Integration of sources in academic writing : A corpus-based study of citation practices in essay writing in two departments at the University of Botswana

Copyright: © 2014. The Authors. Licensee: AOSIS OpenJournals. This work is licensed under the Creative Commons Attribution License. The ability to cite sources appropriately is an important feature of academic writing. Academic writers are expected to integrate ideas of others into their texts and take a stance towards the reported material as they develop their arguments. Despite this importance, research has shown that citation presents considerable difficulties for students, particularly nonnative English speakers. Such difficulties include using citations effectively in writing and understanding them in reading, expressing one’s voice and signalling citations in writing so that there is a clear distinction between one’s ideas and those derived from source materials. This study investigates the types of reporting verbs used by students to refer to the work of others and the extent to which they evaluate the work of others in their writing. It draws from a corpus of approximately 80 000 words from essays written by students in two departments at the University of Botswana (Botswana). The findings show that students used more informing verbs, associated with the neutral passing of information from the source to the reader, without interpreting the information cited, compared with argumentative verbs (which signify an evaluative role). The results of the study underscore the importance of teaching reporting verbs in the English for Academic Purposes classroom and making students aware of their evaluative potential.


Introduction
This article describes and interprets findings from a small corpus of non-native undergraduate essays from two departments at the University of Botswana. The analysis focuses on how students use citations to report the work of other writers, and it specifically focuses on reporting verbs used when citing and the range of reporting verbs used by students to report the work of other writers.
Citation is an important convention of academic writing. It enables writers to develop their arguments persuasively and effectively and thus establish their authority and credibility. Citation can be viewed as 'attribution of propositional content to another source' (Hyland 2002b:115) and 'central to social context of persuasion' (Hyland 1999:342).
In citation, the content can be attributed to another source by the explicit use of reporting verbs. Barton (1993:747) calls this evidentials of citation, which writers use to place a 'perspective on the literature'. Thomas and Hawes (1994:129) refer to an evidential as a 'metalinguistic representation of an idea from other sources'.
The use of reporting verbs also 'represents a significant rhetorical choice' (Hyland 2002:116) because it allows writers not only to report the source material but also to indicate their position or stance towards the cited material. Scollon, Tsang, Li, Yung and Jones (1998) illustrate that boundaries between the writer's voice and the voice of the source author are marked by direct and indirect quotations, which are signalled by reporting verbs.
Research into the use of reporting verbs in citation, and how writers use citations to position themselves in relation to other members of their discipline, has been investigated in depth (Charles 2006;Hyland 1999;Thompson 2001;Thomas & Hawes 1994;Thompson & Yiyun 1991). Research has also documented how citation is influenced by disciplinary culture (Harwood 2009;Hyland 1999).
Much work has also been devoted to the study of citation practices amongst both expert and novice writers. Such studies tend to reflect the complexities of the use of citations and hence highlight the difficulties novice writers experience as they learn to cite materials in their writing.
A study by Mansourizadeh and Ahmad (2011) compared the use and function of citation by experts and novice researchers who belonged to the same research community. They found that novice writers tended to use citation to attribute content to the source whilst expert writers used citations to provide support for their arguments and to justify their claims. They also found out that novice writers tended to use citations in isolation whilst expert writers were able to synthesise several sources in their writing.
Another interesting comparative study was conducted by Barton (1993) to explore how experienced and inexperienced writers used evidentials in their writing to describe a perspective cited from the literature and take a stance. The study sheds light on how reporting verbs can be used in the construction of stance in academic writing. The study was based on a discourse analysis of the use of evidentials in argumentative essays written by experienced academic writers and essays written by students at Wayne State University. Barton collected a sample of written texts from these two groups. The experienced writers came from a variety of academic fields and had written argumentative essays that appeared on the back page of the Chronicle of Higher Education newspaper. Barton argues that the texts produced by experienced writers were essays rather than reports or letters to the editor. Students' essays came from different majors and programs; they were written for a university writing proficiency requirement and were written in an examination situation.
In spite of the difference in experience between the two groups of writers and the writing conditions, Barton (1993) argues that both groups were constructing an argument for a general academic audience and their texts could be said to be similar in terms of genre family (argumentation). The texts were chosen to find out whether there were similarities or differences in the way the two groups of writers constructed an argument. Barton (1993) reports a significant difference in the use of source materials between the two groups. Whilst the experienced writers used evidentials of citation to cite as well as place the literature within a critical perspective, students used evidentials in a neutral manner. Barton (1993) encapsulates it in this quotation: the student writers generally maintain a neutrality in the use of evidentials of citation: the source authors (or the source, passages) state, say, call, refer to, believe and write, evidentials of citation which do not incorporate the perspective of the student writer. (p. 761) The studies on citation practices cited above have focused mainly on scholarly writings found in journals as well as student writing in the form of master's dissertations and doctoral theses. The majority of these studies have focused on the use of citations in texts produced by native Englishspeaking writers. In contrast, this study focuses on citation practices of undergraduate student writers in non-native contexts writing in two different disciplines.

Justification
Students writing a text for a university assignment are usually expected to draw from sources; lecturers attach much importance to how students integrate sources into their writing. The effectiveness of students' arguments is judged by their ability to build on what other people have said and, at the same time, show that they can take a position and be heard. The way students integrate sources into their writing is supposed to 'reflect both the utterer's voice and the voices of those from whom s/he has borrowed the text' (Scollon et al. 1998:228). However, research has shown that citation presents considerable difficulties for students, particularly non-native English speakers writing in English (e.g. Dong 1996;Groom 2000;Mansourizadeh & Ahmad 2011;Pecorari 2003). This is also echoed by Hyland (2002b:116) when he says that 'citation represents a feature which students, particularly non-native speakers, find difficult to either use effectively in their writing or understand correctly in their reading'.
Investigating citation practices in non-native contexts is therefore of interest in this study because it may shed light into the difficulties Botswana second language writers of the English language experience as they learn to cite materials in their writing. It will also inform us of the types of reporting verbs students use to refer to the work of others and the extent to which they evaluate the work of others in their writing. It is also of interest to compare the less experienced writers (media studies students) with the more experienced writers (primary education students). The results of the study will offer valuable pedagogical implications for the teaching of citation in the English for Academic Purposes (EAP) classroom. Therefore, the present study seeks to investigate and compare the types of reporting verbs used by students from two university departments to refer to the work of others and the extent to which they evaluate the work of others in their writing. The aims of the study are: (1) to find out whether there are any differences in the use of reporting verbs when referring to the work of others in essays written by media studies and primary education students and (2) to find out the types of reporting verbs students in the departments of Media Studies and Primary Education use when referring to the work of others.

Research methodology The corpus
The study is based on a small corpus consisting of 40 essays written by undergraduate students from the University of Botswana and is part of a broader study that looked at other aspects of English as a Second Language (ESL) writing. Table 1 shows the size of the corpus. The participants were from two departments, the Department of Media Studies, which is in the Faculty of Humanities, and the Department of Primary Education, which is in the Faculty of Education. The Department of Media Studies offers the Bachelor in Media Studies degree (BMS). The programme introduces students to the world of electronic and print journalism. It opens career opportunities in journalism, both print and electronic, as well as script writing, advertising and public relations. The Department of Primary Education runs in-service programmes for teachers who have been in the field for some time and who have had training up to diploma level. The department offers a Bachelor's degree in Education (B.Ed. Primary). It also offers a Bachelor of Educational Management programme for the improving of educational management. Primary Education students can be regarded as the more experienced writers compared to Media Studies students because they are teachers who have been in the field for a longer time.
The students were selected through convenience sampling and the researcher did not in any way consider selecting a sample that was representative of the entire population.
Only students who were willing to participate in the study submitted their essays and completed a consent form. The students were in the third or fourth year of their study and had all done a writing course in which they were taught the basics of academic writing. The course included referencing and citation skills, which are commonly incorporated in the EAP undergraduate writing course.
The essays students wrote were part of the regular assessment requirements: students were given a writing task that required them to write an academic essay of about 1200-1500 words as homework. The rubrics required students to cite materials they had read to support their arguments, including at least five different sources (see Appendix 1 for the assignment topics).

Analysis of citations
The study employed both qualitative and quantitative approaches, including frequency counts and text analysis. I initially searched for reporting verbs manually in each text. This was done according to the function performed by verbs in a sentence. Before going into the analysis, I shall present a classification of reporting verbs. Swales (1990) distinguishes between integral and nonintegral citations. In integral citations, the name of the cited author appears as part of the sentence, and in non-integral citations the name of the cited author appears in parentheses. According to Petric (2007:240), the integral citations are divided into: (1) those with a controlling verb, (2) naming (where the citation is a noun phrase or its part) and (3) noncitations (where the author's name is not followed by other data, such as publication year). The focus of the analysis of evidentials is on the use of integral citations with a controlling or reporting verb.
In the analysis of citations, I used a classification system used by Thomas and Hawes (1994) for reporting verbs. They classify the reporting verbs according to the kind of activity referred to. They developed three categories of reporting verbs, which are (1) real-world or experimental activity verbs, (2) discourse active verbs and (3) cognition verbs. The category of 'discourse activity verbs' corresponds to a classification by Thompson and Yiyun (1991) who use the term 'textual verbs'. According to Thompson and Yiyun (1991:369), this category refers to 'processes in which verbal expression is an obligatory component'.  Thompson and Yiyun (1991:370), these verbs refer to 'mental or physical processes that are part of research work'. Examples are measure, obtain, quantify and find. They explain that these categories are not watertight and therefore there can be overlaps between the categories, which can make assigning reporting verbs to a single category difficult.
In the analysis I focused on discourse activity verbs or textual verbs. A further division of these verbs is made by Thomas and Hawes (1994): they have two subcategories of discourse verbs, tentative and certainty verbs. According to Thomas and Hawes, certainty verbs state the reported proposition in conclusive and definite terms. They further group verbs of certainty into two sub-classes, which are informing verbs and argument verbs. These were the two classes that I was interested in for the analysis of reporting verbs.
According to Thomas and Hawes (1994:140), informing verbs are associated with the 'neutral passing of information from the source author to the reader via the reporting writer'. The writer does not comment or interpret the information being reported. These verbs do not signal evaluation or the writer's point of view and 'do not seem to have any explicit indication of persuasive intent on the part of the writer' (ibid:143). The verbs mean the same thing as said. Examples are report, note, refer to and so on. The other class of reporting verbs, argument verbs, signal an evaluative role in the sense that the writer interprets the information cited. Thomas and Hawes (ibid:140) stress that 'the verbs do not signal neutrality in the communication of the information as the writer's voice clearly intervenes'. They further argue that the 'semantics of the verbs in this group make it clear that a particular stance is being adopted' (ibid:40). Examples are suggest, maintain, assert and criticise. The use of these verbs makes it clear that the writer is adopting a particular stance. Hyland (2002b:116) also points out that using hedges like suggest opens an 'evaluative space' for the writer.
To locate reporting verbs, I searched the corpus using Wordsmith Tools 5 (Scott 2008) for citations that included a date or page number in parentheses. I recorded a citation only when the text attributed words specifically to an author, thus reflecting the analytical procedure that Hyland (2002b) uses in locating reporting verbs in his study. The paragraph extract below taken from a student's essay shows this: According to Tumara (2001) human beings engage in activities which deplete the natural resources to try to meet their basic needs and raise their standards of living. Warren (1992) points out that although women are equipped with knowledge about environment and natural resource base, and its use, they still engage in harmful environmental activities for critical reasons.
(EPI 442-student) As Thomas and Hawes (1994) indicate that there can be overlaps in assigning reporting verbs to a single category, I had to make decisions and examine the reporting verbs in context to determine whether they were informing verbs or argumentative verbs.
In order to validate the results that I got from the initial analysis of reporting verbs, I used other strategies to ensure the validity of the results. One such strategy is discussed by Dornyei (2007): the research design itself can be used to provide convincing evidence about the validity of the results. This concept of 'triangulation', which involves using multiple methods or perspectives, was used to do reliability checks.
I analysed the essays again after a period of 12 months had elapsed. Using the Wordsmith Tools 5 Concord tool, I did a reverse check, using a list of features I identified in the initial analysis as a starting point for the second analysis. Once I had identified reporting verbs, the corresponding features were searched for using the Concord tool. The concordance provides a list of all occurrences of the search words in a corpus together with the context in which they occur. The concordance output was further subjected to qualitative analysis. This was done by carefully examining the contexts of these features to ensure that they actually functioned as either informing or argumentative verbs. This reflected the analytical procedure that Mur Duenas (2010) uses in his analysis.

Results
The first research question sought to find out whether there were any differences in the use of reporting verbs when referring to the work of others in essays written by Media Studies and Primary Education students. Table 2 shows the total number of reporting verbs in the two sub-corpora and the frequency of reporting verbs per 1000 words. As can be seen from the table, a comparison of the two subcorpora reveals a higher frequency of occurrences (both raw frequencies and frequency per 1000 words) in the use of reporting verbs in the Primary Education corpus than in the Media Studies corpus.
The results indicate 39 occurrence of reporting verbs in the Media Studies corpus. These occurred in nine essays with a maximum use of five and a minimum use of three per essay.
The results also indicate a very high occurrence of reporting verbs in the Primary Education corpus. The results show 158 instances of use of reporting verbs in the whole corpus. These occurred in 18 essays with a maximum use of 18 and a minimum use of two per essay. The two groups differed substantially in the use of reporting verbs.
A t-test for independent samples results showed a significant difference in the use of reporting verbs in the two subcorpora. As can be seen in Table 3, there is a significant difference between the two groups (t = -3.074; df = 38; p < .05). This is an interesting finding because both Media Studies students and Primary Education students were required to use sources in their essays, but we see a significant difference in the frequency of use of reporting verbs by the groups.
The second research question concerns the types of reporting verbs students use when referring to the work of others.
Examining the types of reporting verbs employed by the students helped to determine the rhetorical function of the verbs in use. Generally, the results show that students used both informing and argumentative verbs in their writing but to different degrees.
The pattern we see is that, across the two departments, informing verbs were by far the most common whilst there were substantially fewer argumentative verbs. There were 27 instances of informing verbs and 12 instances of argumentative verbs in the Media Studies corpus. In the Primary Education corpus there were 119 instances of informing verbs and 39 instances of argumentative verbs.
The students used informing verbs, which are associated with neutral passing of information from the source to the reader without any indication of persuasive intent on the part of the writer. Moreover, their use was from frequent than argumentative verbs, which signal the evaluative role in the sense that the writer interprets the information cited. This signalled the degree to which they commit to or detach themselves from the reported proposition. Example 1 shows the use of an informing verb used to accompany a citation: Example 1: Zimbabweans are willing to work and do anything at low prices, therefore occupying job opportunities for the locals. Furthermore, politics perpetuate the situation by debating that the ruling government offers foreigners job opportunities more than the locals. Reynolds (1987 p. 175) says: Living in large complex societies, often composed of several different ethnic groups, has become the fate of most of the world's population today. Nations  and state figure predominantly on the international scene, with politicians vigorously and selfishly pursuing whatever they perceive in their nations interest. (BMS 401 student) In the example above, the student introduces the report from sources in a neutral way without commenting on what has been reported. The student uses the verb 'say' to pass information to the reader without showing any intention to persuade the reader or to adopt a stance towards the material cited. This was a common trend in how students in both groups incorporated cited material. The students relied on a very restricted range of these verbs; the most common informing verbs used were say, according to, point out and state. The results show that writers in both departments displayed a clear preference for adopting a neutral stance to the material cited.
This finding demonstrates that attribution of content to another source was the dominant function of reporting verbs in the students' papers. This function of informing verbs is the simplest to use and 'does not demand advanced rhetorical skills' compared to argumentative verbs (Mansourizadeh & Ahmad 2011).
However, when students used argumentative verbs, they used them effectively to support their argument and to indicate their stance as writers. Example 2 indicates this particular function of argumentative verbs: Example 2: Media representations of women are full of stereotypes which are inevitable, especially in advertising, entertainment and news industries. … In terms of occupation, media coverage continue to rely on men as experts over women in virtually all fields. … Women in the news are likely to be featured in stories about domestic violence than in stories about professional abilities or expertise (Durkin 1985:54). Durkin (1985) further argues that women in politics are equally sidelined. Political and economic success stories are mostly masculine. (BMS 302 student) In Example 2, the writer has presented the original author's point of view as an argument, and, in so doing, advances their own argument that women are sidelined in politics. This suggests that the verb argue has been imposed on the reported information by the writer rather than the source author and hence the writer's voice intervenes on the reported information.

Discussion
The present study sought to explore the types of reporting verbs students use to refer to the work of others and the extent to which they evaluate the work of others in their writing. It also sought to compare the more experienced writers (Primary Education students) with the less experienced writers (Media Studies students). As shown by the results, whilst the two groups of students used reporting verbs in their writing, they were far more prevalent in the Primary Education corpus than in the Media Studies corpus. We have also seen that informing verbs were by far the most common in the whole corpus.
The results suggest a number of possible explanations for the variation in the use of reporting verbs by the two groups of students. They suggest a tendency by Primary Education students to overly rely on the use of citations in their essays and used sources to present knowledge or arrangement of ideas 'merely to fulfill the "requirements" set to pass' (Read, Francis & Robson 2001:395) rather than to present an argument. The results are consistent with what Barton (1993) and Mansourizadeh and Ahmad (2011) found out when they compared the use and function of citation by experts and novice researchers. They found that novice writers tended to use citation to attribute content to the source rather than to provide support for their arguments and to justify their claims. Petric (2007) raises an important point: that students, unlike scholars, write for a different audience and therefore have different writing goals. In that sense, when students write essays it could be to demonstrate their knowledge on a certain topic. Petric (ibid:239) posits that students' 'use of citation for knowledge display and for demonstrating one's ability to apply a theory may be a more prominent function of citation in students' writing'. This could explain why students used informing verbs more often than argumentative verbs.
Another factor that is discussed by Hyland (2002aHyland ( :1094 is that the 'academy's emphasis on analysis and interpretation means that students must position themselves in relation to the materials they discuss, finding a way to express their own contentions and arguments … to get behind their words and stake out a position'. Students are expected to develop ways of arguing and reporting on other views in a manner that is acceptable in the academy. However, Petric (2007:239) argues that 'students refer to the work of others in much higher standing than them; therefore, issues of power are also different, which may affect students' confidence to express evaluation'. Evaluation can also present problems for undergraduate students because 'they are not at the appropriate level of cognitive or intellectual development to do so' (Thompson & Tribble 2001:92).
Despite the small corpus that was used in the present study, overall, the results corroborate findings of earlier researchers, which reflect the difficulties novice writers have in incorporating sources in their work. The results also indicate that Primary Education students seemed to have more difficulties with incorporating source materials than Media Studies students, which is contrary to expectations. Such difficulties include lack of strategies such as paraphrasing and summarising techniques that they could use to embed ideas they get from reading into their writing. The students tended to use less-demanding strategies like 'cutting and pasting' ideas from the sources into their writing in a rather reproductive way.

Pedagogical implications
As presented earlier, citation is an important convention of, and plays a key role in, academic writing. But we have seen that it also presents considerable difficulties for our students. Hyland (2002b:116) asserts that 'this problem is often more than a deficit of vocabulary, but is symptomatic of a larger issue of how to appropriately acknowledge sources in academic writing'. In the EAP writing course, the use of sources in writing is often taught at a technical level: students are introduced to the mechanics of citations such as how to incorporate direct and indirect quotes and how to paraphrase or summarise information. All these are related to helping students avoid plagiarising. This means that, although students are aware that they needed to cite sources that they use in writing, they might not be aware that reporting verbs can be employed differently to perform different functions. The use of reporting verbs is therefore one area in which the students need help. Hyland (2002b) encapsulates it well when he says: although reporting is often treated in EAP materials as the application of standardized conventions and advice about avoiding plagiarism, students need to know more than the necessity and mechanics of referring to existing literature. They also need to present literature in ways that their readers are likely to find convincing. (p. 130) The findings can therefore inform the design of the EAP writing course in that they support incorporating the teaching of reporting verbs into our EAP advanced writing course.
The results also indicate that students used a limited range of linguistic resources to report the work of other writers and students can benefit from becoming more aware of resources available to them to do so.

Conclusion
This study provided a picture of how students at the University of Botswana used reporting verbs in reporting cited material. The findings reported high use of informing verbs, which are associated with neutral passing of information from source to reader. The students used a limited repertoire of argumentative verbs, which signify the evaluative role. The findings of this study have contributed to research on the use of citations by confirming the difficulties non-native students have as they learn to cite materials in their writing. The findings suggest that the University of Botswana EAP lecturers should assist students to become aware of the various linguistic choices they have when they want to report cited material.

Course: BMS 302 -Gender, Sex and Ethnicity in Media
Choose one of the topics below and write an essay of about 1200-1500 words on the topic: a) Discuss the media treatment of one social group of your choice. What factors or contexts do you feel are relevant for understanding the ways in which this group is, or has been, represented? b) Do media images and messages only reflect the world, or do they also create it? Clearly they do both. But many studies demonstrate that media messages do not reflect the world as it really is. There are far more people of colour, disabled people, non-heterosexuals, seniors and poor people in the real world than we see on TV or in the movies. Media portrayal of women often leaves much to be desired, especially women in advertising. Discuss, with reference to both print and broadcast media. c) "Gender-based restrictions on media representations in Africa constitute a form of neo-colonialism which is contrary to African cultural traditions". Explain, with evidence, whether you agree or disagree with this quotation. d) What is now called "globalisation" would not be possible without mass media. Communication through modern technology has the potential to spread democracy. But corporate control of media production simultaneously threatens the ability of citizens to receive unbiased news to have their own messages heard. Discuss.

Course: BMS 401 -Imaging Africa
Choose one of the topics below and write an essay of about 1200-1500 words on the topic: a) To what extent does the national museum of Botswana conform to stereotypes about Africa and to what extent does it provide an alternative Afro-centric view of Botswana? b) If an ancestral mask or a traditional foot stool is taken from its original context and function (e.g. to an art gallery), is its artistic value enhanced or diminished? c) In what ways do African artists and media practitioners provide a counter narrative to colonial and neo colonial imaging of Africa with respect to disaster?

Course: EPI 442 -Environmental Education Conservation Strategies
Choose one of the topics below and write an essay of about 1200-1500 words on the topic: a) Discuss using elaborate examples how indigenous knowledge can contribute towards sustainable use of the environment and its resources. b) Discuss how gender-related activities contribute to natural resource depletion. In your discussion, indicate how the environmental problems caused by such activities could be solved. c) Using examples in Botswana and elsewhere, explain how local communities can be an instrument in conserving the environment.