value for their readers. All in all, according to an extensive search on the Internet, it seems that the present study is a comprehensive research conducted on the highest number of words and articles in high IF nursing journals.
The results of the present study showed that the 1st 3000 word families of English covered 87.55% of all the 8196,953 running words in the journals. Furthermore, the next 22,000 word families together had a coverage of only 6.52% of all the running words [Table 2] and [Table 3]. In other words, of 100 words in the nursing journals, approximately 87 words were among the 1st 3000 word families, and only six words were among the next 22,000 word families. Of the 1st 3000 word families, only 10 word families had not been used in the journals. However, of the next 22,000 word families, only 9681 word families had been used, and 12,319 word families had not been used even once in the journals. This result indicates the high frequency of the 1st 3000 word families compared to other word families.
The results of the present study introduced a list of 1081 word families named the HI-NAWL [Supplementary File] consisting of 3175 word types with 5.24% coverage, which fulfilled the required 98% coverage. Unlike other studies conducted on nursing vocabulary, which have selected their frequent words outside the 1st 2000 word families of English,, the selected words in the present study were outside the 1st 3000 word families of English. Therefore, the coverage of the selected words in the studies by Mukundan and Jin (9.9%) and Yang (13.64%) is higher than that in the present study (5.24%). It should be noted the next 22,000 word families covered only 6.52% of all the tokens in the journals while the 1081 word families in this study covered 5.24% of all the tokens, indicating the importance of the HI-NAWL in this study. Selected academic words for various disciplines including nursing should be outside the 1st 3000 word families of English because these words are expected to be learned before entering the university, or at most after passing the general English credits before passing the specialized credits in the university.
Although the number of the selected words in this study (1081 word families) is higher than the numbers in other studies,,, none of these studies has considered the 98% coverage of words, which is the required coverage for the optimal comprehension of language texts without any help from any other sources. Considering the 98% coverage has resulted in the introduction of a higher number of words in the present study compared to these studies.,,
The results of the frequency of the Coxhead's AWL showed that 569 word families of this 570 word-family list covered 11.75% of all the tokens in the journals, which is a high coverage. The reason is that the words of this list, like the words of the other studies,,,,, are mostly among the first 3000 word families of English with very high frequency.
Downloading the articles, collection of the texts, classifications of the new word families, and the final analyses performed to select the final HI-NAWL in the present study were very time-consuming. Conducting similar studies requires a lot of time, energy, patience, and interest. Furthermore, the full-text articles of some famous nursing journals were not available, and this was one of the limitations of this study. It is recommended that similar studies be conducted on the words of other medical and nonmedical disciplines to extract the required discipline-specific vocabulary.
| Conclusion|| |
It can be concluded that by learning the 1st 3000 word families of English and learning the 1081 word-family HI-NAWL introduced in the present study, nursing students and other nursing groups can comprehend nursing texts in high IF nursing journals without any considerable assistance from other sources. Other words including proper names, marginal words, abbreviations, and letter–number combinations, which do not have any specific meaning loads, do not make any considerable problems. The vocabulary extracted in this research accompanied by appropriate passages and exercises can be compiled in educational books, and these books can be used by nursing students and other nursing groups to expand their academic vocabulary in the field of nursing.
The researcher appreciates the sincere help by the Faculty of Nursing and Midwifery, Lorestan University of Medical Sciences, for the initial approval, and the Deputy for Research of this university for the final approval and funding of this study (no. 1948-2015).
Financial support and sponsorship
Lorestan University of Medical Sciences
Conflicts of interest
Nothing to declare.
| References|| |
Hsu W. Measuring the vocabulary of college general English textbooks and English-medium textbooks of business core courses. Electron J Foreign Lang Teach 2009;6:126-49.
Csomay E, Petrović M. “Yes, your honor!”: A corpus-based study of technical vocabulary in discipline-related movies and TV shows. System 2012;40:305-15.
Konstantakis N. Creating a business word list for teaching business English. Estud Lingüíst Inglesa Apl 2007;7:79-102.
Nor Mohamad AF, Jin NY. Corpus-based studies on nursing textbooks. Adv Lang Lit Stud 2013;4:21-8.
Alizadeh I, Farjami H. Recounting and fine-tuning academic word list for four academic fields. Iran EFL J 2011;7:48-73.
Frazer S. Beyond the Academic Word List: Providing ESP learners with the words they really need. Proceedings of the BAAL Annual Conference; 2008. p. 41-4.
Coxhead A. A new academic word list. TESOL Q 2000;34:213-38.
West M. A General Service List of English Words. London: Longman, Green; 1953.
Tajino A, Dalsky D, Sasao Y. Academic vocabulary reconsidered: An EAP curriculum-design perspective. J Teach Engl Foreign Lang Lit Islam Azad Univ 2009;1:3-21.
Chen Q, Ge G. A corpus-based lexical study on frequency and distribution of Coxhead's AWL word families in medical research articles (RAs). Engl Specific Purposes 2007;26:502-14.
Browne C. A new general service list: The better mousetrap we've been looking for? Vocabulary Learn Instr 2014;3:1-10.
Hyland K, Tse P. Is There an “Academic Vocabulary”? TESOL Q 2007;41;235-53.
Masic I. How to search, write, prepare and publish the scientific papers in the biomedical journals. Acta Inform Med 2011;19:68-79.
Weatherall DJ, Ledingham JG, Warrell DA. On dinosaurs and medical textbooks. Lancet 1995;346:4-5.
Swedlove F. Implications of the impact factor. Can J Occup Ther 2006;73:3-4.
Lokker C, Haynes RB, Chu R, McKibbon KA, Wilczynski NL, Walter SD, et al.
How well are journal and clinical article characteristics associated with the journal impact factor? A retrospective cohort study. J Med Libr Assoc 2012;100:28-33.
Moustafa K. The disaster of the impact factor. Sci Eng Ethics 2015;21:139-42.
Zucker KJ, Cantor JM. The impact factor: The archives breaks from the pack. Arch Sex Behav 2006;35:7-9.
Campbell-Crofts S. The future of nursing journals. Renal Soc Australas J 2012;8:6.
Mungra P, Canziani T. Lexicographic studies in medicine: Academic Word List for clinical case histories. Ibérica 2013;25:39-62.
Frazer S. Building corpora and compiling pedagogical lists for university medical students. Hiroshima Stud Lang Lang Stud 2013;16:65-88.
Takakubo F. Analysis of Vocabulary in English Textbooks for Student Nurses. The Language Teacher; 2003.
Budgell B, Miyazaki M, O'Brien M, Perkins R, Tanaka Y. Developing a corpus of the nursing literature: A pilot study. Japan J Nurs Sci 2007;4:21-5.
Mukundan J, Jin NY. Development of a technical nursing education word list (NEWL). Int J Innov Engl Lang Teach Res 2012;1:105-24.
Yang MN. A nursing academic word list. Engl Specific Purposes 2015;37:27-38.
Oermann MH, Shaw-Kokot J. Impact factors of nursing journals: What nurses need to know. J Contin Educ Nurs 2013;44:293-9.
Nation IS, Webb S. Researching and Analyzing Vocabulary. Boston: Heinle Cengage Learning; 2011.
Hsu W. Bridging the vocabulary gap for EFL medical undergraduates: The establishment of a medical word list. Lang Teach Res 2013;17:454-84.
Bauer L, Nation IS. Word families. Int J Lexicography 1993;6:253-79.
Nation IS. How large a vocabulary is needed for reading and listening? Can Modern Lang Rev 2006;63:59-82.
[Table 1], [Table 2], [Table 3]