April 2008 Volume 4 Issue 2
Article 3.
PDF File
July 2008 home

Article Title
The Foreign Language Engineering Writer:
What Makes a Readable Memo Report?

Author
David F Dalton　
Petroleum Institute, Abu Dhabi

Bio Data
David Dalton is a senior lecturer in communications. He has worked in tertiary institutes in four countries and currently teaches at the Petroleum Institute in Abu Dhabi. His present research interests are the development of critical thinking skills and their application to higher order reading. He believes in student –centered classrooms and that the death of methodology would be no bad thing.

Abstract:
Writing in a foreign language brings its own set of complex difficulties, not least that the structure, notions, and productive skills required by different genres are often not understood by writers writing in their native languages. They have to be taught. The purpose of this exploratory research is to analyze some of the linguistic and organizational elements of a typical degree-level document (memo report) produced in one of the engineering programs at the Petroleum Institute in Abu Dhabi, UAE with a view to describing some of the salient features which contribute to its level of “success” (readability) as judged by the subject teacher grading the document since part of the grade is given for language quality. (Here, teachers made comments related to “accurate grammar” and were able to identify text-based examples which they considered problematic.) 

Texts were analyzed according to quality of noun phrase formation; quality of infinitive clause formation and use in the context of text moves, and lexical range in the employment of the latter. Successful application of these features was a factor in the response of the content teacher to the language quality of the report.
 Data relating to the above will be presented and discussed in relation to how linguistic features can be aligned with the above-noted “success” for the subject lecturer. Recommendations will be made as to how the readability of such documents might be improved by better preparing students to produce them, in particular by developing a better understanding of the relevant genre.

Introduction

Background
The Abu Dhabi National Oil Company (ADNOC) operates in all areas of the oil and gas industry. It was established in 1971 and has steadily widened its endeavors through its subsidiaries, establishing an integrated oil and gas industry in Abu Dhabi. The company manages and oversees oil production of over two million barrels a day and is among the top ten oil and gas companies in the world. It has substantial and impressive upstream and downstream operations in transportation, shipping, marketing, and distribution.
    Five years ago, ADNOC announced the establishment of The Petroleum Institute in Abu Dhabi. The decree establishing the Petroleum Institute provides for an entity that would bring together the best from education and industry to create a world class regional educational center for engineering, applied science, and research. The Institute offers five baccalaureate degree programs: chemical engineering, petroleum engineering, petroleum geosciences engineering, mechanical engineering, and electrical engineering. As the undergraduate programs mature, the PI will eventually evolve into a fully-fledged research and educational institution providing programs leading to Master of Science and Doctor of Philosophy degrees.
   Our students are at the undergraduate age, emirati males and females .English is the medium of instruction as well as the operating language of the oil and gas industry in this region. This implies a range of issues for the students. Writing is perhaps the most complex and difficult (as well as the least liked and most “painfully” acquired) academic skill, particularly for those studying or working in a foreign language. Concern has been expressed in the institute about the quality of some of the written English produced by students. This project is, in part, a response to that concern.
The Exploratory Research
Subject teachers in the degree programs have expressed dissatisfaction with the readability of student writing from a language perspective. In light of the above, an exploratory research project was formulated to develop an analysis and description of some of the dominant linguistic features of one commonly produced text, the memo report. The purpose would then be to allow such description to inform thinking about how to better prepare students to produce effective texts. Implications for teaching and curriculum from this research could then be articulated. This paper describes an exploratory research project focusing on the description and evaluation of a range of memo report texts produced within the Institute. The approach is described in the methodology section. A beginning descriptive model was generated which may be used initially to work with students, teachers, and syllabus designers at the Petroleum Institute in order to improve the quality of student-generated written texts by having a bank of real texts described according to both linguistic and text type criteria, which can be referred to in order to inform curriculum and pedagogical thinking.

Our graduate profile objectives include the following:

• The graduating student will appreciate the critical role played by verbal, written, and graphical communications in engineering practice and project management
• The student will acquire the corresponding skills to communicate with a range of audiences
• The student will acquire the skills to employ information technologies where appropriate.

To meet these objectives, we have a commitment to provide ongoing language and communications support to students in our degree programs. Some of these students inevitably encounter a range of problems with written text. This would be true in a native-speaker environment and becomes more significant where students are studying in a foreign language. Further, within the company (all of our graduates will work for ADNOC), mastery of English language and communications skills constitutes part of the performance review of employees. Production of effective written documentation is therefore significant.        
  Though we are currently in the development stage of this project, eventually, by gathering, categorizing and storing a wide range of functional written documents, we hope to be able to do the following:

• Begin to develop a database of current documents
• Categorize and classify these in terms of text type/genre
• Describe the texts in terms of format, organization, and structure
• Perform discourse analysis for selected field and generic features
• Derive information to inform syllabus priorities, development, and design in order to improve end-product student texts
• Derive information to frame focus for professional development and ways to improve intra-company written communication

This will provide a model which can then be developed further. Given the potential range of stakeholders, the approach cannot be a uniquely linguistic analysis, but rather, needs to have a level of accessibility which will allow inclusion of nonexpert users. It must also have theoretical integrity as well as usefulness in pedagogical and curriculum terms. This is not an easy task and is therefore very much an initial and “first phase” approach.

Literature Review
The field of English for Specific Purposes has grown exponentially from its origins in the 1960’s, as English has increasingly become the lingua franca of international communications. This growth is especially notable in the context of education and commerce. Linguists have shown an understandable interest in analyzing the texts commonly produced in these different environments. From its inception rooted in register (Halliday, 1978), the linguistic analysis of ESP texts has broadened to include syntactic choices, pragmatics, and elements of text organization and structure. Since texts are interactive, with a relationship between writer and audience revolving around the communicative purpose of the writer, chosen genre, and expectations of the reader, (amongst other things), we require a wide range of tools for their analysis. This is particularly true if our description is motivated by the desire to act on this analysis in order to improve things, such as writer awareness, teaching and curriculum, and syllabus design (Celce-Murcia & Olshtain, 2001), as is the case in this paper. The essentially pragmatic nature of EAP (Benesch, 2001) encourages text analysis based on the kind of features this study focuses on.
   Much purely linguistic analysis, while highly informative and contributing to our developed understanding of written discourse, does not easily translate to learners or other “nonexpert” interested parties. Hyland & Hamp-Lyons (2002) have pointed out the problems of believing in a universality of academic conventions, and Clarke (1992) indicated that while students may be required to write in a particular fashion, they are not always clear what this should be and why. The need for pragmatic approaches which include reader expectations is therefore well-established (Hyland, 2002).Since this research hopes to have an “applications” focus, an approach to text description would need to be functional, broad, and accessible.Some approaches are more transferable than others and therefore more potentially useful while working with students and colleagues in the context of professional development. Dudley-Evans’ work on genre analysis (Dudley-Evans, 1994, 2001) is particularly significant in this area and takes a rather pragmatic approach which is also accessible to learners (this is doubly important in our environment since we are dealing with foreign language users, and much of the research looks at native English speakers.) For example, the concept of a “move” (Dudley-Evans, 1994, p. 220) in the context of the discussion section of a report is one which is not difficult for foreign language{sic} students to grasp and permits a functional analysis of their own texts which can be of significant use in a drafting process. This has informed part of the current work.
   The development of an analytical framework for use by a variety of constituents including teachers and students therefore requires the question of linguistic analysis of text to be approached from a functional perspective. For example, nominalization is more functional in a science text than in everyday conversation and therefore the choice of using a nominal group, or noun-phrase, rather than a clause, to express a semantic process is more “natural” to the text. Further, this also allows us description of text as “chunks” of information,which is arguably how learners acquire language as well as informing how writers write. The noun phrase is already a widely researched area of linguistic enquiry. Here it will be described as one of a range of features which impacts and improves readability. Infinitive clauses are less well-researched. However, an interesting 2002 study of instructions, annual reports, and junk mail revealed that “more than one sentence in four contains an infinitive clause” (Myers, 2002, p.59) and argues that this feature has not been taken seriously in writing.  It has been analyzed in this study as it is a stand-out feature of specific sections of the report.
   The academic writing process lends itself to reflection and explicit learning in a way that speaking does not, primarily because it is “a highly self conscious, reflective, non spontaneous activity” (Lewis, 1993, p.101) within which there is a natural opportunity to analyze. A broad functional description of texts, as this study begins to do, may therefore more adequately meet the needs of teachers and students to engage in this reflection and ultimately produce more effective texts.As Halliday has stated: "The value of a theory lies in the use that can be made of it, and I have always considered a theory of language to be essentially consumer oriented" (1985, p. 7). We are therefore following something of a developing tradition in ESP to approach questions of text from the point of view of the end user and to expand this to include a range of stakeholders who have a vested interest in students developing a more complete understanding of how to produce effective written communication.

Methodology
Since our students are foreign language users, we took the simple view that the primary reader of the text (audience) should be the major judge of the success of that text. The students are involved in a degree program, and their immediate clients are their teachers. Further, these readers are a key element of the discourse community to which the text and writer belong. (Dudley-Evans, 1994)
   As we wanted to look at a range of quality in order to have models, we were guided by the grade and perception of the subject teachers as to the notion of “excellent” and “poor.”  This would give us a standard of acceptability in their context and allow us to describe features of the optimum text and suboptimum texts within that context. Initially a range of eleven texts were collected to use as a corpus for the analysis. Five of these were selected for linguistic analysis of noun phrases. A further six texts were used for analysis of infinitive clauses and text moves.
The Communication Department had been requested to become involved in providing language support to the Institute’s degree programs as a result of concerns expressed about the quality of the writing of some of the students. Individual teachers were assigned to specific programs. The researcher worked within the Department of Chemical Engineering. It thus seemed logical to use texts (the genre was a memo report) from that program to build the corpus, and therefore a purposeful sample was selected from one class. The students are in the first year of the degree program and have attained a 500+ in the TOEFL. Our primary concern is not to generate a large quantity of numerical data, but rather to describe qualitative elements of text, and though the sample is not a large one, it is sufficient for some primary conclusions and to provide the basis for further investigation (see McKenna,1997, p.193).
   As many writers have pointed out (McKenna, 1997; Berwick, 2001) but perhaps most notably Quirk, Greenbaum, Leech, & Svartvik (1980): “Scientific writing differs greatly from other styles in having a distinctly higher proportion of noun phrases with complexity” (p.  34). Taking the noun phrase as a driver is therefore an established starting point and allows an analysis and description of pre-and post-modification which would indicate development of the discourse and be accessible as a text feature for description to the stakeholders. The infinitive clause was also identified as a high frequency element of these texts and was therefore selected as a focus for analysis. Criteria for evaluation are described below.

Process
A sample of texts (five) was chosen from a chemical engineering class. These exemplified a range of quality from “poor” to “excellent” as graded by the subject teacher. Part of the grade is awarded on the quality of the language content. The various realizations of noun phrases was marked and enumerated. Complex noun phrases were broken down into their noun phrase components, since counting them singularly and in their entirety would, in a sense, “penalize” the more sophisticated writer.The second feature looked at was the infinitive clause as an analysis of the reports revealed that this was a significant feature of the purpose and process elements of the report.
   Further, analysis is done concerning lexical items. As Halliday (1994, p. 350) has pointed out: “Typically, written language becomes complex by being lexically dense: it packs a large number of lexical items into each clause; whereas spoken language becomes complex by being grammatically intricate…” These lexical items in written text are not the field-specific items, as this range of lexis tends to be high frequency and known. Rather, it is the discourse vocabulary which tends to be problematic. This sub-technical vocabulary (Cowan, 1974;  Nation, 2005; Chung & Nation, 2003, and Mudraya, 2006) is important as it is not field-specific. According to King, “it is often overlooked by teachers and is a cause of problems for students” (King, 1989,  p. 13). Such language can be marked for sectional relevance in different text types. For example, as stated, the infinitive clause is a high frequency feature in the process/method section of the report, and often exemplifies such vocabulary.
   The analysis was therefore mainly driven by isolating some of the syntactical and lexical features which made the text successful according to the reader. It was thus not externally or extraneously imposed, but derived from levels of acceptable products within a specific context. This provided us with a simple model which could then be applied to texts from other engineering courses to see if it held true. If this was found to be the case, then it may be further generalized to other areas.
   As stated above, the analysis had to be more than simply linguistic. A description of the “culture” of the collated texts was therefore agreed on through analysis of the documents and discussion with content teachers as to their expectations of how the reports should look.

By this is meant the:

• format and conventions
• structure
• organization

   Each section of each text was then described in terms of expected content load (information) and purpose. A general description was given of dominant linguistic features. Further, description was given in the case of infinitive clauses as to how these function in the context of moves within the purpose and method section of the reports. (See page 3 and discussion). Such a description could be carried out with a range of linguistic features.This would result in a model encompassing a comprehensive description of texts from the different perspectives described, which is thought to be practical given the stated purpose of the research project and the potential stakeholders.
As part of this, a template for text description was designed, which is displayed in Appendix 1.
   Also, a writer profile was created. This is significant in our environment as we have students entering from quite varied backgrounds. Some are direct entry, and others have come through various stages of our foundation program. All have achieved a 500+ on the TEOFL. Such a profile can be used in making curriculum and administrative recommendations, such as initiatives relating to grouping students for specific language focus work. It would also allow us to track changes in student writing over time and thereby give valuable institutional research information and which could inform syllabus changes.  A possible template for this is outlined in Appendix 3.
   The documents, data and coding were stored in a Lotus Domino database (the system is adequate for the purposes of this research;  however, more detailed linguistic analysis of text would indicate the use of a different program, such as N6 or Text Quest.)  Documents relate to the provision of a model and partial answer to the question what makes an “excellent,”  “very good,”  “good,” or “average/poor” student memo report text in the PI context.  (Grouped together here as texts, being awarded 60 % D is considered a failure in some courses, but acceptable in others.)   As stated in the methodology section, the measure here was the response of the subject teacher to the language (readability) of the document (comments on the involvement of language teachers will be made in the discussion.)  The identification and description of patterns and features which answer this question would reasonably allow careful generalization of the approach to other documents.
   Subject teachers may not always be able to specify what they mean by readability in relation to linguistic features, but in informal discussion, three of the teachers teaching the Chemical Engineering program were able to affirm that the text elements analyzed were important for “easy” reading of the documents. The issue of different response to the same text from subject and language teachers has been addressed in an interesting study by Napierkowski (2003) but is not part of the scope of this paper.

Criteria
The criteria described below were derived from discussion with subject teachers as signficant factors in readability. Communication teachers agreed that these features are key drivers of meaning in these text types. Broadly, it is considered a reasonable starting point that a text which is effective in a foreign language educational environment should have a number of characteristics which promote effective communication of the message(s) to the subject-informed reader. These text characteristics include: quantity and quality, which in turn includes accuracy and appropriateness. Quantity is related to length of text defined by simple word count. In the case of memo texts reviewed for this study, preferred length from the point of view of the content teacher was between 800 and 1000 words. As White has pointed out, this relates to how much information the reader expects (White,2001, p. 68).
   Accuracy in the context of quality relates to the number of words in error-free text units. (Robb, Ross, & Shortreed 1986). Appropriateness here relates to the employment of infinitive clauses in the expected text section. Further description is given below.

Quantity
Word length guidelines are a common feature of academic writing in general and specifically useful for those writing in a foreign language.
   For a text demonstrating understanding of a concept to be effective, there clearly has to be a sufficient amount of text. This is especially the case in the foreign language context, where a very limited text might strongly suggest the writer has poor communicative competence for whatever reasons.

Quality
Quantity alone, however, is obviously no guide to the effectiveness of a text. An effective text, something which not only conveys the intended message(s), but which is also easy to read, will have an appropriate variety of structures and lexis, and forms of expression. There should not be, for example, repetition of a few formulaic phrases but rather, a variety of forms (syntactical/lexical) which aid the reading experience and hence the effectiveness of expression should be in the evidence.  Such features are a given as evidenced by their appearance as assessment criteria in writing rubrics in tertiary institutes internationally (Stevens & Levi, 2004; Walvrood 1998)

Accuracy
One aspect of quality is the structural accuracy of the items used (Polio, 1997). This is, again, vital in the foreign language context, as, although a number of infelicities can be tolerated and not interfere with the meaning(s)/message(s) intended through the text, there comes a point where the reader's effort is so great that meaning and message are lost, or at least difficult to discern. Hence, accurate use is employed as a feature of readability (see Criteria above).

Appropriateness
That there are salient features in different genres of text would not be challenged, and there are now many publications aimed at foreign language text book writers, teachers, and students to assist in developing awareness of this ( Carter et. al, 2000; Hughes, 2005).The appropriateness, or relevance of, the structures, lexis, and forms of expression used are clearly germane to this. Certain specific elements of grammar and vocabulary are “natural” to particular sections of the report which has influenced the focus here on infinitive clauses (Thomson, 1996). Since the most common uses of the infinitive are to indicate the purpose or intention of an action, in technical writing higher frequency would be expected (appropriate) in the objectives and purpose sections of a report (see p.13) or within certain contexts such as procedure explanation (experiment).
   For each of the three broad categories of “good,”“average,” and “poor,” the purpose of the model is to demonstrate some of the linguistic features which lead to the application of these evaluative communication categories and to give a part description of functionality by a focus on text moves (Swales, 1990; Dudley-Evans 1994).

Features
The linguistic features used are noun phrases and infinitive (purpose) phrases. There is a huge body of research on the former, and a significant recent doctoral study by Godby clearly demonstrates the significance of noun phrases in the specific context of engineering writing (Godby, 2002).
As stated in the literature review, infinitive clauses are less well researched. They are however, high frequency in certain contexts (Myers, 2002, p.61). Infinitives were chosen here because (as previously stated) they are stand-out feature of the process section of the successful document.
   De Haan (1992) has pointed out that for infrequent structures to be effectively tracked, samples would need to be large. This is presumably to do with more extended pattern identification. This was not a concern in our case as structures were observably frequent in the small sample range selected. A large corpus was not therefore necessary (see p.4).

Units of Measurement
Noun phrases
Noun phrases have been divided into those which have post-modification (complex) and those which do not (simple). The former include any noun phrase which has some form of post-modification postpositive adjective, prepositional phrase, or other phrase. The former include those phrases which end at the head, irrespective of the nature of the pre-modification. Within this classification, any conjunct noun phrases none of whose heads are post-modified are also classified as simple. Finally, noun phrases occurring inside other noun phrases are not counted separately but are included within the largest (complex) noun phrase.
   Noun phrases are marked up and counted principally for two features: the (non)existence of post-modification and number of words in the phrase.

The first feature has four possible values:

np = simple noun phrase with no syntactic errors
npx = simple noun phrase with at least one syntactic error
nppm = complex noun phrase with no syntactic errors
nppmx = complex noun phrase with at least one syntactic error*
*Intra-phrase grammatical errors as identified by the researchers.

The second feature is indicated thus:
np_? = simple noun phrase where? indicates the number of words in the phrase
np__? = complex noun phrase where? indicates the number of words in the   phrase
   A final feature, which may be extended in future work, is a minimal marking of functionality for pronominal noun phrases. The incidence of these is very small, understandable in text from the kinds of writers in the PI, but a small number of usages of “it” and “this” do occur. (Further work in this area might include a breakdown of the kinds on noun phrases in the texts into the various clausal elements: subject, object, etc.).
Infinitive clauses
The same principle of inclusion applies to infinitive clauses. These phrases were further analyzed according to number and variety of verb head.

Findings
Data
Our data are as shown in Figures 1 to 7 below.
Noun phrases
The basis for noun phrase analysis includes total words in a text, the number of words in noun phrases, percentage of such, the numbers of noun phrases (total, simple, and complex),  and the number and percentages of well-formed (structurally accurate) total, simple, and complex noun phrases. They are shown in Figure 1.
Chemical engineering texts

Figure 1: Noun Phrase Data

Infinitive clauses
For these clauses, again the total number of words, the number of such clauses, the number of words in these clauses, the percentage of this, and the numerical range (variety) of the verb heading each clause were analyzed. This is shown in Figure 2 below.

Figure 2: Infinitive (Purpose) Clause Data

 *Of the initial batch of twenty eight base texts from which the five were selected, 42 % were in this category, thus two examples were selected.

Lexical Verbs in Infinitive Clauses

Figure 3: Lexical Verbs in Texts
Text 1: Excellent

Influence on readability was indicated by variety, appropriateness,
and frequency of use. Students are encouraged to use a variety
of lexis in a meaningful way (see Process Section).
as this is one of the factors which influence the nature of reader
response (+/-) to any given text.

Text 2: Average/Poor

A high frequency of successfully employed infinitive clauses in the sections of the report indicated in Figure 3 improved readability for these sections and therefore contributed to text quality overall (see discussion).

Figure 4: Electrical Engineering Data

*Report sections were the same as in Figure 3.

Figure 5: Text Moves

*The accuracy of the content information is not commented on here.

As can be seen from Figure5 above, the infinitive clause is apparent as a driver of moves within sections of the report as displayed. This lends a further level of description to the data presented in Figure 3.
   The trend further held true for reservoir fluid lab reports in petroleum engineering where sections of a poor and excellent text were analyzed. Text moves section results are presented below and commented on in the discussion.

Figure 6:  Text Moves Petroleum Engineering Lab Report
Text 1: Poor

Figure 7:  Text Moves Petroleum Engineering Lab Report
Text 2: Excellent

The full text, is displayed in Appendix 2.

Discussion
Returning briefly to the concept of an excellent, good, and average/poor text, it may be that clients other than the subject teachers need to be included in discussion of these descriptions. Language teachers are an obvious agent here as they “often undergo long term socialization and enculturation within the discourse community and writing assessment is central to their professional day-to-day lives.” (Napierkowski, 2003, p. 6). Of the significance of the items identified for analysis and the approach, we would claim the following.
   The “excellent” text was noticeably longer than its counterparts (899 words). Even though these are memo reports, they cannot be successful if text length is compromised. Features such as variety and frequency are clearly undermined by a reduced text. There was a significantly higher percentage (almost double) of words in noun phrases between the “excellent” and “very good” text. Well- formed simple and complex noun phrases were above 98% in the former, falling to 88% in the latter in a considerably shorter text.
   An example containing no syntactic (grammar) errors from this text is given below in Figure 8. The quality is typical of the text.

The “poor” text has the fewest number of words (339) and is therefore likelier to contain a higher proportion of noun phrases of some kind: forty-three simple noun phrases in this case. The proportion of complex to simple noun phrases as shown in Figure 1 is higher at the top end (81/32) and lower at the bottom (43/10).
A typical example with error from this text is given below.

   The general trend across the five texts is almost consistently downward from the “optimum” (excellent) text. However, text length alone is obviously not a definitive marker of a well-written text. The more accomplished writer who produced the “excellent” text wrote more, but also had a greater number of words in noun phrases and a greater number of such phrases, both complex and simple. More importantly, these phrases were substantially better-formed overall. Simply, his language competence and proficiency on this evidence is higher, and his text evidences the feature of readability resulting in a more positive response from the subject teacher.
    For the second feature of infinitive clauses, the results were slightly idiosyncratic with a “very good” text having a smaller word count, fewer Tot-Inf-Cl (Total Infinitive Clauses)and less lexical range than a “good” text. One issue here is that two grades are given which results in the generic grade; one for content and one for language. This particular text received the maximum grade for content but a lower grade than the “good” texts for language. This was the only case where the anomaly occurred, but it did slightly distort the results. It is notable, though, that the readability of this text was viewed less positively, and therefore the linguistic analysis does in fact hold. Further research should avoid the problem identified above by considering only the language grade and further exploring the responses of the subject and language teacher to the same text (Napierkowski, 2003, p. 12). The relationship between content and language grades would clearly be a useful focus for further research.
   Differences between the top and bottom end of the scale are certainly revealing. Both the writer of the ‘excellent’ text and the writer of the “average/poor” text demonstrated a high level of appropriateness by concentrating use within the introduction, and method/stages sections of the document-traditionally where a high frequency of this feature would be predicted given the “proposal” and
“report” (purpose) nature of these elements of text. (Thomson, 1996, p. 503).
    Where they differed substantially was in the range of verbs utilized and the effectiveness of their use (quality, accuracy, and appropriateness). A wider variety of appropriate items led to a more positive perception of the readability of the better text.It made for more interesting and persuasive reading. A sample from the introduction is given below.

    The weaker text was limited to five items with one of these used three times (see Findings). The lexical range of the IC verbs in text 1 (excellent) indicates a lexical flexibility from the writer which was not present in text 2.  This greater variety helped give the text greater readability. It engaged the reader more. Reader expectations are not only about what information we expect to see, but on a more subconscious level, being presented with the language we expect which has been chosen in order to communicate that content. (see Appendix 2)
   The two language teachers who read these texts specifically mentioned how the choice of appropriate (expected) vocabulary and grammar in the relevant sections of the report contributed to ease of readability especially with reference to the variety and effectiveness of lexical verbs in infinitive clauses in the purpose and method section of the “excellent” text in Figure 3 (criteria of quality, appropriateness, and accuracy identified in the “concept” section). Such “sectional” analysis and description of documents might therefore be extended and seen as valuable from teaching and learning perspectives. One of the issues for our students ( and perhaps generally for students writing in a second or foreign language) is the absence of an understanding that appropriate choices of grammar are not simply about “correct structure,”  but also effective style and communication and meeting audience expectation. (Connor, 1996).
   When the approach was then tested with “good” and “average” lab reports from the electrical engineering programme, a similar profile emerged (see Figure 3). In the introduction and approach sections of the reports, an “average” text contained 244 words with seven infinitive clauses. A “good” text contained 345 words for the same sections with twelve infinitive clauses. The former displayed much less lexical variety and frequency in the infinitive verbs with “test” being used four times. Variety in the better text was much more noticeable. Accuracy of application was the same in both cases (100%). This is perhaps to be expected given relative grammatical and structural complexity.
   The pattern further held when applied to petroleum engineering lab reports where a similar description can be made, indicating that the approach could be generalized (see Figures 6 and 7). It is significant that the feature appears with high frequency as a driver of text moves as indicated in Figures 5, 6, and 7. Further, it was applied appropriately by writers of different quality texts to make similar text moves. The main difference between these texts was one of lexical variety, frequency, and range in the verbs employed (quality) and the quantity of text produced (see Criteria).
   The two features chosen are significant for analysis of academic style memo/report writing. Noun phrases are perhaps more pervasive and generalizable, but infinitive clauses are also germane in our text area. Further analysis of noun phrases might focus on their functions within clauses and sentences in relation to their complexity, the various and varying natures of post-modifying elements, coordinated phrases, and so forth. In addition, other types of adverbial phrases could be considered such as those indicating sequence/process (temporal or logical), hypotheticals, and concessives.
   Nevertheless, we have two important features pointing in the same direction.

Conclusion
While accepting that the above does not constitute an exhaustive analysis and is exploratory in nature, it has shown itself to be useful as the foundation of a model of text description which could be extended and generalized to other texts across a range of academic programs in our institute with a view to influencing preparation of students to write such documents more effectively.  Clearly there were differences in terms of the language proficiency of the writers, which is influenced by factors other than those analyzed. However, sensitizing weaker writers to the contextual nature of elements of text discourse such as those looked at here should result in a better report and improve readability. In turn this should elicit a more positive response from the subject teacher and result in a better grade. It would also (perhaps more importantly) improve the writer’s overall language awareness.
    We have been able to describe the texts in terms of significant features which enhance readability as indicated by the assessment of the texts by chemical engineering teachers. It is felt there is enough evidence here to claim that such a multifaceted description of texts would be useful in illustrating and documenting text types and using such information to inform teaching and curriculum development. As stated, anecdotal evidence from teachers in PI degree programs has indicated a problem with regard to the quality of writing. Analysis of a range of texts has highlighted some of the linguistic issues.  The extent of the problem (e.g., numbers of students, level of proficiency, and so on) still needs to be clarified as it is not within the scope of this investigation.

Recommendations
It is important to emphasize that tertiary education has a relatively short history in the UAE, but much of great importance has been achieved in that time. Responses to second and foreign language learning issues have had much longer to develop in other countries. The following observations should be seen in that context.
    The need for ongoing language support and development in degree programs is clear. In the quest for better student-produced texts, cross-curricular teacher collaboration is a significant element. There are various ways this can be explored and many models of good practice. The three-phase approach of collaboration, cooperation, and team teaching described by Dudley-Evans (2001) is a very practical example here. A synthesis of some of these elements (specifically collaboration and cooperation) has informed the teaching relationship between the Communication Department and the Department of Chemical Engineering in our institute leading to the development of a parallel syllabus (see Appendix 4). In the opinion of the department head, the quality of student documentation at the end of our first semester of working together was much improved. Attitudinal shifts are important here. As Sharpling states (2002,  p.87): “Collaborative and cooperative knowledge cannot in itself be gained from a fixed-term training course alone. Rather, it is dispositional, involving a genuine desire to investigate the conceptual and discourse framework of the subject students are studying, leading to a more systematic attempt to find out how a discipline works”. Further, content teachers in foreign language medium institutes should be encouraged to develop an interest in the interface between language and content. Our most successful ventures (for learners) have occurred with content teachers who understand and are interested in this relationship.
    On the text level, comprehensive description of the culture of each document should be given to students in order for them to have a detailed map before they begin. A possible format/template for this is presented in Appendix 1. The intention is to give a description of the purpose and structure of the document and guidance as to the linguistic and discourse elements appropriate to different sections of the text, such as those looked at in this study. The student then has a schema before embarking on the writing process. This can be utilized in consultation with instructors in the preparation of documents.
    Further, it is important to encourage students to take a functional view of why particular language choices are significant in terms of meeting reader expectations and how the language works in order to do that. (Clarke, 1992). Description of documents from the perspective of text moves is a very accessible and practical approach here. The features described in this study could easily be discussed in this context. Too often grammar is presented as simply structural and therefore divorced from meaning and purpose. Learners require a more functional understanding in order to select relevant structures and vocabulary for their purpose. Given reasonable concerns about notions of universality in the context of academic discourse, (Hyland & Hamp-Lyons, 2002, Hyland & Tse, 2007) it is also worth sensitizing students to notions of range, frequency, (Mudraya, 2006) collocation and meaning within the context of specific disciplines and genres. Teachers may work on helping students develop a more focused lexical repertoire derived from their particular discipline, as suggested by these authors. This again reinforces the reflective and pragmatic nature of ESP.
    The concept of ownership is also significant. Students need to be encouraged to effectively proof read and edit documents. The skills involved are complex and developed over time.  It is often (wrongly) assumed that this is something which students know how to do rather than being a specific suite of abilities which need to be taught and valued as an important element in the set of skills utilized in the production of professional documentation.

References
Benesch, S. (2001). Critical EAP:  Theory, politics and practice. New Jersey: Lawrence Erblaum.

Berwick, R, C. (2001).Language: From words to meaning. Recitation given at MIT. Oct 26 2001. Retrieved Oct 12, 2006 from:
http://www.ai.mit.edu/projects/berwick/recitation7.pdf

Carter, R., Hughes, R., & McCathy, M.(2000). Exploring grammar in context. Cambridge: Cambridge University Press. UK.

Celce-Murcia, M & Olshtain, E. (2001). Discourse and content in language teaching. Cambridge: Cambridge University Press.

Chung, T. & Nation, P. (2003). Teaching vocabulary in specialized texts. Reading in a foreign language, 15(2), 103-116.

Clarke, R. (1992). Principles and practice of critical language awareness in the classroom.In Fairclough, N. (Ed).Critical language awareness (pp 117-140).

Connor, U. (1996) Contrastive rhetoric: Cross-cultural aspects of second-language writing. New York: Cambridge University Press.

Cowan, J. R. (1974). Lexical and syntactic research for the design of EFL reading materials. TESOL Quarterly, 8(4), 389-400.

De Haan, P. (1992). The optimum corpus sample size? In G. Leitner (Ed.), New directions in English language corpora, (pp 3-19). Berlin: Mouton de Gruyter.

Dudley-Evans, T. (1994) Genre analysis: An approach to text analysis for ESP. In M. Coulthard (Ed.) Advances in written text analysis. (pp 219-228). London: Routledge.

Dudley-Evans, T. (2001). Team-teaching in EAP: Changes and adaptations in the Birmingham Approach. In J. Flowerdew & M. Peacock (Eds.), Research perspectives on English for academic purposes (pp 225-238). Cambridge: Cambridge University Press.

Godby, C.J (2002) A computational study of lexicalized noun phrases in English.
Doctoral dissertation, Ohio State University. USA

Halliday, M. A. K. (1978).Language as social semiotic. London: Edward Arnold.

Halliday, M. A. K. (1985). Systemic background. In J. D. Benson, & W. S. Greaves, (Eds.) Systemic Perspectives on Discourse, Volume 1. Selected Theoretical Papers from the Ninth International Systemic Workshop, 1- 15. Norwood, NJ: Ablex Publishing Corporation.

Halliday, M. A. K. (1994).An introduction to functional grammar. (2nd ed.) London: Arnold.

Hughes, R. (2005) Exploring grammar in writing. Cambridge: Cambridge University Press.

Hyland, K. (2002). Teaching and researching writing. London: Longman.

Hyland, K. & Hamp-Lyons, L. (2002). EAP: issues and directions. Journal of English for Academic Purposes 1(1), 1-12.

Hyland, K. & Tse, P. (2007). Is there an “academic vocabulary”? TESOL Quarterly, 41(2), 235-253.

King, P. (1989). The uncommon core. : Some discourse features of student writing. System, 17, 13-20.

Lewis, M. (1993). The lexical approach. London. LTP

McKenna, B. (1997) How engineers write: an empirical study of engineering report writing. Applied Linguistics, 18, (2), 189-211. England: Oxford University Press.

Mudraya,O. (2006) Engineering English: A lexical frequency instructional model. TESOL Quarterly 8(4), 389-400.

Myers, M. (2002). A new look at infinitives in business and technical writing .Journal of Technical Writing and Communication.32, (1), 59-65.

Napierkowski, H. (2003) Addressing instructional documents: A comparison of writing faculty, engineering faculty and workplace technical communication practices. Paper presented at the Annual Meeting on College Composition and Communication. New York, Mar 19 -22, 2003.

Nation, P. (2005) Second language vocabulary. In Braun, K (Ed.) Encyclopedia of language and linguistics, 2nd edition, 13, (pp 448-454).

Polio. G.C (1997). Measures of linguistic accuracy in second language writing research. Language Learning, 47(1), 101-43. 

Quirk R., Greenbaum G., Leech G. & Svartvik, J. (1980). A grammar of contemporary English. London: Longman.

Robb, T., Ross, S., & Shortreed, I. (1986). Salience of feedback on error and its effect writing quality. TESOL Quarterly, 20(1), 83-93.

Sharpling P. (2002) Learning to teach English for academic purposes: Some current training and development issues. English Language Teacher Education and Development, 6(1), (p. 82 – 94).

Stevens.D. & Levi, A. (2004) An introduction to rubrics: An Assessment tool to save grading time, convey effective feedback and promote student learning. USA: Stylus Publishing.

Swales, J. (1990) Genre analysis:  English in academic and research settings. Cambridge: Cambridge University Press

Thomson, G. (1996) Voices in the text: Discourse perspective on language reports. Applied Linguistics, 17, (4), 501-530.

Walvrood, B.E. & Anderson, V.J. (1998) Effective grading: A tool for learning and assessment.  San Francisco: Jossey-Bass.

White, R. (2001) Adapting Grice’s maxims in the teaching of writing.ELT Journal 55, (1), 62-69.