The Preparatory Year English courses conform to the university’s American-style semester credit system: there are two semesters (English 001 and 002), each of which involves 20 class hours a week for 16 weeks. The courses are based on and fairly strictly limited to the course books, which have evolved in-house over the years as a result of input from most of the university’s English teachers, all native English speakers experienced and qualified in EFL or ESL and predominantly British or American. The core of the courses is the Reading component, whose course books present two thematically related short readings each week and materials intended to exploit them for various reading skills. The other components (Writing, Listening, Grammar, Vocabulary and Oral) are, to varying degrees, related thematically and linguistically to this core. The significance of this for the present study is that all the students have been exposed fairly extensively to a common core of vocabulary items related not only to the topics in the Reading course books but also to the rhetorical modes practised in the Writing course books. This vocabulary may be identified in the written data.

Up to 20% of students enrolling in the Preparatory Year do not embark on freshman courses; a few of these drop all courses for personal (usually family) reasons, but most leave the university with failing or insufficiently high grades, or make a tactical withdrawal in anticipation of such grades. The remaining 80%, having proved, among other things, their ability and English language proficiency, enrol in English 101, a one semester course consisting of three class hours a week for 16 weeks. These students are the subject of this study. It is normal to take this course in the first (in American nomenclature fall) semester of the academic year. However, there are many students who enrol in the second (spring) semester, either because they are on the ‘fast track’, having been exempted from the first Preparatory Year English course, or because they are on the ‘slow track’, having had to repeat one of the Preparatory Year courses. Thus, the second semester includes a more divergent sample of good and poor students.

There are three major advantages in choosing these students for a study of misspellings. First, they have a relatively homogenous and definable history of EFL learning, especially in terms of vocabulary. Second, they are all taking the same course and taking the same (within each semester) or similar (between semesters) examinations, which provide standardized data. Finally, and most importantly, the students have developed the ability to produce suitable scripts. The level of handwriting proficiency allows nearly all misspellings to be identified as such and the level of language ability allows an interpretation of what words the misspellings are in fact misspellings of. Also, the frequency of misspellings is low enough to allow a fairly large sample of students to be studied without being completely overwhelmed by the total number of misspellings.

The data for the present study was provided by students in the two 101 courses in the academic year 1989-90. This data can be divided into two types: objective grades from multiple choice examinations, and scripts, with assigned grades, from writing examinations. While the objective grades offer a population-wide standardized measure of relative student achievement, the scripts and misspellings are from those 123 students from the total population who were in my classes during the year, and they constitute the core data.
 

Measures of student achievement

To an even greater extent than with the Preparatory Year English courses, the in-house produced 101 course books define the curriculum and are strictly reflected in the examinations. Although the official title of the course is ‘English Composition 1’, the two course books are entitled Academic Reading and Composition 1 and 2, and there are both composition and reading examinations. As far as reading is concerned, the first course book, which leads to the midterm examination, concentrates on reading freshman level textbooks while the second, which leads to the final examination, is preparation for reading journal articles.

Both the midterm and final objective examinations contain three sections: reading, grammar, and dictionary. The grammar and dictionary sections, which constitute under half of the questions, are similar in both examinations. The grammar questions are of two types: multiple choice answers to complete the blank in a sentence or sentences, and the choice of the incorrect underlined part of a sentence or sentences. The dictionary questions test the ability to use various features of entries in The American Heritage Dictionary, which is a course book. Students must use entries reprinted on the examination question paper to find, for example, the correct multiple entry or numbered definition for a word in a sentence, a suitable phrasal verb, idiom or synonym, and information from biographical and geographical entries. Proficiency and efficiency in reading are a sine qua non for answering questions in the grammar and dictionary sections, and success can legitimately be considered an indirect reflection of the students’ reading ability.

Whereas the grammar and dictionary sections are similar in the midterm and final examinations, the reading section in the two examinations differs. For the midterm examination, students must answer questions on the contents pages, glossary, index and a chapter reprinted in booklet format from a previously unseen freshman level textbook. These questions test reading proficiency more directly than the grammar and vocabulary sections. The skills required range from skimming and scanning to intensive reading, and word recognition is important to varying degrees in all these skills. Since the time allowed for the examination is limited, efficient word recognition is most directly rewarded in answering questions on the contents pages and index, and to a lesser extent the glossary questions, which involve some intensive reading. The questions on the chapter also require word recognition firstly in skimming for headings and key words to locate the section of the chapter where the answer may be found and secondly in scanning text as an efficientpreliminarstep to intensive reading.

The reading section of the final objective examination differs from that of the midterm examination in two significant ways. First, the questions are based on the cover, contents page and a half to one page article from a special interest magazine such as Middle East Education & Training or Korea Newsreview. Second, the students have a copy of the article for about a week before the examination to prepare as they will. Therefore, there is a much lower premium on efficient word recognition than in the midterm examination since the text is not only vastly shorter but also quite familiar to the students. However, between a quarter and a third of the questions are based on the cover and contents page, which are previously unseen, and here again, with limited time for the examination, efficient word recognition is vital.

For reasons of examination security, we are obliged to prepare a new objective examination for each session. There is little attempt made to standardize the level of difficulty between examinations beyond a general intuition about what our students have been and will be able to do. In fact, the mean raw scores for the total population in the four examinations under consideration range from 54.09% to 59.44%, a respectable, if felicitous, difference of only 4.35 percentage points. Nevertheless, the raw scores are converted to produce a spread of grades A, B, C, D, and F, which are the standard grades throughout the American educational system, grade A being the highest and grade F a failing grade. The number of students achieving these grades is in a roughly 10:20:40:20:10 proportion. Not only does this address the problem of possible different levels of difficulty between examinations but it also gives students a fighting chance to pass the course, since anything below 60% is defined as a failing grade in all undergraduate English courses.

The total number of questions in the four objective examinations ranges from 35 to 40. The proportion of questions of different types, for example skimming, intensive reading, grammar, dictionary, also varies. Therefore, the examinations may be said to reflect reading proficiency to varying degrees. The second semester saw the introduction of a new component in the course. This was based on a list of word prefixes and word roots, and their meanings, which students were given and told to learn. The examination questions for this component asked the meanings of words, such as revoke, which were presumably unknown to the students. These questions are only minimally related to reading skills, and the mean raw score of 45% suggests that finding the correct answer depends principally on luck. Exactly a quarter of the examination is devoted to questions on word prefixes and roots.

The results of the objective examinations, and particularly the grades resulting from raw score conversion, can be used as a measure of relative student proficiency in reading. In addition, they can be expected to reflect to some extent the students’ proficiency in word recognition.
 

Core data: the writing tasks

The writing component of English 101 centres on rhetorical modes such as identification, definition, classification, exemplification, process description, and comparison and contrast, most of which have already been extensively studied, practiced and examined in the Preparatory Year. However, whereas the Preparatory Year uses information in tables or illustrations as a stimulus for writing, English 101 focuses on texts, from which information must be extracted and whose words are hopefully paraphrased rather than plagiarized. The use of texts is intended as an introduction to English 102, entitled Researching and Reporting, during which students must find articles on which to base a short research report.

A composition examination is taken at the same sitting and immediately after the objective examination. As with the reading examination, the midterm and final composition examinations differ in important respects. In the midterm examination, that section of the textbook chapter for which there were intensive reading questions is the basis of answers to a composition question. On the other hand, the final composition examination is based on an article from a special interest magazine, but not on the same article used for the reading questions. The students are given three articles about a week before the examination; one of these is used for the reading questions, one for the composition, and one is not used. Which article will be used for which examination is not disclosed to the students and they must prepare all three articles as if for either of the examinations. Consequently, the text used for the midterm composition examination is unseen, apart from any time spent referring to it during the immediately preceding objective examination, whereas the text for the final composition examination has been prepared for a week.

There are two further differences between the midterm and final composition examinations: the nature of the texts and the examination questions. The textbook extract for the midterm examination is written in academic expository style, and is therefore organizationally, lexically and syntactically familiar to our students, who have studied similar texts extensively and intensively, not only throughout the Preparatory Year English courses but also during their freshman courses in, for example, physics, chemistry and calculus.

On the other hand, the articles chosen for the final examination display, to varying degrees, features of a journalistic style. This involves a non-linear organization of ideas, an eclectic range of lexis and tortuous syntax, all of which cause problems for our students. The articles are written, apparently under pressure of time and the indiscriminate hand of the sub-editor, to incorporate a variety of facts and quotations into a current news item which will entertain to at least as great an extent as it will inform. The following paragraph, from the article for the second semester final examination, exhibits various features that must cause difficulty for our students.
 

At the level of discourse, the topic of this paragraph is signalled by “However” to contrast sentences (1) and (2) with the rest of the paragraph. However, this is obscured by vocabulary such as “Gulf regionals and Gulf expatriates”, which apparently contrasts with “other Western or European vacationers”. In addition, it is far from clear what “the objectives and social implications” are or that they are obliquely referred to in the following sentences. Also, it is not immediately apparent that sentence (6) is the “One critical factor” of sentence (5). Other problems could be caused by, for example, “only” in sentence (4), which should surely read “only a portion” rather than “only suitable”. In sentence (6), it is difficult to imagine the purpose of “genuinely” in “genuinely traditional society”; it is presumably an elegant variation of “very”. In the same sentence, one can imagine students interpreting the phrase “its customs and mores” as “its customs and many other things”, ‘mores’ being a plural form of ‘more’.

It is not only the types of text and the circumstances under which the students receive them that differ between the midterm and final examinations. The questions themselves reflect the different types of text. The midterm examination question refers explicitly to the rhetorical functions studied during the first half of the semester, and it can generally be answered satisfactorily with almost exclusive reference to information, and particularly vocabulary, set out in the textbook chapter. The first semester midterm composition examination exemplifies this most clearly:
 

This question closely follows the information, and the order of that information, in the textbook chapter. The second semester midterm composition examination is not so clear cut:
 

While the first two parts can be answered directly from the text, the last part involves a certain amount of imagination.

The final composition examination question calls for a much greater use of the students’ own ideas and vocabulary. However, it is usually based on the rhetorical functions studied in the second half of the semester: either comparison and contrast, or cause and effect. The first semester final composition examination is based on an article about science students at Kuwait University and the difficulties they face, one of which is the use of English as the medium of instruction:
 

The ‘relevant parts’ of the text are few and the significant points are referred to obliquely. The same is true of the second semester final composition examination:
 

In fact, the article for this examination provides more relevant points than the first semester final text, but they are referred to even more obliquely.

The composition examination scripts are assigned grades in the following way. A standardization committee of three teachers selects scripts to represent each of the five grades on the basis of four explicit criteria: content, organization, language use and mechanics. In other words, four questions are asked. First, is the question answered explicitly, relevantly and sufficiently completely? Second, is the composition organized logically and explicitly both at the paragraph and the inter- and intrasentential levels? Third, is the language grammatical? Finally, does the composition follow the accepted norms of layout, punctuation, handwriting and spelling? In theory, these four criteria are supposed to have equal weight, and are supposed to produce a standard between semesters. In fact, there is a fifth, implicit criterion: each examination will produce a spread of grades which includes the highest and the lowest, and the mean will be grade C, which is ‘average’ by definition. Individual scripts are compared with the standardized scripts and graded twice, once by the student’s teacher and once by another teacher. Any difference in the grade awarded to a script by the two teachers is resolved by reference again to the standard scripts. Therefore, the grades for the composition examinations can be used as a general measure of relative student achievement in the writing tasks outlined above.

To sum up, the data for the present study consists of midterm and final objective and composition examination grades, and scripts from the composition examinations written by 123 English 101 students. This data will be analysed quantitatively, and then the misspellings will be further analysed.