Arabic-specific design
Designing an Arabic-Aware Language Aptitude Test
A sound assessment does not simply translate an international test. It preserves the construct while rebuilding tasks around Arabic sound, word formation, grammar, and literacy.
Why direct translation is not enough
The best-known language aptitude batteries were built largely through English-language materials and for particular educational contexts. Their constructs may travel; their items often do not. A spelling clue that works because English spelling and pronunciation diverge cannot simply be translated into Arabic and remain the same task. Likewise, a grammatical example built around English word order may stop measuring the intended relationship when moved word for word into Arabic.
Dana’s thesis treats adaptation as construct-preserving redesign. It reviews five international instruments or versions—the MLAT, PLAB, MLAT-O, MLAT-E, and MLAT-ES—then asks how their underlying abilities can be represented through Arabic. The Spanish MLAT-E adaptation is especially important as a precedent: language, instructions, and stimulus material can change while the architecture of the aptitude task remains recognizable.
Five domains, refined into 17 criteria
The proposed framework began with 32 draft criteria drawn from prior research, test literature, contrastive analysis, and consultation with Arabic-language and assessment specialists. Expert review merged overlapping statements, removed criteria that did not fit a domain, and clarified wording. The final list contains 17 criteria across five domains.
The distribution makes the theory tangible. Phonetic coding receives five criteria; phonetic sensitivity four; vocabulary and spelling three; grammatical sensitivity two; and memory three. These are not five independent school subjects. Together they describe how a learner notices form, builds associations, detects patterns, and retrieves new language.
| Domain | Criteria | What the tasks are intended to reveal |
|---|---|---|
| Phonetic coding | 5 | Recognizing heard symbols, retaining their order, and mapping sounds to numbers, names, or other equivalents. |
| Phonetic sensitivity | 4 | Discriminating Arabic sounds and articulation points, recalling sound, sensing rhythm, and working with morphological patterns. |
| Vocabulary / spelling | 3 | Linking words and meanings, using a sufficient lexical base, and identifying an item that does not belong. |
| Grammatical sensitivity | 2 | Recognizing and analyzing grammatical function, then imitating or matching the function of a morpheme or word. |
| Memory | 3 | Learning words under time pressure and retrieving them through translation or image association. |
Where Arabic changes the task
In phonetic coding, the framework retains the MLAT idea of learning an artificial code and mapping it to heard forms, but the instructions and combinations follow Arabic presentation. In phonetic sensitivity, proposed items draw attention to closely articulated Arabic sounds, rhyme, and the recognition of a changed letter inside a familiar-looking word. The aim is not to reward accent imitation; it is to see how accurately the learner notices, holds, and differentiates unfamiliar sound patterns.
Vocabulary work shifts away from an English-specific “distorted spelling” clue. It uses relationships, roots, odd-one-out decisions, and definitions. This is an important design move because Arabic’s derivational system makes root-and-pattern awareness a plausible way to examine pattern sensitivity—provided later trials confirm that the task measures aptitude rather than accumulated Arabic knowledge.
For grammar, the proposed tasks ask learners to match grammatical roles and infer functions through constructed examples. Arabic features such as definiteness, adjective order, nominal structures, and gendered possessive suffixes inform the examples. Memory is assessed through two routes: rapid word-to-image association and rapid word-to-translation association.
An Arabic-looking item is not automatically an Arabic-valid item. The task must still isolate the intended ability.
A complete 200-item blueprint
The thesis converts the five-domain model into a detailed scoring table of 200 one-point items. Phonetic coding accounts for 50 items; phonetic sensitivity and spelling for 25; vocabulary for 35; grammatical analysis for 60; and memory for 30. The emphasis on grammar and coding reflects the author’s reading of established aptitude batteries and the proposed demands of the Arabic design.
| Section | Items | Share |
|---|---|---|
| Phonetic coding | 50 | 25% |
| Phonetic sensitivity & spelling | 25 | 12.5% |
| Vocabulary | 35 | 17.5% |
| Grammatical analysis | 60 | 30% |
| Memory | 30 | 15% |
Language design is also human design
The framework adds psychological, cultural, and educational safeguards. Questions should move from easier to more complex, instructions should be explicit, difficult formats should include examples, and the testing environment should support focused performance. Results that expose weaknesses should remain confidential and be used to plan support.
Cultural knowledge should not quietly become a prerequisite for success. The thesis advises against direct questions that depend on prior familiarity with Arab or Islamic culture, denigration of other cultures, or claims of Arabic superiority. This is not a minor editorial preference. If a purported aptitude task rewards a particular cultural background, it no longer gives a clean account of the ability it claims to measure.
The result is a strong design proposition: borrow theory internationally, rebuild evidence locally, and make fairness part of the specification from the beginning. The next obligation is empirical—to test whether these Arabic-aware tasks work as intended across learners with different first languages and educational histories.