The following is a review of the current  assessment system functions in Saudi Arabia at the two main levels of study: (1) state schools (2) university level.

 The school assessment system

The general educational system, curricula and textbooks in Saudi Arabia are the same all over the country. Although the private schools may offer more academic subjects and have better facilities, the input does not really differ much from that in the state schools because private schools are required to teach the same academic subjects, using the same textbooks that are used in the state schools. Thus, it can be said that the participants in this research had a similar schooling history.

Within the school system, there is no grouping by ability at any stage for any academic subject. However, progression is not strictly by age, as in the UK, each year to the higher class, since children can be held back for a year (or more) to repeat the same class if they do not do well enough in the exam. In the Saudi twelve-year schooling system, English is not taught in the six years of elementary school.  It starts in grade one in middle school, at the age of twelve. Each academic subject, including English, is taught over an academic year, which is divided into two terms and each term ends with an achievement examination. Both the end of first term and the end of year examination are identical in weight, length and total scores. The final assessment of a subject depends on the total marks obtained over the two terms. Students’ participation in class, homework and extra activities, is assessed as a component of their term and end of year grades. This pattern of assessment runs through the school curriculum in all subjects taught. However, 80% of the assessment is based upon examinations. The total percentage score consists of the following:

* 10%= mid-term quizzes; 6% Homework; 2% Participation and 2% Activities (own initiative)

 The general education assessment scheme in Saudi Arabia

The contents of the exam, including vocabulary, are based on what has been taught in the syllabus. However, it is not really possible to regard their construction as guaranteeing that the exams are criterion-referenced in any exact way, i.e. the vocabulary taught in the period covered is not strictly randomly sampled for the exams. For the EFL examinations, all examination instructions are in English. The test construction, examining and grading are all done internally by the schools’ teachers, who are left to rely on their own judgment to ensure the tests are reliable and valid, and that the choice of items is such that a student of some notional ‘just adequante’ ability in the school’s sense will obtain easily a pass score of 40. Only the grade three second term final examinations in Secondary schools (the leaving certificate) are set by the Ministry of Education and are marked by teachers from other schools, coordinated by inspectors. The English exams normally include reading comprehension, writing, grammar and vocabulary. The formats used in the examinations and the design of test items depend entirely on the test-maker. Referring to the general education in Saudi Arabia, Al-Akloby (2001), however, claims

The vocabulary tested has been previously taught in class, but the test items were poorly designed and can often be answered without any real knowledge of words. There was no backwash to prompt thorough vocabulary learning for the examination (Al-Akloby, 2001: 198).

The students must achieve a mark of 40% of the total assessment in the elementary schools in order to pass to a higher grade. In intermediate schools, 40% is pass mark for English, but in many other subjects 50% is the norm. However, in the secondary school 50% is the pass mark required for all academic subjects, in order for the students to progress to the next year in the school. One quarter, 7.5, out of the total 30 marks in the final examination is a minimum pass score for all students. The efforts needed to pass an exam test vary according to students' first term performance. A high first term score means there is less of a need to score well in second term examinations in order to pass. A lower first term score necessitates greater efforts in the second.

The university assessment system

In the Saudi university educational system, each academic subject (module) is taught over a fourteen-week term, then an achievement examination follows.  The examination construction, administration and grading are all done by the course instructors. However, if there are various groups of students who study the same course and are taught by several instructors, the department may ask one or more of the course instructors to construct one version of the final examination to be taken by all the groups at the same time. The contents of the examinations are usually a sample of the syllabus knowledge, which is supposed to represent most aspects of what was taught. However, the sample is usually personally selected rather than random. The aim is to establish how well the individuals have achieved the taught objectives as part of the course implementation. Apart from the subjective judgment of the test-makers, there are no criteria for examining whether or not the difficulty level of the exams set for the same subject by different teachers or even by the same teacher in different terms is identical, and usually the teachers do not vet each other's exams. The test-maker chooses the test formats, the layout and the content of the exam. For English Assessement, vocabulary is tested in mid-term quizzes as well as the final exams. Our preliminary study (Chapter III) shows that testing vocabulary knowledge is an essential part of the EFL examinations in Saudi Arabia.

The total score for any subject, including English, is 100. The assessment is divided, generally, over three periods of examinations, which are two mid-term quizzes (40%) and a final examination (60%). The student must achieve at least 60 % to pass the course and the higher the scores s/he achieves, the higher the grade s/he obtains. These grades are:
 

 The grading system in the Saudi universities

If a student fails to pass a particular module, s/he will need to restudy this module over a further term/s and undergo another assessment.

Conclusion of the review   

Based on the preceding review, it can be said that most common type of exam used at both levels (schools and university) in Saudi Arabia is the achievement exam, which is in line with Al-Akloby’s (2002) claim that the school examinations are intended as achievement exams of what has been taught to the students each school year. Thus, it can be easily concluded that the EFL exams most encountered by the subjects of this study are solely teacher-made achievement tests, on which the following section will throw some light.

The achievement test among other assessment types

An educational test, including an achievement test, is defined as a systematic procedure for observing and describing one or more characteristics of a person with the aid of either a numerical scale or category system (Nitko, 1996).  According to Bachman’s (1990) classification of the different categories of assessment, the nature of an achievement test, as used in schools and universities, is related to both evaluation and measurement.  The following figure illustrates the relationships between evaluation, measurement and test.

 Relationships among assessment types as adopted from Bachman (1990)

According to Bachman (1990), the general meanings of each of the three terms are:

·        Test: to elicit a specific sample of an individual’s characteristic via a measurement instrument.

·        Measurement: to quantify a characteristic of a person according to explicit rules.

·        Evaluation: to make a decision

Area 3 of the above figure shows the combination of the three types of assessment relevant in other situation. Saudi exams elicit a specific sample of an individual’s characteristic by a measurement instrument (Test). Then, quantification  (Measurement) is made on the intended characteristic according to explicit rules in order to be represent ability as numbers or codes.  Then a decision (Evaluation), e.g. pass/fail or go to a higher class, is made and allocated to each individual test taker’s characteristics based on the measurement rules. An example of area 4 would be where an L₂ researcher used a test, a proficiency test for instance, to determine the L₂ general level of proficiency of a group of subjects with no pedagogical or other evaluation in view. However, if the researcher allocated the proficiency levels to a group by having experts rate students on a systematic scale, and not by testing them; it would be considered an example of area 5. Using a teacher’s point of view to rank a student’s ability for the purpose of allocating his/her grade would fall into area 2. In area 1, no measurement instrument is used nor a result quantified. An example of this is to rely on a qualitative report of students’ performance in order to improve the learning context.

Functions of achievement tests

Achievement tests can be grouped into two main functional categories: formative and summative (Nitko, 1996; Airasian, 1997). The purpose of formative assessment is to monitor students’ progress and to provide ongoing feedback to students and teachers during a course or learning unit. Formative assessments are conducted during the course of instruction, when changes or improvements can be made. Formative tests usually describe students' achievement on a small number of specific objectives covered in a relatively short period. Summative assessments usually come at the end of an instructional period, a term, a semester or a year. They usually test the extent to which students have mastered general skills and a wider range of objectives and/or subject matter than a formative assessment does. In Saudi Arabia, most tests are used for summative purposes to assign grades to students for administrative purposes, to dicied on promotion for a higher class. Of course, they could be exploited for formative purposes to guide revision of next term, but they usually are not. Relevant literature also indicates that this is the case in many other countries (Oescher and Kirby, 1990).